diff --git a/.gitattributes b/.gitattributes index 00ea38b8676a0c2250c5d2462094d3ba92ca787c..30211e5c5bda6a9a61ee958816c4efd897ba0aed 100644 --- a/.gitattributes +++ b/.gitattributes @@ -185,3 +185,12 @@ babylm_hop_words4_10M_seed0/runs/checkpoint-1000/tokenizer.json filter=lfs diff= babylm_hop_words4_10M_seed0/runs/checkpoint-300/tokenizer.json filter=lfs diff=lfs merge=lfs -text babylm_hop_words4_10M_seed0/runs/checkpoint-1600/tokenizer.json filter=lfs diff=lfs merge=lfs -text babylm_hop_words4_10M_seed0/runs/checkpoint-2100/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text +babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1100/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1100/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1100/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1100/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1100/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1100/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1100/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1100/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1100/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/generation_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/latest b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/latest new file mode 100644 index 0000000000000000000000000000000000000000..873b08355ae55e303606c474ab5fba25f9e8262a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/latest @@ -0,0 +1 @@ +global_step1200 \ No newline at end of file diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/rng_state_0.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..3259c6605b4fec3d2abdf15c7c3ccbdef85f5d96 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b6aefb2eb9cf046f058c1e4bd20c70066eedd4535bf2e9f2b3f178ddbfbfd9e +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/rng_state_1.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..9f4723485ff30930b59224e8afe0cb0b2b3c51bc --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f34c8972483cae48cde438f54c14179a5393e31298d634393f2e166c6a9bfb3e +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..6f245457df6ae96d9dae4e4269752c2f036e6ee3 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0d0068d515416f5010564895d620c337c347a04bf8d5382015b5833708d52850 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/scheduler.pt b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..62f3744e649f425ca3a4fab9d69476ea69341eb0 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b8ed41fc8621931587e709b0292b7cf3eaa7f79d92078bc0d8932ba58b050e95 +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/special_tokens_map.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..4a8bb5389171fd35c11a544b215d1125b227b7e0 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/trainer_state.json @@ -0,0 +1,9393 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.3150684931506849, + "eval_steps": 10, + "global_step": 1200, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.5818793773651123, + "learning_rate": 4.569455727051178e-06, + "loss": 1.85, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 3.0734691619873047, + "learning_rate": 4.567424857839156e-06, + "loss": 1.8682, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.496253252029419, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9007, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.360891342163086, + "learning_rate": 4.5633631194151105e-06, + "loss": 1.8262, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.5731117725372314, + "learning_rate": 4.561332250203087e-06, + "loss": 1.8369, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.700558662414551, + "learning_rate": 4.559301380991065e-06, + "loss": 1.7825, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.561306953430176, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.8174, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.618889570236206, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.8812, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.4671308994293213, + "learning_rate": 4.553208773354996e-06, + "loss": 1.7959, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.4571878910064697, + "learning_rate": 4.551177904142974e-06, + "loss": 1.8943, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.010777711868286, + "eval_runtime": 43.7424, + "eval_samples_per_second": 22.861, + "eval_steps_per_second": 0.96, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.4786667823791504, + "learning_rate": 4.5491470349309505e-06, + "loss": 1.8147, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.2955000400543213, + "learning_rate": 4.547116165718929e-06, + "loss": 1.8547, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.3873913288116455, + "learning_rate": 4.545085296506905e-06, + "loss": 1.8156, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.2449734210968018, + "learning_rate": 4.543054427294882e-06, + "loss": 1.8053, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.3969857692718506, + "learning_rate": 4.54102355808286e-06, + "loss": 1.8085, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.3446357250213623, + "learning_rate": 4.538992688870837e-06, + "loss": 1.8355, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.408762216567993, + "learning_rate": 4.536961819658814e-06, + "loss": 1.8319, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.362962245941162, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.8422, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.402984857559204, + "learning_rate": 4.532900081234769e-06, + "loss": 1.8125, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.3964452743530273, + "learning_rate": 4.530869212022746e-06, + "loss": 1.8417, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.006908416748047, + "eval_runtime": 44.2701, + "eval_samples_per_second": 22.589, + "eval_steps_per_second": 0.949, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.595233201980591, + "learning_rate": 4.528838342810723e-06, + "loss": 1.8275, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.326117753982544, + "learning_rate": 4.526807473598701e-06, + "loss": 1.8405, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.4904465675354004, + "learning_rate": 4.524776604386678e-06, + "loss": 1.8172, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.4591610431671143, + "learning_rate": 4.522745735174655e-06, + "loss": 1.876, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.319800853729248, + "learning_rate": 4.520714865962632e-06, + "loss": 1.8294, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.28790020942688, + "learning_rate": 4.51868399675061e-06, + "loss": 1.8537, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.284822702407837, + "learning_rate": 4.516653127538587e-06, + "loss": 1.8087, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.273820638656616, + "learning_rate": 4.514622258326564e-06, + "loss": 1.7992, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.2025654315948486, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.8248, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 2.3218843936920166, + "learning_rate": 4.510560519902519e-06, + "loss": 1.8647, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.0048956871032715, + "eval_runtime": 44.5972, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.515766143798828, + "learning_rate": 4.508529650690496e-06, + "loss": 1.7673, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.4376487731933594, + "learning_rate": 4.506498781478473e-06, + "loss": 1.7755, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.2801272869110107, + "learning_rate": 4.504467912266451e-06, + "loss": 1.8273, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.4707891941070557, + "learning_rate": 4.502437043054428e-06, + "loss": 1.8971, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.301301956176758, + "learning_rate": 4.500406173842405e-06, + "loss": 1.8345, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.2317373752593994, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.8248, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.447373867034912, + "learning_rate": 4.49634443541836e-06, + "loss": 1.8148, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.337245225906372, + "learning_rate": 4.494313566206336e-06, + "loss": 1.8811, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3717031478881836, + "learning_rate": 4.492282696994314e-06, + "loss": 1.8311, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.287677526473999, + "learning_rate": 4.490251827782291e-06, + "loss": 1.7728, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.004889965057373, + "eval_runtime": 43.8132, + "eval_samples_per_second": 22.824, + "eval_steps_per_second": 0.959, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.441445827484131, + "learning_rate": 4.488220958570269e-06, + "loss": 1.7927, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.3191535472869873, + "learning_rate": 4.486190089358245e-06, + "loss": 1.759, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.3118820190429688, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8509, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.4711859226226807, + "learning_rate": 4.4821283509342e-06, + "loss": 1.8402, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.3142173290252686, + "learning_rate": 4.480097481722178e-06, + "loss": 1.8238, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.2665774822235107, + "learning_rate": 4.478066612510154e-06, + "loss": 1.7826, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.1692490577697754, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.8138, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.412783145904541, + "learning_rate": 4.474004874086109e-06, + "loss": 1.9421, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3692686557769775, + "learning_rate": 4.471974004874087e-06, + "loss": 1.7722, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.3224151134490967, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.8753, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.0016305446624756, + "eval_runtime": 45.564, + "eval_samples_per_second": 21.947, + "eval_steps_per_second": 0.922, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.51043438911438, + "learning_rate": 4.467912266450041e-06, + "loss": 1.8325, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.4568657875061035, + "learning_rate": 4.465881397238018e-06, + "loss": 1.8081, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.3279001712799072, + "learning_rate": 4.463850528025995e-06, + "loss": 1.8439, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.6041171550750732, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.8623, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.3780531883239746, + "learning_rate": 4.45978878960195e-06, + "loss": 1.8269, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.511500597000122, + "learning_rate": 4.457757920389927e-06, + "loss": 1.7749, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.707789659500122, + "learning_rate": 4.455727051177904e-06, + "loss": 1.8157, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.5495495796203613, + "learning_rate": 4.453696181965882e-06, + "loss": 1.8534, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.255876302719116, + "learning_rate": 4.451665312753859e-06, + "loss": 1.8377, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.415097951889038, + "learning_rate": 4.449634443541836e-06, + "loss": 1.8764, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 1.9943474531173706, + "eval_runtime": 42.2985, + "eval_samples_per_second": 23.642, + "eval_steps_per_second": 0.993, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.2604713439941406, + "learning_rate": 4.4476035743298135e-06, + "loss": 1.7898, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.5643842220306396, + "learning_rate": 4.445572705117791e-06, + "loss": 1.7953, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.412548780441284, + "learning_rate": 4.443541835905768e-06, + "loss": 1.8611, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.5423877239227295, + "learning_rate": 4.441510966693745e-06, + "loss": 1.8049, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.4986631870269775, + "learning_rate": 4.439480097481723e-06, + "loss": 1.8759, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.8349273204803467, + "learning_rate": 4.4374492282697e-06, + "loss": 1.7416, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.452017068862915, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8268, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.6425273418426514, + "learning_rate": 4.433387489845654e-06, + "loss": 1.8334, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.328740358352661, + "learning_rate": 4.431356620633632e-06, + "loss": 1.7931, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.262209415435791, + "learning_rate": 4.429325751421609e-06, + "loss": 1.8151, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 1.9917123317718506, + "eval_runtime": 42.4309, + "eval_samples_per_second": 23.568, + "eval_steps_per_second": 0.99, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.2984049320220947, + "learning_rate": 4.427294882209586e-06, + "loss": 1.8187, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.3910019397735596, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.7512, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.4494471549987793, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8202, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.3857171535491943, + "learning_rate": 4.421202274573518e-06, + "loss": 1.8279, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.4526500701904297, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.7136, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.2216501235961914, + "learning_rate": 4.417140536149473e-06, + "loss": 1.7574, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.7639758586883545, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8403, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.3177521228790283, + "learning_rate": 4.413078797725427e-06, + "loss": 1.7898, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.527902364730835, + "learning_rate": 4.411047928513404e-06, + "loss": 1.8184, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.34061336517334, + "learning_rate": 4.409017059301382e-06, + "loss": 1.811, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 1.9882216453552246, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.6462934017181396, + "learning_rate": 4.406986190089358e-06, + "loss": 1.8501, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.2725913524627686, + "learning_rate": 4.404955320877336e-06, + "loss": 1.8262, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.390084981918335, + "learning_rate": 4.402924451665313e-06, + "loss": 1.7842, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.314305305480957, + "learning_rate": 4.400893582453291e-06, + "loss": 1.7692, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.2828257083892822, + "learning_rate": 4.398862713241267e-06, + "loss": 1.8031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.3493599891662598, + "learning_rate": 4.396831844029245e-06, + "loss": 1.8183, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.423365831375122, + "learning_rate": 4.394800974817222e-06, + "loss": 1.8012, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.438015937805176, + "learning_rate": 4.392770105605199e-06, + "loss": 1.8217, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.357172727584839, + "learning_rate": 4.390739236393176e-06, + "loss": 1.8382, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.3383474349975586, + "learning_rate": 4.388708367181154e-06, + "loss": 1.841, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 1.9878787994384766, + "eval_runtime": 44.0694, + "eval_samples_per_second": 22.691, + "eval_steps_per_second": 0.953, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.459385633468628, + "learning_rate": 4.386677497969131e-06, + "loss": 1.8377, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.356788158416748, + "learning_rate": 4.384646628757108e-06, + "loss": 1.8718, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.3217222690582275, + "learning_rate": 4.3826157595450855e-06, + "loss": 1.8073, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.3134331703186035, + "learning_rate": 4.380584890333063e-06, + "loss": 1.7675, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.3273653984069824, + "learning_rate": 4.37855402112104e-06, + "loss": 1.7169, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.611034393310547, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8702, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.2077300548553467, + "learning_rate": 4.3744922826969946e-06, + "loss": 1.7951, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.3393311500549316, + "learning_rate": 4.372461413484972e-06, + "loss": 1.8136, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.3306660652160645, + "learning_rate": 4.370430544272949e-06, + "loss": 1.8372, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.4440250396728516, + "learning_rate": 4.368399675060926e-06, + "loss": 1.844, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 1.98439621925354, + "eval_runtime": 44.4595, + "eval_samples_per_second": 22.492, + "eval_steps_per_second": 0.945, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.205542802810669, + "learning_rate": 4.366368805848904e-06, + "loss": 1.8182, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.6030094623565674, + "learning_rate": 4.364337936636881e-06, + "loss": 1.7524, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.4000444412231445, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8269, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.319526433944702, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.7722, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.1624808311462402, + "learning_rate": 4.358245329000813e-06, + "loss": 1.8081, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.26615309715271, + "learning_rate": 4.35621445978879e-06, + "loss": 1.8049, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.4895036220550537, + "learning_rate": 4.354183590576767e-06, + "loss": 1.7932, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.437161445617676, + "learning_rate": 4.352152721364745e-06, + "loss": 1.772, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.344069480895996, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8213, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.360734224319458, + "learning_rate": 4.348090982940699e-06, + "loss": 1.8268, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 1.9846479892730713, + "eval_runtime": 44.947, + "eval_samples_per_second": 22.248, + "eval_steps_per_second": 0.934, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.59487247467041, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.8303, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.6512107849121094, + "learning_rate": 4.344029244516654e-06, + "loss": 1.7979, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.2900476455688477, + "learning_rate": 4.341998375304631e-06, + "loss": 1.8707, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.409494638442993, + "learning_rate": 4.339967506092607e-06, + "loss": 1.8075, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.527205467224121, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.7884, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.33461332321167, + "learning_rate": 4.335905767668562e-06, + "loss": 1.7839, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.3555047512054443, + "learning_rate": 4.33387489845654e-06, + "loss": 1.8245, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.397148609161377, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.8115, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.232139825820923, + "learning_rate": 4.329813160032495e-06, + "loss": 1.7747, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.1684389114379883, + "learning_rate": 4.327782290820471e-06, + "loss": 1.748, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 1.980297565460205, + "eval_runtime": 42.9183, + "eval_samples_per_second": 23.3, + "eval_steps_per_second": 0.979, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.2454638481140137, + "learning_rate": 4.325751421608449e-06, + "loss": 1.83, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.3264827728271484, + "learning_rate": 4.323720552396426e-06, + "loss": 1.7856, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.298804998397827, + "learning_rate": 4.321689683184404e-06, + "loss": 1.78, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2202727794647217, + "learning_rate": 4.31965881397238e-06, + "loss": 1.7875, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.494479179382324, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.8363, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3520524501800537, + "learning_rate": 4.315597075548335e-06, + "loss": 1.8266, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.373878002166748, + "learning_rate": 4.313566206336312e-06, + "loss": 1.8348, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.4232168197631836, + "learning_rate": 4.311535337124289e-06, + "loss": 1.77, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.2572121620178223, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.7914, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.5817952156066895, + "learning_rate": 4.307473598700244e-06, + "loss": 1.8099, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 1.9784427881240845, + "eval_runtime": 43.2981, + "eval_samples_per_second": 23.096, + "eval_steps_per_second": 0.97, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.646040916442871, + "learning_rate": 4.305442729488221e-06, + "loss": 1.7452, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.370220184326172, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8399, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.4599130153656006, + "learning_rate": 4.301380991064176e-06, + "loss": 1.8273, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 2.4135093688964844, + "learning_rate": 4.299350121852153e-06, + "loss": 1.852, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.371021270751953, + "learning_rate": 4.29731925264013e-06, + "loss": 1.8339, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.6858229637145996, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.792, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.4101266860961914, + "learning_rate": 4.293257514216085e-06, + "loss": 1.7556, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.769761800765991, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8223, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.5777804851531982, + "learning_rate": 4.289195775792039e-06, + "loss": 1.8406, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.376864433288574, + "learning_rate": 4.287164906580017e-06, + "loss": 1.7846, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 1.9758412837982178, + "eval_runtime": 43.9353, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.7461960315704346, + "learning_rate": 4.285134037367994e-06, + "loss": 1.776, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.549576759338379, + "learning_rate": 4.283103168155971e-06, + "loss": 1.7644, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.568074941635132, + "learning_rate": 4.281072298943948e-06, + "loss": 1.7571, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.8195722103118896, + "learning_rate": 4.279041429731926e-06, + "loss": 1.7959, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.471623182296753, + "learning_rate": 4.277010560519903e-06, + "loss": 1.7584, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.664689302444458, + "learning_rate": 4.27497969130788e-06, + "loss": 1.7909, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.684504985809326, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.8205, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.5247528553009033, + "learning_rate": 4.270917952883835e-06, + "loss": 1.8365, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.757972478866577, + "learning_rate": 4.268887083671812e-06, + "loss": 1.7665, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.6032614707946777, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8702, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 1.9716291427612305, + "eval_runtime": 43.5482, + "eval_samples_per_second": 22.963, + "eval_steps_per_second": 0.964, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.5324370861053467, + "learning_rate": 4.264825345247766e-06, + "loss": 1.738, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.730712890625, + "learning_rate": 4.262794476035744e-06, + "loss": 1.7918, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.5554966926574707, + "learning_rate": 4.26076360682372e-06, + "loss": 1.7218, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.4774436950683594, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.7823, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.77431058883667, + "learning_rate": 4.256701868399675e-06, + "loss": 1.794, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.258828639984131, + "learning_rate": 4.254670999187653e-06, + "loss": 1.805, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.384399175643921, + "learning_rate": 4.252640129975629e-06, + "loss": 1.818, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3278932571411133, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.7719, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.207348108291626, + "learning_rate": 4.248578391551584e-06, + "loss": 1.7362, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.4525978565216064, + "learning_rate": 4.246547522339562e-06, + "loss": 1.778, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 1.9694374799728394, + "eval_runtime": 42.5977, + "eval_samples_per_second": 23.475, + "eval_steps_per_second": 0.986, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.377835273742676, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.7926, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.4623866081237793, + "learning_rate": 4.242485783915517e-06, + "loss": 1.7366, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.542219400405884, + "learning_rate": 4.240454914703493e-06, + "loss": 1.7409, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.1955819129943848, + "learning_rate": 4.23842404549147e-06, + "loss": 1.7596, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.5834269523620605, + "learning_rate": 4.236393176279448e-06, + "loss": 1.7343, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.219660520553589, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8089, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.283200740814209, + "learning_rate": 4.232331437855402e-06, + "loss": 1.8248, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.274127244949341, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.825, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.3550686836242676, + "learning_rate": 4.228269699431357e-06, + "loss": 1.7957, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.588027238845825, + "learning_rate": 4.226238830219334e-06, + "loss": 1.7865, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 1.9662212133407593, + "eval_runtime": 43.2661, + "eval_samples_per_second": 23.113, + "eval_steps_per_second": 0.971, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.2445170879364014, + "learning_rate": 4.224207961007311e-06, + "loss": 1.791, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.399174690246582, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.7944, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.586184501647949, + "learning_rate": 4.220146222583266e-06, + "loss": 1.7856, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.4174695014953613, + "learning_rate": 4.218115353371243e-06, + "loss": 1.7842, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2684640884399414, + "learning_rate": 4.21608448415922e-06, + "loss": 1.7482, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 3.1782288551330566, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8286, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.4046335220336914, + "learning_rate": 4.212022745735175e-06, + "loss": 1.7228, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 3.3722083568573, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8044, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.293466329574585, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.7415, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.28663969039917, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8282, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 1.9627619981765747, + "eval_runtime": 43.6092, + "eval_samples_per_second": 22.931, + "eval_steps_per_second": 0.963, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.255870819091797, + "learning_rate": 4.203899268887084e-06, + "loss": 1.7983, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.4964773654937744, + "learning_rate": 4.201868399675061e-06, + "loss": 1.8141, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.3372740745544434, + "learning_rate": 4.199837530463039e-06, + "loss": 1.7142, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.3506968021392822, + "learning_rate": 4.197806661251016e-06, + "loss": 1.7898, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.242410898208618, + "learning_rate": 4.195775792038993e-06, + "loss": 1.7677, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.3864502906799316, + "learning_rate": 4.19374492282697e-06, + "loss": 1.7097, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.482349157333374, + "learning_rate": 4.191714053614948e-06, + "loss": 1.8052, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.317652940750122, + "learning_rate": 4.189683184402924e-06, + "loss": 1.7692, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.2777392864227295, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8246, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.3506035804748535, + "learning_rate": 4.185621445978879e-06, + "loss": 1.7576, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 1.9610412120819092, + "eval_runtime": 43.3119, + "eval_samples_per_second": 23.088, + "eval_steps_per_second": 0.97, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.3011536598205566, + "learning_rate": 4.183590576766857e-06, + "loss": 1.7891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.2614853382110596, + "learning_rate": 4.181559707554833e-06, + "loss": 1.8258, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.219175100326538, + "learning_rate": 4.179528838342811e-06, + "loss": 1.7494, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.4239423274993896, + "learning_rate": 4.177497969130788e-06, + "loss": 1.7611, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.470712423324585, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8281, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.504060745239258, + "learning_rate": 4.173436230706742e-06, + "loss": 1.7641, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.326601982116699, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.7671, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.4280030727386475, + "learning_rate": 4.169374492282697e-06, + "loss": 1.7573, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.455716371536255, + "learning_rate": 4.167343623070675e-06, + "loss": 1.809, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.3029844760894775, + "learning_rate": 4.165312753858652e-06, + "loss": 1.77, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 1.9604709148406982, + "eval_runtime": 44.4627, + "eval_samples_per_second": 22.491, + "eval_steps_per_second": 0.945, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.3833982944488525, + "learning_rate": 4.163281884646629e-06, + "loss": 1.7652, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.2392966747283936, + "learning_rate": 4.161251015434607e-06, + "loss": 1.7904, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.4546353816986084, + "learning_rate": 4.159220146222583e-06, + "loss": 1.7324, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.13592529296875, + "learning_rate": 4.157189277010561e-06, + "loss": 1.759, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.381657361984253, + "learning_rate": 4.155158407798538e-06, + "loss": 1.7871, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.4142940044403076, + "learning_rate": 4.153127538586516e-06, + "loss": 1.7832, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.305697441101074, + "learning_rate": 4.151096669374492e-06, + "loss": 1.7619, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.193429708480835, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.7738, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.3007619380950928, + "learning_rate": 4.147034930950447e-06, + "loss": 1.7839, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.6337807178497314, + "learning_rate": 4.145004061738425e-06, + "loss": 1.7748, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 1.9597856998443604, + "eval_runtime": 43.8644, + "eval_samples_per_second": 22.798, + "eval_steps_per_second": 0.957, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.4294793605804443, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.7001, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.512155532836914, + "learning_rate": 4.140942323314379e-06, + "loss": 1.7334, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.290808916091919, + "learning_rate": 4.138911454102356e-06, + "loss": 1.7522, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.483161211013794, + "learning_rate": 4.136880584890333e-06, + "loss": 1.798, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2973921298980713, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.7847, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.253084659576416, + "learning_rate": 4.132818846466288e-06, + "loss": 1.7812, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.306342840194702, + "learning_rate": 4.130787977254265e-06, + "loss": 1.7886, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.5153164863586426, + "learning_rate": 4.128757108042242e-06, + "loss": 1.7954, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.4122278690338135, + "learning_rate": 4.12672623883022e-06, + "loss": 1.8525, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.4520716667175293, + "learning_rate": 4.124695369618197e-06, + "loss": 1.7648, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 1.9547661542892456, + "eval_runtime": 45.3774, + "eval_samples_per_second": 22.037, + "eval_steps_per_second": 0.926, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2959611415863037, + "learning_rate": 4.122664500406174e-06, + "loss": 1.7626, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.2280471324920654, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.7954, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.3870065212249756, + "learning_rate": 4.118602761982129e-06, + "loss": 1.7377, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.292612075805664, + "learning_rate": 4.116571892770106e-06, + "loss": 1.7, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.4607021808624268, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8338, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.532824993133545, + "learning_rate": 4.112510154346061e-06, + "loss": 1.7549, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2139623165130615, + "learning_rate": 4.110479285134038e-06, + "loss": 1.78, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.4749462604522705, + "learning_rate": 4.108448415922015e-06, + "loss": 1.8034, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.609265089035034, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.7941, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.367753028869629, + "learning_rate": 4.10438667749797e-06, + "loss": 1.6653, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 1.9539440870285034, + "eval_runtime": 48.8438, + "eval_samples_per_second": 20.473, + "eval_steps_per_second": 0.86, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.9723567962646484, + "learning_rate": 4.102355808285947e-06, + "loss": 1.8317, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.388806104660034, + "learning_rate": 4.100324939073924e-06, + "loss": 1.7597, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4882986545562744, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.7708, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.7586395740509033, + "learning_rate": 4.096263200649879e-06, + "loss": 1.7804, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.237025260925293, + "learning_rate": 4.094232331437856e-06, + "loss": 1.7415, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.421330213546753, + "learning_rate": 4.092201462225833e-06, + "loss": 1.734, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.7281360626220703, + "learning_rate": 4.090170593013811e-06, + "loss": 1.7716, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.2200379371643066, + "learning_rate": 4.088139723801787e-06, + "loss": 1.7266, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.465440034866333, + "learning_rate": 4.086108854589765e-06, + "loss": 1.741, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.39835262298584, + "learning_rate": 4.084077985377742e-06, + "loss": 1.757, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 1.9514222145080566, + "eval_runtime": 52.8068, + "eval_samples_per_second": 18.937, + "eval_steps_per_second": 0.795, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.526684522628784, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8195, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.341954469680786, + "learning_rate": 4.080016246953696e-06, + "loss": 1.7196, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.3755977153778076, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8115, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.3952431678771973, + "learning_rate": 4.075954508529651e-06, + "loss": 1.812, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.554089069366455, + "learning_rate": 4.073923639317629e-06, + "loss": 1.8184, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.2319302558898926, + "learning_rate": 4.071892770105605e-06, + "loss": 1.7448, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.284856081008911, + "learning_rate": 4.069861900893583e-06, + "loss": 1.7154, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.4081835746765137, + "learning_rate": 4.06783103168156e-06, + "loss": 1.7311, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.6115968227386475, + "learning_rate": 4.065800162469538e-06, + "loss": 1.7436, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.368004322052002, + "learning_rate": 4.063769293257514e-06, + "loss": 1.7811, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 1.9508955478668213, + "eval_runtime": 46.8711, + "eval_samples_per_second": 21.335, + "eval_steps_per_second": 0.896, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.4607419967651367, + "learning_rate": 4.061738424045492e-06, + "loss": 1.7563, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.2978763580322266, + "learning_rate": 4.059707554833469e-06, + "loss": 1.7755, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.3316237926483154, + "learning_rate": 4.057676685621446e-06, + "loss": 1.8226, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.3699169158935547, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.7755, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.195596218109131, + "learning_rate": 4.053614947197401e-06, + "loss": 1.6916, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.4524662494659424, + "learning_rate": 4.051584077985378e-06, + "loss": 1.774, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.573888063430786, + "learning_rate": 4.049553208773355e-06, + "loss": 1.7953, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.5157253742218018, + "learning_rate": 4.047522339561333e-06, + "loss": 1.7597, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.5588185787200928, + "learning_rate": 4.04549147034931e-06, + "loss": 1.7553, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.8950047492980957, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8647, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 1.945882797241211, + "eval_runtime": 45.481, + "eval_samples_per_second": 21.987, + "eval_steps_per_second": 0.923, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.217116117477417, + "learning_rate": 4.041429731925264e-06, + "loss": 1.7465, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.5601565837860107, + "learning_rate": 4.039398862713242e-06, + "loss": 1.7575, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.4540457725524902, + "learning_rate": 4.037367993501219e-06, + "loss": 1.8291, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.327181100845337, + "learning_rate": 4.035337124289196e-06, + "loss": 1.7409, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.485457181930542, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.7192, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.4613442420959473, + "learning_rate": 4.031275385865151e-06, + "loss": 1.7613, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.303217649459839, + "learning_rate": 4.029244516653128e-06, + "loss": 1.7799, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.3830621242523193, + "learning_rate": 4.027213647441105e-06, + "loss": 1.7045, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.342332601547241, + "learning_rate": 4.025182778229083e-06, + "loss": 1.7246, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.4014711380004883, + "learning_rate": 4.02315190901706e-06, + "loss": 1.7621, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 1.95199453830719, + "eval_runtime": 45.4754, + "eval_samples_per_second": 21.99, + "eval_steps_per_second": 0.924, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.6015400886535645, + "learning_rate": 4.021121039805037e-06, + "loss": 1.7536, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.866978406906128, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8405, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.389810085296631, + "learning_rate": 4.017059301380992e-06, + "loss": 1.7365, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.280442953109741, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8273, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.509330987930298, + "learning_rate": 4.012997562956945e-06, + "loss": 1.7127, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.452078104019165, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.7457, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.368337631225586, + "learning_rate": 4.0089358245329e-06, + "loss": 1.7412, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.4529192447662354, + "learning_rate": 4.006904955320878e-06, + "loss": 1.7434, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.647761821746826, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8013, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.387424945831299, + "learning_rate": 4.002843216896833e-06, + "loss": 1.7948, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 1.946029543876648, + "eval_runtime": 43.9666, + "eval_samples_per_second": 22.745, + "eval_steps_per_second": 0.955, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.9217846393585205, + "learning_rate": 4.000812347684809e-06, + "loss": 1.7384, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.400040864944458, + "learning_rate": 3.998781478472787e-06, + "loss": 1.7496, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.4050440788269043, + "learning_rate": 3.996750609260764e-06, + "loss": 1.7165, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 3.6001529693603516, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8679, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 2.4040098190307617, + "learning_rate": 3.992688870836718e-06, + "loss": 1.7671, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.5645527839660645, + "learning_rate": 3.990658001624696e-06, + "loss": 1.7374, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.799394130706787, + "learning_rate": 3.988627132412673e-06, + "loss": 1.796, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.5481607913970947, + "learning_rate": 3.98659626320065e-06, + "loss": 1.753, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.8851330280303955, + "learning_rate": 3.984565393988627e-06, + "loss": 1.7505, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.6668694019317627, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.7604, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 1.939759612083435, + "eval_runtime": 44.5019, + "eval_samples_per_second": 22.471, + "eval_steps_per_second": 0.944, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.320935010910034, + "learning_rate": 3.980503655564582e-06, + "loss": 1.7749, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.829897880554199, + "learning_rate": 3.978472786352559e-06, + "loss": 1.7763, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.8420908451080322, + "learning_rate": 3.976441917140536e-06, + "loss": 1.7598, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.8183, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.781, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 3.9271891117095947, + "learning_rate": 3.972380178716491e-06, + "loss": 1.7949, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.277897834777832, + "learning_rate": 3.970349309504468e-06, + "loss": 1.707, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 3.241063117980957, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.7175, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.514458417892456, + "learning_rate": 3.966287571080423e-06, + "loss": 1.7764, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.420212745666504, + "learning_rate": 3.9642567018684e-06, + "loss": 1.7618, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 1.9401386976242065, + "eval_runtime": 44.1672, + "eval_samples_per_second": 22.641, + "eval_steps_per_second": 0.951, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.3980538845062256, + "learning_rate": 3.962225832656377e-06, + "loss": 1.6922, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.892646312713623, + "learning_rate": 3.960194963444355e-06, + "loss": 1.7571, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.499260902404785, + "learning_rate": 3.958164094232332e-06, + "loss": 1.7501, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.409695625305176, + "learning_rate": 3.956133225020309e-06, + "loss": 1.7553, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.640069007873535, + "learning_rate": 3.954102355808286e-06, + "loss": 1.7526, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.9101760387420654, + "learning_rate": 3.952071486596264e-06, + "loss": 1.7483, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.2579474449157715, + "learning_rate": 3.950040617384241e-06, + "loss": 1.7379, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 4.039298057556152, + "learning_rate": 3.948009748172218e-06, + "loss": 1.7988, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.3232030868530273, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.7592, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.697164297103882, + "learning_rate": 3.943948009748173e-06, + "loss": 1.7622, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 1.9367746114730835, + "eval_runtime": 43.7013, + "eval_samples_per_second": 22.883, + "eval_steps_per_second": 0.961, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.480297803878784, + "learning_rate": 3.94191714053615e-06, + "loss": 1.8006, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.3624603748321533, + "learning_rate": 3.939886271324127e-06, + "loss": 1.7152, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.417187213897705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.7077, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.6575164794921875, + "learning_rate": 3.935824532900082e-06, + "loss": 1.7744, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.4896557331085205, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8052, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.475693941116333, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.7008, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.3210372924804688, + "learning_rate": 3.929731925264013e-06, + "loss": 1.7339, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.433785915374756, + "learning_rate": 3.927701056051991e-06, + "loss": 1.7544, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 3.181574583053589, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.7229, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.704399347305298, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.7086, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 1.9335075616836548, + "eval_runtime": 47.0248, + "eval_samples_per_second": 21.265, + "eval_steps_per_second": 0.893, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 3.5027623176574707, + "learning_rate": 3.921608448415922e-06, + "loss": 1.7622, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.450943946838379, + "learning_rate": 3.9195775792039e-06, + "loss": 1.7286, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.4603374004364014, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.7725, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.8217649459838867, + "learning_rate": 3.915515840779855e-06, + "loss": 1.7859, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.3308372497558594, + "learning_rate": 3.913484971567831e-06, + "loss": 1.6874, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.624776840209961, + "learning_rate": 3.911454102355808e-06, + "loss": 1.7632, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.4097838401794434, + "learning_rate": 3.909423233143786e-06, + "loss": 1.7576, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.487410545349121, + "learning_rate": 3.907392363931763e-06, + "loss": 1.831, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.647245168685913, + "learning_rate": 3.90536149471974e-06, + "loss": 1.8448, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.397664785385132, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.868, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 1.930379033088684, + "eval_runtime": 51.1662, + "eval_samples_per_second": 19.544, + "eval_steps_per_second": 0.821, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.277433395385742, + "learning_rate": 3.901299756295695e-06, + "loss": 1.7394, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 3.1312496662139893, + "learning_rate": 3.899268887083672e-06, + "loss": 1.7422, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.5011534690856934, + "learning_rate": 3.897238017871649e-06, + "loss": 1.7484, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.8716166019439697, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.7253, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.4346914291381836, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8197, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.348475933074951, + "learning_rate": 3.891145410235581e-06, + "loss": 1.7737, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.3996007442474365, + "learning_rate": 3.889114541023558e-06, + "loss": 1.7333, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.355262279510498, + "learning_rate": 3.887083671811536e-06, + "loss": 1.6902, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.335932731628418, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8206, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.2893190383911133, + "learning_rate": 3.88302193338749e-06, + "loss": 1.7533, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 1.9290764331817627, + "eval_runtime": 46.9373, + "eval_samples_per_second": 21.305, + "eval_steps_per_second": 0.895, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.739842176437378, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.7318, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.587916374206543, + "learning_rate": 3.878960194963445e-06, + "loss": 1.7979, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.406548500061035, + "learning_rate": 3.876929325751422e-06, + "loss": 1.6899, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.440321922302246, + "learning_rate": 3.874898456539399e-06, + "loss": 1.7546, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.320030450820923, + "learning_rate": 3.872867587327377e-06, + "loss": 1.7254, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.3795316219329834, + "learning_rate": 3.870836718115354e-06, + "loss": 1.728, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.299802541732788, + "learning_rate": 3.868805848903331e-06, + "loss": 1.7464, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.469719409942627, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.765, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.612109899520874, + "learning_rate": 3.864744110479286e-06, + "loss": 1.7827, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.623173952102661, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8019, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 1.9268505573272705, + "eval_runtime": 46.77, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.4146616458892822, + "learning_rate": 3.86068237205524e-06, + "loss": 1.7498, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.361994743347168, + "learning_rate": 3.858651502843217e-06, + "loss": 1.6926, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.493215560913086, + "learning_rate": 3.856620633631195e-06, + "loss": 1.736, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.4341812133789062, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8152, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.391331195831299, + "learning_rate": 3.852558895207149e-06, + "loss": 1.7717, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.419492244720459, + "learning_rate": 3.850528025995126e-06, + "loss": 1.7283, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.349090576171875, + "learning_rate": 3.848497156783104e-06, + "loss": 1.7317, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.298647880554199, + "learning_rate": 3.84646628757108e-06, + "loss": 1.7389, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.4612667560577393, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.7333, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.3139944076538086, + "learning_rate": 3.842404549147035e-06, + "loss": 1.6944, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 1.924438714981079, + "eval_runtime": 45.8276, + "eval_samples_per_second": 21.821, + "eval_steps_per_second": 0.916, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 2.17958664894104, + "learning_rate": 3.840373679935013e-06, + "loss": 1.7474, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.337859869003296, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.7345, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4777672290802, + "learning_rate": 3.836311941510967e-06, + "loss": 1.77, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.280158519744873, + "learning_rate": 3.834281072298944e-06, + "loss": 1.7069, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2387664318084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.7889, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.373771905899048, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8268, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.4950132369995117, + "learning_rate": 3.828188464662876e-06, + "loss": 1.7586, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.3296444416046143, + "learning_rate": 3.826157595450853e-06, + "loss": 1.6991, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.6653530597686768, + "learning_rate": 3.82412672623883e-06, + "loss": 1.7505, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.485128402709961, + "learning_rate": 3.822095857026808e-06, + "loss": 1.7196, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 1.9221889972686768, + "eval_runtime": 46.6581, + "eval_samples_per_second": 21.433, + "eval_steps_per_second": 0.9, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.3726181983947754, + "learning_rate": 3.820064987814785e-06, + "loss": 1.7296, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.233992099761963, + "learning_rate": 3.818034118602762e-06, + "loss": 1.7313, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.5388147830963135, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.7435, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.4787333011627197, + "learning_rate": 3.813972380178717e-06, + "loss": 1.7733, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.232564926147461, + "learning_rate": 3.811941510966694e-06, + "loss": 1.693, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.3882386684417725, + "learning_rate": 3.809910641754671e-06, + "loss": 1.6529, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.4785983562469482, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.7456, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.685098886489868, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.7162, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.353523015975952, + "learning_rate": 3.803818034118603e-06, + "loss": 1.7852, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.342163562774658, + "learning_rate": 3.80178716490658e-06, + "loss": 1.7767, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 1.9235820770263672, + "eval_runtime": 47.1327, + "eval_samples_per_second": 21.217, + "eval_steps_per_second": 0.891, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.550818920135498, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.7328, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.3204798698425293, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.7058, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.4752719402313232, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8051, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.5202794075012207, + "learning_rate": 3.793663688058489e-06, + "loss": 1.7966, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.26521635055542, + "learning_rate": 3.791632818846467e-06, + "loss": 1.7879, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.331515312194824, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.7814, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.37178635597229, + "learning_rate": 3.787571080422421e-06, + "loss": 1.7635, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.321417808532715, + "learning_rate": 3.785540211210398e-06, + "loss": 1.7513, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.485989809036255, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.7306, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.7497379779815674, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8409, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 1.9275448322296143, + "eval_runtime": 46.2088, + "eval_samples_per_second": 21.641, + "eval_steps_per_second": 0.909, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 4.9214019775390625, + "learning_rate": 3.77944760357433e-06, + "loss": 1.7834, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 4.505277156829834, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.7179, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.5772030353546143, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.7766, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.254185438156128, + "learning_rate": 3.773354995938262e-06, + "loss": 1.7697, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 3.632233142852783, + "learning_rate": 3.771324126726239e-06, + "loss": 1.7968, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.174461841583252, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.7346, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.3983266353607178, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.6534, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 3.393791913986206, + "learning_rate": 3.765231519090171e-06, + "loss": 1.7561, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.2853987216949463, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.6963, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.649789810180664, + "learning_rate": 3.761169780666125e-06, + "loss": 1.7395, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 1.9242008924484253, + "eval_runtime": 46.6017, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 3.417872190475464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.784, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.316666841506958, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.7154, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.486161231994629, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.7903, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 4.538443565368652, + "learning_rate": 3.753046303818034e-06, + "loss": 1.7711, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.9772889614105225, + "learning_rate": 3.751015434606012e-06, + "loss": 1.8046, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 7.2353715896606445, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.7189, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.736, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.8026, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 12.551762580871582, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.7788, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 3.0898046493530273, + "learning_rate": 3.742891957757921e-06, + "loss": 1.7878, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 1.9646916389465332, + "eval_runtime": 44.1583, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 8.120022773742676, + "learning_rate": 3.740861088545898e-06, + "loss": 1.768, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 6.485774040222168, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.746, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.85135555267334, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.6968, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 5.8551435470581055, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.6796, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 4.45880126953125, + "learning_rate": 3.732737611697807e-06, + "loss": 1.7219, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 3.0435831546783447, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.6959, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.8194847106933594, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.8152, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 3.1266117095947266, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.6937, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 3.642854690551758, + "learning_rate": 3.724614134849716e-06, + "loss": 1.7992, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 3.344496965408325, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.7583, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 1.9293720722198486, + "eval_runtime": 50.8602, + "eval_samples_per_second": 19.662, + "eval_steps_per_second": 0.826, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.7666447162628174, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7441, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.8819239139556885, + "learning_rate": 3.718521527213648e-06, + "loss": 1.6314, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.6030640602111816, + "learning_rate": 3.716490658001625e-06, + "loss": 1.7389, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.705246925354004, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.7408, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 3.181328535079956, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7092, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 3.0449445247650146, + "learning_rate": 3.710398050365557e-06, + "loss": 1.7399, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.650517225265503, + "learning_rate": 3.708367181153534e-06, + "loss": 1.6861, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.7593672275543213, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.6701, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 3.0486466884613037, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.7191, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.502850294113159, + "learning_rate": 3.702274573517466e-06, + "loss": 1.6982, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 1.9228183031082153, + "eval_runtime": 49.565, + "eval_samples_per_second": 20.176, + "eval_steps_per_second": 0.847, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.400770664215088, + "learning_rate": 3.700243704305443e-06, + "loss": 1.6666, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.5583295822143555, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.7971, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.519505739212036, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.7076, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 3.106309413909912, + "learning_rate": 3.694151096669375e-06, + "loss": 1.7279, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.516906499862671, + "learning_rate": 3.692120227457352e-06, + "loss": 1.6265, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.603166103363037, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.7122, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.842773199081421, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.7312, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.8860342502593994, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7229, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.517254114151001, + "learning_rate": 3.683996750609261e-06, + "loss": 1.7342, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.430025100708008, + "learning_rate": 3.681965881397238e-06, + "loss": 1.6817, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 1.9156773090362549, + "eval_runtime": 44.6903, + "eval_samples_per_second": 22.376, + "eval_steps_per_second": 0.94, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.340075969696045, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.7044, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.4758968353271484, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.6935, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.878903865814209, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.7021, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.388366222381592, + "learning_rate": 3.673842404549147e-06, + "loss": 1.7229, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.5789167881011963, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8112, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.743072509765625, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.7239, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.5485000610351562, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.7641, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.53190541267395, + "learning_rate": 3.665718927701056e-06, + "loss": 1.6996, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.3354969024658203, + "learning_rate": 3.663688058489034e-06, + "loss": 1.6623, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.37371826171875, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.7109, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 1.9122159481048584, + "eval_runtime": 43.8844, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.406137228012085, + "learning_rate": 3.659626320064988e-06, + "loss": 1.6958, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.4220657348632812, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7196, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.6203134059906006, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.6014, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.5388472080230713, + "learning_rate": 3.65353371242892e-06, + "loss": 1.7272, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.672658681869507, + "learning_rate": 3.651502843216897e-06, + "loss": 1.6786, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4366531372070312, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7198, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.4872887134552, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.7082, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.371206283569336, + "learning_rate": 3.645410235580829e-06, + "loss": 1.6964, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.496103286743164, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.716, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.658316135406494, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.7263, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 1.9115047454833984, + "eval_runtime": 43.8788, + "eval_samples_per_second": 22.79, + "eval_steps_per_second": 0.957, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.5931882858276367, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.7586, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.4697139263153076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.7232, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.7274153232574463, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.6917, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.683138132095337, + "learning_rate": 3.633225020308692e-06, + "loss": 1.7518, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.4510817527770996, + "learning_rate": 3.63119415109667e-06, + "loss": 1.6533, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.6939656734466553, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.6545, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.572864532470703, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.6965, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.4816951751708984, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.6851, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.640394926071167, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7336, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.6707329750061035, + "learning_rate": 3.621039805036556e-06, + "loss": 1.7182, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 1.91274094581604, + "eval_runtime": 44.35, + "eval_samples_per_second": 22.548, + "eval_steps_per_second": 0.947, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.6947290897369385, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.7255, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.5496068000793457, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.6209, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.6196374893188477, + "learning_rate": 3.614947197400488e-06, + "loss": 1.7277, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.4534807205200195, + "learning_rate": 3.612916328188465e-06, + "loss": 1.7404, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.5428082942962646, + "learning_rate": 3.610885458976442e-06, + "loss": 1.6618, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.5431482791900635, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.6743, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.707956552505493, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.7483, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.5643982887268066, + "learning_rate": 3.604792851340374e-06, + "loss": 1.7304, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.4612927436828613, + "learning_rate": 3.602761982128351e-06, + "loss": 1.6561, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.3140358924865723, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.6834, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 1.9124692678451538, + "eval_runtime": 44.0522, + "eval_samples_per_second": 22.7, + "eval_steps_per_second": 0.953, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4636526107788086, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.7395, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.541180372238159, + "learning_rate": 3.596669374492283e-06, + "loss": 1.6579, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.621335983276367, + "learning_rate": 3.59463850528026e-06, + "loss": 1.7332, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.363450050354004, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.7047, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.8609447479248047, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7416, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.457338571548462, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.7516, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.3943030834198, + "learning_rate": 3.586515028432169e-06, + "loss": 1.6436, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.4719340801239014, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.6988, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3442330360412598, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.7165, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.3391900062561035, + "learning_rate": 3.580422420796101e-06, + "loss": 1.6263, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 1.9100929498672485, + "eval_runtime": 44.2504, + "eval_samples_per_second": 22.599, + "eval_steps_per_second": 0.949, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.533566474914551, + "learning_rate": 3.578391551584078e-06, + "loss": 1.7141, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.464540481567383, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.7335, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.4789979457855225, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7074, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.8252034187316895, + "learning_rate": 3.57229894394801e-06, + "loss": 1.6808, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.5870537757873535, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.7171, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.349393129348755, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.6687, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.3343207836151123, + "learning_rate": 3.566206336311942e-06, + "loss": 1.6351, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.558645248413086, + "learning_rate": 3.564175467099919e-06, + "loss": 1.708, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.7598931789398193, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.6797, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.6048543453216553, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7117, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 1.906927466392517, + "eval_runtime": 44.5189, + "eval_samples_per_second": 22.462, + "eval_steps_per_second": 0.943, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.6124815940856934, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.7328, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.3410215377807617, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.6932, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.5247764587402344, + "learning_rate": 3.554021121039805e-06, + "loss": 1.6628, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.642958164215088, + "learning_rate": 3.551990251827783e-06, + "loss": 1.7035, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.3978517055511475, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.6796, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.3969287872314453, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.6626, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.5723509788513184, + "learning_rate": 3.545897644191714e-06, + "loss": 1.7135, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.545154571533203, + "learning_rate": 3.543866774979692e-06, + "loss": 1.6626, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.501058340072632, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.7112, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.695483684539795, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.6773, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 1.9063464403152466, + "eval_runtime": 44.2034, + "eval_samples_per_second": 22.623, + "eval_steps_per_second": 0.95, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 3.0584375858306885, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.7242, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.445220947265625, + "learning_rate": 3.535743298131601e-06, + "loss": 1.7349, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.427220106124878, + "learning_rate": 3.533712428919578e-06, + "loss": 1.677, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.5148212909698486, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.7007, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.644319534301758, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.7442, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.6118004322052, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.7509, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.5224156379699707, + "learning_rate": 3.525588952071487e-06, + "loss": 1.7279, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.4668405055999756, + "learning_rate": 3.523558082859464e-06, + "loss": 1.6754, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 3.8330910205841064, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.6815, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.569058895111084, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.6518, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 1.906802773475647, + "eval_runtime": 50.8521, + "eval_samples_per_second": 19.665, + "eval_steps_per_second": 0.826, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.705265522003174, + "learning_rate": 3.517465475223396e-06, + "loss": 1.7172, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 3.2667057514190674, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7641, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 3.5119211673736572, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.7178, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.7348034381866455, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.6815, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 3.019892454147339, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.6417, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 3.2486684322357178, + "learning_rate": 3.507311129163282e-06, + "loss": 1.6888, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.897071361541748, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7121, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.7342352867126465, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.6339, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.691178560256958, + "learning_rate": 3.501218521527214e-06, + "loss": 1.7069, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.548682689666748, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7218, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 1.9029825925827026, + "eval_runtime": 48.287, + "eval_samples_per_second": 20.71, + "eval_steps_per_second": 0.87, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.422056198120117, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.6486, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.4367074966430664, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.759, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.48002290725708, + "learning_rate": 3.493095044679123e-06, + "loss": 1.5902, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.5030295848846436, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.7108, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.5892179012298584, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.6154, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.4282920360565186, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.6807, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.4826700687408447, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7284, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.416597843170166, + "learning_rate": 3.482940698619009e-06, + "loss": 1.6972, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.4664933681488037, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.7119, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.541856288909912, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.6752, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 1.8999215364456177, + "eval_runtime": 46.3528, + "eval_samples_per_second": 21.574, + "eval_steps_per_second": 0.906, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4803109169006348, + "learning_rate": 3.476848090982941e-06, + "loss": 1.776, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.4363205432891846, + "learning_rate": 3.474817221770918e-06, + "loss": 1.7196, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.4377009868621826, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.6937, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.425791025161743, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.6547, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3216712474823, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.6789, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.65527081489563, + "learning_rate": 3.466693744922827e-06, + "loss": 1.7231, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.3415145874023438, + "learning_rate": 3.464662875710805e-06, + "loss": 1.6805, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.3864822387695312, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.7153, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.490896224975586, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.6866, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.339711904525757, + "learning_rate": 3.458570268074736e-06, + "loss": 1.6732, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 1.8985264301300049, + "eval_runtime": 46.7854, + "eval_samples_per_second": 21.374, + "eval_steps_per_second": 0.898, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.393709659576416, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.6549, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.564114809036255, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.7019, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3826403617858887, + "learning_rate": 3.452477660438668e-06, + "loss": 1.6427, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.3185465335845947, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.6916, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4503891468048096, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.6793, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.386446714401245, + "learning_rate": 3.4463850528026e-06, + "loss": 1.7186, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.6404714584350586, + "learning_rate": 3.444354183590577e-06, + "loss": 1.6605, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.342374324798584, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.6219, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.7503273487091064, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.693, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.5416243076324463, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7323, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 1.9016659259796143, + "eval_runtime": 48.5456, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.5787603855133057, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.7057, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.649528980255127, + "learning_rate": 3.434199837530463e-06, + "loss": 1.6348, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.504089832305908, + "learning_rate": 3.432168968318441e-06, + "loss": 1.7202, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.5624678134918213, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7351, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.641331911087036, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.7453, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.599500894546509, + "learning_rate": 3.426076360682372e-06, + "loss": 1.6942, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.668599843978882, + "learning_rate": 3.42404549147035e-06, + "loss": 1.6337, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.75667667388916, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.6988, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.641526937484741, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.7641, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.4972548484802246, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.7269, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 1.8979382514953613, + "eval_runtime": 45.6555, + "eval_samples_per_second": 21.903, + "eval_steps_per_second": 0.92, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.5454585552215576, + "learning_rate": 3.415922014622259e-06, + "loss": 1.7294, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.5921928882598877, + "learning_rate": 3.413891145410236e-06, + "loss": 1.7145, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.4028453826904297, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.6985, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.604522705078125, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.6586, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.528970241546631, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.734, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.565030574798584, + "learning_rate": 3.405767668562145e-06, + "loss": 1.6997, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.7735602855682373, + "learning_rate": 3.403736799350122e-06, + "loss": 1.6543, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.5116474628448486, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.7711, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.4827239513397217, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.6831, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.544309377670288, + "learning_rate": 3.397644191714054e-06, + "loss": 1.6905, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 1.8945367336273193, + "eval_runtime": 46.6027, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.456636905670166, + "learning_rate": 3.395613322502031e-06, + "loss": 1.646, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.5900352001190186, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.6967, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 3.3344573974609375, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.6947, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.5621678829193115, + "learning_rate": 3.389520714865963e-06, + "loss": 1.7084, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.5676684379577637, + "learning_rate": 3.38748984565394e-06, + "loss": 1.6788, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.600691318511963, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.6811, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.5324196815490723, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.661, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.7559397220611572, + "learning_rate": 3.381397238017872e-06, + "loss": 1.7473, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.526397466659546, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.3752379417419434, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.689, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 1.8995447158813477, + "eval_runtime": 48.0083, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.5902793407440186, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.6899, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 3.157573938369751, + "learning_rate": 3.373273761169781e-06, + "loss": 1.6792, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.628035545349121, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.7003, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.4810822010040283, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.6489, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.4859938621520996, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.6768, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.6773345470428467, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7046, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.9695260524749756, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.7341, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.8146679401397705, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.7472, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.534985065460205, + "learning_rate": 3.359057676685622e-06, + "loss": 1.6199, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.9351718425750732, + "learning_rate": 3.357026807473599e-06, + "loss": 1.6656, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 1.8962808847427368, + "eval_runtime": 47.0443, + "eval_samples_per_second": 21.257, + "eval_steps_per_second": 0.893, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.4404256343841553, + "learning_rate": 3.354995938261576e-06, + "loss": 1.6207, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.6691436767578125, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.6627, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.5436172485351562, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.6919, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.7167909145355225, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.6434, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.575446367263794, + "learning_rate": 3.346872461413485e-06, + "loss": 1.6554, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.3874504566192627, + "learning_rate": 3.344841592201463e-06, + "loss": 1.6691, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.66317081451416, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.6864, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.4396398067474365, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.7217, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.5002803802490234, + "learning_rate": 3.338748984565394e-06, + "loss": 1.7556, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.679577350616455, + "learning_rate": 3.336718115353372e-06, + "loss": 1.6822, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 1.8942477703094482, + "eval_runtime": 44.3804, + "eval_samples_per_second": 22.532, + "eval_steps_per_second": 0.946, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.6119370460510254, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.662, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.6024818420410156, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.687, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 4.128854751586914, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.6898, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3910412788391113, + "learning_rate": 3.32859463850528e-06, + "loss": 1.6274, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.8630096912384033, + "learning_rate": 3.326563769293258e-06, + "loss": 1.6554, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 4.201050281524658, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.6676, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.492168664932251, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.6319, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 3.6807360649108887, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.6776, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.7271926403045654, + "learning_rate": 3.318440292445167e-06, + "loss": 1.6383, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.5567307472229004, + "learning_rate": 3.316409423233144e-06, + "loss": 1.7325, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 1.892768383026123, + "eval_runtime": 50.909, + "eval_samples_per_second": 19.643, + "eval_steps_per_second": 0.825, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.5373101234436035, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.689, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.602923631668091, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.6861, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.507110118865967, + "learning_rate": 3.310316815597076e-06, + "loss": 1.7575, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.361786127090454, + "learning_rate": 3.308285946385053e-06, + "loss": 1.6267, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5360615253448486, + "learning_rate": 3.30625507717303e-06, + "loss": 1.6911, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.6238744258880615, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.7193, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.5358500480651855, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.6691, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.7285404205322266, + "learning_rate": 3.300162469536962e-06, + "loss": 1.7249, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.346160650253296, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.6958, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.5415399074554443, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.7544, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 1.8925012350082397, + "eval_runtime": 46.4957, + "eval_samples_per_second": 21.507, + "eval_steps_per_second": 0.903, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.5924015045166016, + "learning_rate": 3.294069861900894e-06, + "loss": 1.6901, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.4649693965911865, + "learning_rate": 3.292038992688871e-06, + "loss": 1.7234, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.3648035526275635, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.7324, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.5146145820617676, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.631, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.5919511318206787, + "learning_rate": 3.285946385052803e-06, + "loss": 1.706, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.299527168273926, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.608, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.480637788772583, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.7415, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.458183526992798, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.6806, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.4864983558654785, + "learning_rate": 3.277822908204712e-06, + "loss": 1.6373, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.3432068824768066, + "learning_rate": 3.275792038992689e-06, + "loss": 1.6832, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 1.8899296522140503, + "eval_runtime": 44.3309, + "eval_samples_per_second": 22.558, + "eval_steps_per_second": 0.947, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.365891218185425, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.6492, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.463261842727661, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.6847, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.3977205753326416, + "learning_rate": 3.269699431356621e-06, + "loss": 1.6789, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.4172558784484863, + "learning_rate": 3.267668562144598e-06, + "loss": 1.6495, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.663381814956665, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.7139, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.3702306747436523, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.6876, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 3.1759073734283447, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.7465, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.6344592571258545, + "learning_rate": 3.259545085296507e-06, + "loss": 1.6339, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.364490032196045, + "learning_rate": 3.257514216084484e-06, + "loss": 1.7394, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.5639748573303223, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.6928, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 1.8937300443649292, + "eval_runtime": 43.9356, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 3.152968645095825, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.6693, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.45776104927063, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.6999, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.5289807319641113, + "learning_rate": 3.249390739236393e-06, + "loss": 1.6523, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.4168646335601807, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.6823, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.5115208625793457, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.6816, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.685535192489624, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.6307, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.253084897994995, + "learning_rate": 3.241267262388302e-06, + "loss": 1.662, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.4825408458709717, + "learning_rate": 3.23923639317628e-06, + "loss": 1.6444, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.691168785095215, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.6912, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.978756904602051, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7536, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 1.8903727531433105, + "eval_runtime": 43.4099, + "eval_samples_per_second": 23.036, + "eval_steps_per_second": 0.968, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.317105531692505, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.6441, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 3.1030166149139404, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.7205, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.595792770385742, + "learning_rate": 3.229082047116166e-06, + "loss": 1.677, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.3315582275390625, + "learning_rate": 3.227051177904143e-06, + "loss": 1.6779, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.4262566566467285, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.7003, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.4438798427581787, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.6081, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.573531150817871, + "learning_rate": 3.220958570268075e-06, + "loss": 1.6489, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.477809429168701, + "learning_rate": 3.218927701056052e-06, + "loss": 1.6025, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.5499629974365234, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.7337, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.4699153900146484, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.6443, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 1.8867121934890747, + "eval_runtime": 42.9343, + "eval_samples_per_second": 23.291, + "eval_steps_per_second": 0.978, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.4200708866119385, + "learning_rate": 3.212835093419984e-06, + "loss": 1.6743, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.418069839477539, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.723, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.452284336090088, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.6358, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.4241347312927246, + "learning_rate": 3.206742485783916e-06, + "loss": 1.638, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.461860418319702, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.7017, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.591986894607544, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7065, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.478470802307129, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.6262, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.49831223487854, + "learning_rate": 3.198619008935825e-06, + "loss": 1.6871, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.485332727432251, + "learning_rate": 3.196588139723802e-06, + "loss": 1.6864, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.533710479736328, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.6794, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 1.88768470287323, + "eval_runtime": 43.5948, + "eval_samples_per_second": 22.938, + "eval_steps_per_second": 0.963, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.368609666824341, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.6877, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.4875223636627197, + "learning_rate": 3.190495532087734e-06, + "loss": 1.7406, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.5075156688690186, + "learning_rate": 3.188464662875711e-06, + "loss": 1.6545, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.428731918334961, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.6213, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.467618465423584, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.6263, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.561612129211426, + "learning_rate": 3.182372055239643e-06, + "loss": 1.6987, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.5035271644592285, + "learning_rate": 3.18034118602762e-06, + "loss": 1.6902, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.538311243057251, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.6523, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.486321210861206, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.6665, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.433173894882202, + "learning_rate": 3.174248578391552e-06, + "loss": 1.6605, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 1.8853652477264404, + "eval_runtime": 43.3417, + "eval_samples_per_second": 23.072, + "eval_steps_per_second": 0.969, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.4658470153808594, + "learning_rate": 3.172217709179529e-06, + "loss": 1.6615, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.452450752258301, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.6332, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.5310161113739014, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.6846, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.4504501819610596, + "learning_rate": 3.166125101543461e-06, + "loss": 1.6996, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.6004488468170166, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.6302, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.5000431537628174, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.6742, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.4869067668914795, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.6797, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.5018959045410156, + "learning_rate": 3.15800162469537e-06, + "loss": 1.673, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.6645195484161377, + "learning_rate": 3.155970755483347e-06, + "loss": 1.6908, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.3708560466766357, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.683, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 1.8829350471496582, + "eval_runtime": 43.5383, + "eval_samples_per_second": 22.968, + "eval_steps_per_second": 0.965, + "step": 1200 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.740771291234304e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1200/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/rng_state_0.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..491d7aa6873ac345ad7d2d8238b3751df06748aa --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:dc77a0f3d732b76deec0b9f9ee77f2ad22c4c8a301af0167ff41a26d22d5beb3 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/rng_state_1.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..0c1653c04de59e53b4d05673f04ed4b2c7039c73 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7a2973146fd35e5e1a515f6332b8a2b33728c4360853ce2c03e1a7a90b79c2cc +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..4cf5bfa682e62dbe5c2e496536a84208dc8433d0 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8d1658c107e9585fdf02f4eb4b846349794600f9e1c2f5151d959824c6021559 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/scheduler.pt b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..25680b651f6f84b789bafbb387d5670b713521f0 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:38b68aa5e184f913f58a85a7a9033a1733ce1331af7ab0763ca9988b515959ae +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/tokenizer.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..dfd970187355ef9a2c49e3c09aec6760d0822b36 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/trainer_state.json @@ -0,0 +1,11733 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.643835616438356, + "eval_steps": 10, + "global_step": 1500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.5818793773651123, + "learning_rate": 4.569455727051178e-06, + "loss": 1.85, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 3.0734691619873047, + "learning_rate": 4.567424857839156e-06, + "loss": 1.8682, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.496253252029419, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9007, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.360891342163086, + "learning_rate": 4.5633631194151105e-06, + "loss": 1.8262, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.5731117725372314, + "learning_rate": 4.561332250203087e-06, + "loss": 1.8369, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.700558662414551, + "learning_rate": 4.559301380991065e-06, + "loss": 1.7825, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.561306953430176, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.8174, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.618889570236206, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.8812, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.4671308994293213, + "learning_rate": 4.553208773354996e-06, + "loss": 1.7959, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.4571878910064697, + "learning_rate": 4.551177904142974e-06, + "loss": 1.8943, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.010777711868286, + "eval_runtime": 43.7424, + "eval_samples_per_second": 22.861, + "eval_steps_per_second": 0.96, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.4786667823791504, + "learning_rate": 4.5491470349309505e-06, + "loss": 1.8147, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.2955000400543213, + "learning_rate": 4.547116165718929e-06, + "loss": 1.8547, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.3873913288116455, + "learning_rate": 4.545085296506905e-06, + "loss": 1.8156, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.2449734210968018, + "learning_rate": 4.543054427294882e-06, + "loss": 1.8053, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.3969857692718506, + "learning_rate": 4.54102355808286e-06, + "loss": 1.8085, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.3446357250213623, + "learning_rate": 4.538992688870837e-06, + "loss": 1.8355, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.408762216567993, + "learning_rate": 4.536961819658814e-06, + "loss": 1.8319, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.362962245941162, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.8422, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.402984857559204, + "learning_rate": 4.532900081234769e-06, + "loss": 1.8125, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.3964452743530273, + "learning_rate": 4.530869212022746e-06, + "loss": 1.8417, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.006908416748047, + "eval_runtime": 44.2701, + "eval_samples_per_second": 22.589, + "eval_steps_per_second": 0.949, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.595233201980591, + "learning_rate": 4.528838342810723e-06, + "loss": 1.8275, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.326117753982544, + "learning_rate": 4.526807473598701e-06, + "loss": 1.8405, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.4904465675354004, + "learning_rate": 4.524776604386678e-06, + "loss": 1.8172, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.4591610431671143, + "learning_rate": 4.522745735174655e-06, + "loss": 1.876, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.319800853729248, + "learning_rate": 4.520714865962632e-06, + "loss": 1.8294, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.28790020942688, + "learning_rate": 4.51868399675061e-06, + "loss": 1.8537, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.284822702407837, + "learning_rate": 4.516653127538587e-06, + "loss": 1.8087, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.273820638656616, + "learning_rate": 4.514622258326564e-06, + "loss": 1.7992, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.2025654315948486, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.8248, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 2.3218843936920166, + "learning_rate": 4.510560519902519e-06, + "loss": 1.8647, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.0048956871032715, + "eval_runtime": 44.5972, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.515766143798828, + "learning_rate": 4.508529650690496e-06, + "loss": 1.7673, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.4376487731933594, + "learning_rate": 4.506498781478473e-06, + "loss": 1.7755, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.2801272869110107, + "learning_rate": 4.504467912266451e-06, + "loss": 1.8273, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.4707891941070557, + "learning_rate": 4.502437043054428e-06, + "loss": 1.8971, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.301301956176758, + "learning_rate": 4.500406173842405e-06, + "loss": 1.8345, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.2317373752593994, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.8248, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.447373867034912, + "learning_rate": 4.49634443541836e-06, + "loss": 1.8148, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.337245225906372, + "learning_rate": 4.494313566206336e-06, + "loss": 1.8811, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3717031478881836, + "learning_rate": 4.492282696994314e-06, + "loss": 1.8311, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.287677526473999, + "learning_rate": 4.490251827782291e-06, + "loss": 1.7728, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.004889965057373, + "eval_runtime": 43.8132, + "eval_samples_per_second": 22.824, + "eval_steps_per_second": 0.959, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.441445827484131, + "learning_rate": 4.488220958570269e-06, + "loss": 1.7927, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.3191535472869873, + "learning_rate": 4.486190089358245e-06, + "loss": 1.759, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.3118820190429688, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8509, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.4711859226226807, + "learning_rate": 4.4821283509342e-06, + "loss": 1.8402, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.3142173290252686, + "learning_rate": 4.480097481722178e-06, + "loss": 1.8238, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.2665774822235107, + "learning_rate": 4.478066612510154e-06, + "loss": 1.7826, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.1692490577697754, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.8138, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.412783145904541, + "learning_rate": 4.474004874086109e-06, + "loss": 1.9421, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3692686557769775, + "learning_rate": 4.471974004874087e-06, + "loss": 1.7722, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.3224151134490967, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.8753, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.0016305446624756, + "eval_runtime": 45.564, + "eval_samples_per_second": 21.947, + "eval_steps_per_second": 0.922, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.51043438911438, + "learning_rate": 4.467912266450041e-06, + "loss": 1.8325, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.4568657875061035, + "learning_rate": 4.465881397238018e-06, + "loss": 1.8081, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.3279001712799072, + "learning_rate": 4.463850528025995e-06, + "loss": 1.8439, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.6041171550750732, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.8623, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.3780531883239746, + "learning_rate": 4.45978878960195e-06, + "loss": 1.8269, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.511500597000122, + "learning_rate": 4.457757920389927e-06, + "loss": 1.7749, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.707789659500122, + "learning_rate": 4.455727051177904e-06, + "loss": 1.8157, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.5495495796203613, + "learning_rate": 4.453696181965882e-06, + "loss": 1.8534, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.255876302719116, + "learning_rate": 4.451665312753859e-06, + "loss": 1.8377, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.415097951889038, + "learning_rate": 4.449634443541836e-06, + "loss": 1.8764, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 1.9943474531173706, + "eval_runtime": 42.2985, + "eval_samples_per_second": 23.642, + "eval_steps_per_second": 0.993, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.2604713439941406, + "learning_rate": 4.4476035743298135e-06, + "loss": 1.7898, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.5643842220306396, + "learning_rate": 4.445572705117791e-06, + "loss": 1.7953, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.412548780441284, + "learning_rate": 4.443541835905768e-06, + "loss": 1.8611, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.5423877239227295, + "learning_rate": 4.441510966693745e-06, + "loss": 1.8049, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.4986631870269775, + "learning_rate": 4.439480097481723e-06, + "loss": 1.8759, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.8349273204803467, + "learning_rate": 4.4374492282697e-06, + "loss": 1.7416, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.452017068862915, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8268, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.6425273418426514, + "learning_rate": 4.433387489845654e-06, + "loss": 1.8334, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.328740358352661, + "learning_rate": 4.431356620633632e-06, + "loss": 1.7931, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.262209415435791, + "learning_rate": 4.429325751421609e-06, + "loss": 1.8151, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 1.9917123317718506, + "eval_runtime": 42.4309, + "eval_samples_per_second": 23.568, + "eval_steps_per_second": 0.99, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.2984049320220947, + "learning_rate": 4.427294882209586e-06, + "loss": 1.8187, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.3910019397735596, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.7512, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.4494471549987793, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8202, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.3857171535491943, + "learning_rate": 4.421202274573518e-06, + "loss": 1.8279, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.4526500701904297, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.7136, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.2216501235961914, + "learning_rate": 4.417140536149473e-06, + "loss": 1.7574, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.7639758586883545, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8403, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.3177521228790283, + "learning_rate": 4.413078797725427e-06, + "loss": 1.7898, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.527902364730835, + "learning_rate": 4.411047928513404e-06, + "loss": 1.8184, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.34061336517334, + "learning_rate": 4.409017059301382e-06, + "loss": 1.811, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 1.9882216453552246, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.6462934017181396, + "learning_rate": 4.406986190089358e-06, + "loss": 1.8501, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.2725913524627686, + "learning_rate": 4.404955320877336e-06, + "loss": 1.8262, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.390084981918335, + "learning_rate": 4.402924451665313e-06, + "loss": 1.7842, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.314305305480957, + "learning_rate": 4.400893582453291e-06, + "loss": 1.7692, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.2828257083892822, + "learning_rate": 4.398862713241267e-06, + "loss": 1.8031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.3493599891662598, + "learning_rate": 4.396831844029245e-06, + "loss": 1.8183, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.423365831375122, + "learning_rate": 4.394800974817222e-06, + "loss": 1.8012, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.438015937805176, + "learning_rate": 4.392770105605199e-06, + "loss": 1.8217, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.357172727584839, + "learning_rate": 4.390739236393176e-06, + "loss": 1.8382, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.3383474349975586, + "learning_rate": 4.388708367181154e-06, + "loss": 1.841, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 1.9878787994384766, + "eval_runtime": 44.0694, + "eval_samples_per_second": 22.691, + "eval_steps_per_second": 0.953, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.459385633468628, + "learning_rate": 4.386677497969131e-06, + "loss": 1.8377, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.356788158416748, + "learning_rate": 4.384646628757108e-06, + "loss": 1.8718, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.3217222690582275, + "learning_rate": 4.3826157595450855e-06, + "loss": 1.8073, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.3134331703186035, + "learning_rate": 4.380584890333063e-06, + "loss": 1.7675, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.3273653984069824, + "learning_rate": 4.37855402112104e-06, + "loss": 1.7169, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.611034393310547, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8702, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.2077300548553467, + "learning_rate": 4.3744922826969946e-06, + "loss": 1.7951, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.3393311500549316, + "learning_rate": 4.372461413484972e-06, + "loss": 1.8136, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.3306660652160645, + "learning_rate": 4.370430544272949e-06, + "loss": 1.8372, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.4440250396728516, + "learning_rate": 4.368399675060926e-06, + "loss": 1.844, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 1.98439621925354, + "eval_runtime": 44.4595, + "eval_samples_per_second": 22.492, + "eval_steps_per_second": 0.945, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.205542802810669, + "learning_rate": 4.366368805848904e-06, + "loss": 1.8182, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.6030094623565674, + "learning_rate": 4.364337936636881e-06, + "loss": 1.7524, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.4000444412231445, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8269, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.319526433944702, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.7722, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.1624808311462402, + "learning_rate": 4.358245329000813e-06, + "loss": 1.8081, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.26615309715271, + "learning_rate": 4.35621445978879e-06, + "loss": 1.8049, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.4895036220550537, + "learning_rate": 4.354183590576767e-06, + "loss": 1.7932, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.437161445617676, + "learning_rate": 4.352152721364745e-06, + "loss": 1.772, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.344069480895996, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8213, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.360734224319458, + "learning_rate": 4.348090982940699e-06, + "loss": 1.8268, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 1.9846479892730713, + "eval_runtime": 44.947, + "eval_samples_per_second": 22.248, + "eval_steps_per_second": 0.934, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.59487247467041, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.8303, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.6512107849121094, + "learning_rate": 4.344029244516654e-06, + "loss": 1.7979, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.2900476455688477, + "learning_rate": 4.341998375304631e-06, + "loss": 1.8707, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.409494638442993, + "learning_rate": 4.339967506092607e-06, + "loss": 1.8075, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.527205467224121, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.7884, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.33461332321167, + "learning_rate": 4.335905767668562e-06, + "loss": 1.7839, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.3555047512054443, + "learning_rate": 4.33387489845654e-06, + "loss": 1.8245, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.397148609161377, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.8115, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.232139825820923, + "learning_rate": 4.329813160032495e-06, + "loss": 1.7747, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.1684389114379883, + "learning_rate": 4.327782290820471e-06, + "loss": 1.748, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 1.980297565460205, + "eval_runtime": 42.9183, + "eval_samples_per_second": 23.3, + "eval_steps_per_second": 0.979, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.2454638481140137, + "learning_rate": 4.325751421608449e-06, + "loss": 1.83, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.3264827728271484, + "learning_rate": 4.323720552396426e-06, + "loss": 1.7856, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.298804998397827, + "learning_rate": 4.321689683184404e-06, + "loss": 1.78, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2202727794647217, + "learning_rate": 4.31965881397238e-06, + "loss": 1.7875, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.494479179382324, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.8363, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3520524501800537, + "learning_rate": 4.315597075548335e-06, + "loss": 1.8266, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.373878002166748, + "learning_rate": 4.313566206336312e-06, + "loss": 1.8348, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.4232168197631836, + "learning_rate": 4.311535337124289e-06, + "loss": 1.77, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.2572121620178223, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.7914, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.5817952156066895, + "learning_rate": 4.307473598700244e-06, + "loss": 1.8099, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 1.9784427881240845, + "eval_runtime": 43.2981, + "eval_samples_per_second": 23.096, + "eval_steps_per_second": 0.97, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.646040916442871, + "learning_rate": 4.305442729488221e-06, + "loss": 1.7452, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.370220184326172, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8399, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.4599130153656006, + "learning_rate": 4.301380991064176e-06, + "loss": 1.8273, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 2.4135093688964844, + "learning_rate": 4.299350121852153e-06, + "loss": 1.852, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.371021270751953, + "learning_rate": 4.29731925264013e-06, + "loss": 1.8339, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.6858229637145996, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.792, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.4101266860961914, + "learning_rate": 4.293257514216085e-06, + "loss": 1.7556, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.769761800765991, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8223, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.5777804851531982, + "learning_rate": 4.289195775792039e-06, + "loss": 1.8406, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.376864433288574, + "learning_rate": 4.287164906580017e-06, + "loss": 1.7846, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 1.9758412837982178, + "eval_runtime": 43.9353, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.7461960315704346, + "learning_rate": 4.285134037367994e-06, + "loss": 1.776, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.549576759338379, + "learning_rate": 4.283103168155971e-06, + "loss": 1.7644, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.568074941635132, + "learning_rate": 4.281072298943948e-06, + "loss": 1.7571, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.8195722103118896, + "learning_rate": 4.279041429731926e-06, + "loss": 1.7959, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.471623182296753, + "learning_rate": 4.277010560519903e-06, + "loss": 1.7584, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.664689302444458, + "learning_rate": 4.27497969130788e-06, + "loss": 1.7909, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.684504985809326, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.8205, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.5247528553009033, + "learning_rate": 4.270917952883835e-06, + "loss": 1.8365, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.757972478866577, + "learning_rate": 4.268887083671812e-06, + "loss": 1.7665, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.6032614707946777, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8702, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 1.9716291427612305, + "eval_runtime": 43.5482, + "eval_samples_per_second": 22.963, + "eval_steps_per_second": 0.964, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.5324370861053467, + "learning_rate": 4.264825345247766e-06, + "loss": 1.738, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.730712890625, + "learning_rate": 4.262794476035744e-06, + "loss": 1.7918, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.5554966926574707, + "learning_rate": 4.26076360682372e-06, + "loss": 1.7218, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.4774436950683594, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.7823, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.77431058883667, + "learning_rate": 4.256701868399675e-06, + "loss": 1.794, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.258828639984131, + "learning_rate": 4.254670999187653e-06, + "loss": 1.805, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.384399175643921, + "learning_rate": 4.252640129975629e-06, + "loss": 1.818, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3278932571411133, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.7719, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.207348108291626, + "learning_rate": 4.248578391551584e-06, + "loss": 1.7362, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.4525978565216064, + "learning_rate": 4.246547522339562e-06, + "loss": 1.778, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 1.9694374799728394, + "eval_runtime": 42.5977, + "eval_samples_per_second": 23.475, + "eval_steps_per_second": 0.986, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.377835273742676, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.7926, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.4623866081237793, + "learning_rate": 4.242485783915517e-06, + "loss": 1.7366, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.542219400405884, + "learning_rate": 4.240454914703493e-06, + "loss": 1.7409, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.1955819129943848, + "learning_rate": 4.23842404549147e-06, + "loss": 1.7596, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.5834269523620605, + "learning_rate": 4.236393176279448e-06, + "loss": 1.7343, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.219660520553589, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8089, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.283200740814209, + "learning_rate": 4.232331437855402e-06, + "loss": 1.8248, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.274127244949341, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.825, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.3550686836242676, + "learning_rate": 4.228269699431357e-06, + "loss": 1.7957, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.588027238845825, + "learning_rate": 4.226238830219334e-06, + "loss": 1.7865, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 1.9662212133407593, + "eval_runtime": 43.2661, + "eval_samples_per_second": 23.113, + "eval_steps_per_second": 0.971, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.2445170879364014, + "learning_rate": 4.224207961007311e-06, + "loss": 1.791, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.399174690246582, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.7944, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.586184501647949, + "learning_rate": 4.220146222583266e-06, + "loss": 1.7856, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.4174695014953613, + "learning_rate": 4.218115353371243e-06, + "loss": 1.7842, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2684640884399414, + "learning_rate": 4.21608448415922e-06, + "loss": 1.7482, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 3.1782288551330566, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8286, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.4046335220336914, + "learning_rate": 4.212022745735175e-06, + "loss": 1.7228, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 3.3722083568573, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8044, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.293466329574585, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.7415, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.28663969039917, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8282, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 1.9627619981765747, + "eval_runtime": 43.6092, + "eval_samples_per_second": 22.931, + "eval_steps_per_second": 0.963, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.255870819091797, + "learning_rate": 4.203899268887084e-06, + "loss": 1.7983, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.4964773654937744, + "learning_rate": 4.201868399675061e-06, + "loss": 1.8141, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.3372740745544434, + "learning_rate": 4.199837530463039e-06, + "loss": 1.7142, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.3506968021392822, + "learning_rate": 4.197806661251016e-06, + "loss": 1.7898, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.242410898208618, + "learning_rate": 4.195775792038993e-06, + "loss": 1.7677, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.3864502906799316, + "learning_rate": 4.19374492282697e-06, + "loss": 1.7097, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.482349157333374, + "learning_rate": 4.191714053614948e-06, + "loss": 1.8052, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.317652940750122, + "learning_rate": 4.189683184402924e-06, + "loss": 1.7692, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.2777392864227295, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8246, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.3506035804748535, + "learning_rate": 4.185621445978879e-06, + "loss": 1.7576, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 1.9610412120819092, + "eval_runtime": 43.3119, + "eval_samples_per_second": 23.088, + "eval_steps_per_second": 0.97, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.3011536598205566, + "learning_rate": 4.183590576766857e-06, + "loss": 1.7891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.2614853382110596, + "learning_rate": 4.181559707554833e-06, + "loss": 1.8258, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.219175100326538, + "learning_rate": 4.179528838342811e-06, + "loss": 1.7494, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.4239423274993896, + "learning_rate": 4.177497969130788e-06, + "loss": 1.7611, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.470712423324585, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8281, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.504060745239258, + "learning_rate": 4.173436230706742e-06, + "loss": 1.7641, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.326601982116699, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.7671, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.4280030727386475, + "learning_rate": 4.169374492282697e-06, + "loss": 1.7573, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.455716371536255, + "learning_rate": 4.167343623070675e-06, + "loss": 1.809, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.3029844760894775, + "learning_rate": 4.165312753858652e-06, + "loss": 1.77, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 1.9604709148406982, + "eval_runtime": 44.4627, + "eval_samples_per_second": 22.491, + "eval_steps_per_second": 0.945, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.3833982944488525, + "learning_rate": 4.163281884646629e-06, + "loss": 1.7652, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.2392966747283936, + "learning_rate": 4.161251015434607e-06, + "loss": 1.7904, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.4546353816986084, + "learning_rate": 4.159220146222583e-06, + "loss": 1.7324, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.13592529296875, + "learning_rate": 4.157189277010561e-06, + "loss": 1.759, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.381657361984253, + "learning_rate": 4.155158407798538e-06, + "loss": 1.7871, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.4142940044403076, + "learning_rate": 4.153127538586516e-06, + "loss": 1.7832, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.305697441101074, + "learning_rate": 4.151096669374492e-06, + "loss": 1.7619, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.193429708480835, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.7738, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.3007619380950928, + "learning_rate": 4.147034930950447e-06, + "loss": 1.7839, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.6337807178497314, + "learning_rate": 4.145004061738425e-06, + "loss": 1.7748, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 1.9597856998443604, + "eval_runtime": 43.8644, + "eval_samples_per_second": 22.798, + "eval_steps_per_second": 0.957, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.4294793605804443, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.7001, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.512155532836914, + "learning_rate": 4.140942323314379e-06, + "loss": 1.7334, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.290808916091919, + "learning_rate": 4.138911454102356e-06, + "loss": 1.7522, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.483161211013794, + "learning_rate": 4.136880584890333e-06, + "loss": 1.798, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2973921298980713, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.7847, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.253084659576416, + "learning_rate": 4.132818846466288e-06, + "loss": 1.7812, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.306342840194702, + "learning_rate": 4.130787977254265e-06, + "loss": 1.7886, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.5153164863586426, + "learning_rate": 4.128757108042242e-06, + "loss": 1.7954, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.4122278690338135, + "learning_rate": 4.12672623883022e-06, + "loss": 1.8525, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.4520716667175293, + "learning_rate": 4.124695369618197e-06, + "loss": 1.7648, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 1.9547661542892456, + "eval_runtime": 45.3774, + "eval_samples_per_second": 22.037, + "eval_steps_per_second": 0.926, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2959611415863037, + "learning_rate": 4.122664500406174e-06, + "loss": 1.7626, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.2280471324920654, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.7954, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.3870065212249756, + "learning_rate": 4.118602761982129e-06, + "loss": 1.7377, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.292612075805664, + "learning_rate": 4.116571892770106e-06, + "loss": 1.7, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.4607021808624268, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8338, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.532824993133545, + "learning_rate": 4.112510154346061e-06, + "loss": 1.7549, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2139623165130615, + "learning_rate": 4.110479285134038e-06, + "loss": 1.78, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.4749462604522705, + "learning_rate": 4.108448415922015e-06, + "loss": 1.8034, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.609265089035034, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.7941, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.367753028869629, + "learning_rate": 4.10438667749797e-06, + "loss": 1.6653, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 1.9539440870285034, + "eval_runtime": 48.8438, + "eval_samples_per_second": 20.473, + "eval_steps_per_second": 0.86, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.9723567962646484, + "learning_rate": 4.102355808285947e-06, + "loss": 1.8317, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.388806104660034, + "learning_rate": 4.100324939073924e-06, + "loss": 1.7597, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4882986545562744, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.7708, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.7586395740509033, + "learning_rate": 4.096263200649879e-06, + "loss": 1.7804, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.237025260925293, + "learning_rate": 4.094232331437856e-06, + "loss": 1.7415, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.421330213546753, + "learning_rate": 4.092201462225833e-06, + "loss": 1.734, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.7281360626220703, + "learning_rate": 4.090170593013811e-06, + "loss": 1.7716, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.2200379371643066, + "learning_rate": 4.088139723801787e-06, + "loss": 1.7266, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.465440034866333, + "learning_rate": 4.086108854589765e-06, + "loss": 1.741, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.39835262298584, + "learning_rate": 4.084077985377742e-06, + "loss": 1.757, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 1.9514222145080566, + "eval_runtime": 52.8068, + "eval_samples_per_second": 18.937, + "eval_steps_per_second": 0.795, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.526684522628784, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8195, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.341954469680786, + "learning_rate": 4.080016246953696e-06, + "loss": 1.7196, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.3755977153778076, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8115, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.3952431678771973, + "learning_rate": 4.075954508529651e-06, + "loss": 1.812, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.554089069366455, + "learning_rate": 4.073923639317629e-06, + "loss": 1.8184, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.2319302558898926, + "learning_rate": 4.071892770105605e-06, + "loss": 1.7448, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.284856081008911, + "learning_rate": 4.069861900893583e-06, + "loss": 1.7154, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.4081835746765137, + "learning_rate": 4.06783103168156e-06, + "loss": 1.7311, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.6115968227386475, + "learning_rate": 4.065800162469538e-06, + "loss": 1.7436, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.368004322052002, + "learning_rate": 4.063769293257514e-06, + "loss": 1.7811, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 1.9508955478668213, + "eval_runtime": 46.8711, + "eval_samples_per_second": 21.335, + "eval_steps_per_second": 0.896, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.4607419967651367, + "learning_rate": 4.061738424045492e-06, + "loss": 1.7563, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.2978763580322266, + "learning_rate": 4.059707554833469e-06, + "loss": 1.7755, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.3316237926483154, + "learning_rate": 4.057676685621446e-06, + "loss": 1.8226, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.3699169158935547, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.7755, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.195596218109131, + "learning_rate": 4.053614947197401e-06, + "loss": 1.6916, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.4524662494659424, + "learning_rate": 4.051584077985378e-06, + "loss": 1.774, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.573888063430786, + "learning_rate": 4.049553208773355e-06, + "loss": 1.7953, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.5157253742218018, + "learning_rate": 4.047522339561333e-06, + "loss": 1.7597, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.5588185787200928, + "learning_rate": 4.04549147034931e-06, + "loss": 1.7553, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.8950047492980957, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8647, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 1.945882797241211, + "eval_runtime": 45.481, + "eval_samples_per_second": 21.987, + "eval_steps_per_second": 0.923, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.217116117477417, + "learning_rate": 4.041429731925264e-06, + "loss": 1.7465, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.5601565837860107, + "learning_rate": 4.039398862713242e-06, + "loss": 1.7575, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.4540457725524902, + "learning_rate": 4.037367993501219e-06, + "loss": 1.8291, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.327181100845337, + "learning_rate": 4.035337124289196e-06, + "loss": 1.7409, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.485457181930542, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.7192, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.4613442420959473, + "learning_rate": 4.031275385865151e-06, + "loss": 1.7613, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.303217649459839, + "learning_rate": 4.029244516653128e-06, + "loss": 1.7799, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.3830621242523193, + "learning_rate": 4.027213647441105e-06, + "loss": 1.7045, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.342332601547241, + "learning_rate": 4.025182778229083e-06, + "loss": 1.7246, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.4014711380004883, + "learning_rate": 4.02315190901706e-06, + "loss": 1.7621, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 1.95199453830719, + "eval_runtime": 45.4754, + "eval_samples_per_second": 21.99, + "eval_steps_per_second": 0.924, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.6015400886535645, + "learning_rate": 4.021121039805037e-06, + "loss": 1.7536, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.866978406906128, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8405, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.389810085296631, + "learning_rate": 4.017059301380992e-06, + "loss": 1.7365, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.280442953109741, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8273, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.509330987930298, + "learning_rate": 4.012997562956945e-06, + "loss": 1.7127, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.452078104019165, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.7457, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.368337631225586, + "learning_rate": 4.0089358245329e-06, + "loss": 1.7412, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.4529192447662354, + "learning_rate": 4.006904955320878e-06, + "loss": 1.7434, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.647761821746826, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8013, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.387424945831299, + "learning_rate": 4.002843216896833e-06, + "loss": 1.7948, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 1.946029543876648, + "eval_runtime": 43.9666, + "eval_samples_per_second": 22.745, + "eval_steps_per_second": 0.955, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.9217846393585205, + "learning_rate": 4.000812347684809e-06, + "loss": 1.7384, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.400040864944458, + "learning_rate": 3.998781478472787e-06, + "loss": 1.7496, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.4050440788269043, + "learning_rate": 3.996750609260764e-06, + "loss": 1.7165, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 3.6001529693603516, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8679, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 2.4040098190307617, + "learning_rate": 3.992688870836718e-06, + "loss": 1.7671, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.5645527839660645, + "learning_rate": 3.990658001624696e-06, + "loss": 1.7374, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.799394130706787, + "learning_rate": 3.988627132412673e-06, + "loss": 1.796, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.5481607913970947, + "learning_rate": 3.98659626320065e-06, + "loss": 1.753, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.8851330280303955, + "learning_rate": 3.984565393988627e-06, + "loss": 1.7505, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.6668694019317627, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.7604, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 1.939759612083435, + "eval_runtime": 44.5019, + "eval_samples_per_second": 22.471, + "eval_steps_per_second": 0.944, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.320935010910034, + "learning_rate": 3.980503655564582e-06, + "loss": 1.7749, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.829897880554199, + "learning_rate": 3.978472786352559e-06, + "loss": 1.7763, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.8420908451080322, + "learning_rate": 3.976441917140536e-06, + "loss": 1.7598, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.8183, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.781, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 3.9271891117095947, + "learning_rate": 3.972380178716491e-06, + "loss": 1.7949, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.277897834777832, + "learning_rate": 3.970349309504468e-06, + "loss": 1.707, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 3.241063117980957, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.7175, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.514458417892456, + "learning_rate": 3.966287571080423e-06, + "loss": 1.7764, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.420212745666504, + "learning_rate": 3.9642567018684e-06, + "loss": 1.7618, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 1.9401386976242065, + "eval_runtime": 44.1672, + "eval_samples_per_second": 22.641, + "eval_steps_per_second": 0.951, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.3980538845062256, + "learning_rate": 3.962225832656377e-06, + "loss": 1.6922, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.892646312713623, + "learning_rate": 3.960194963444355e-06, + "loss": 1.7571, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.499260902404785, + "learning_rate": 3.958164094232332e-06, + "loss": 1.7501, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.409695625305176, + "learning_rate": 3.956133225020309e-06, + "loss": 1.7553, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.640069007873535, + "learning_rate": 3.954102355808286e-06, + "loss": 1.7526, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.9101760387420654, + "learning_rate": 3.952071486596264e-06, + "loss": 1.7483, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.2579474449157715, + "learning_rate": 3.950040617384241e-06, + "loss": 1.7379, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 4.039298057556152, + "learning_rate": 3.948009748172218e-06, + "loss": 1.7988, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.3232030868530273, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.7592, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.697164297103882, + "learning_rate": 3.943948009748173e-06, + "loss": 1.7622, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 1.9367746114730835, + "eval_runtime": 43.7013, + "eval_samples_per_second": 22.883, + "eval_steps_per_second": 0.961, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.480297803878784, + "learning_rate": 3.94191714053615e-06, + "loss": 1.8006, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.3624603748321533, + "learning_rate": 3.939886271324127e-06, + "loss": 1.7152, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.417187213897705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.7077, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.6575164794921875, + "learning_rate": 3.935824532900082e-06, + "loss": 1.7744, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.4896557331085205, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8052, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.475693941116333, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.7008, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.3210372924804688, + "learning_rate": 3.929731925264013e-06, + "loss": 1.7339, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.433785915374756, + "learning_rate": 3.927701056051991e-06, + "loss": 1.7544, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 3.181574583053589, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.7229, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.704399347305298, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.7086, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 1.9335075616836548, + "eval_runtime": 47.0248, + "eval_samples_per_second": 21.265, + "eval_steps_per_second": 0.893, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 3.5027623176574707, + "learning_rate": 3.921608448415922e-06, + "loss": 1.7622, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.450943946838379, + "learning_rate": 3.9195775792039e-06, + "loss": 1.7286, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.4603374004364014, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.7725, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.8217649459838867, + "learning_rate": 3.915515840779855e-06, + "loss": 1.7859, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.3308372497558594, + "learning_rate": 3.913484971567831e-06, + "loss": 1.6874, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.624776840209961, + "learning_rate": 3.911454102355808e-06, + "loss": 1.7632, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.4097838401794434, + "learning_rate": 3.909423233143786e-06, + "loss": 1.7576, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.487410545349121, + "learning_rate": 3.907392363931763e-06, + "loss": 1.831, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.647245168685913, + "learning_rate": 3.90536149471974e-06, + "loss": 1.8448, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.397664785385132, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.868, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 1.930379033088684, + "eval_runtime": 51.1662, + "eval_samples_per_second": 19.544, + "eval_steps_per_second": 0.821, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.277433395385742, + "learning_rate": 3.901299756295695e-06, + "loss": 1.7394, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 3.1312496662139893, + "learning_rate": 3.899268887083672e-06, + "loss": 1.7422, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.5011534690856934, + "learning_rate": 3.897238017871649e-06, + "loss": 1.7484, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.8716166019439697, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.7253, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.4346914291381836, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8197, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.348475933074951, + "learning_rate": 3.891145410235581e-06, + "loss": 1.7737, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.3996007442474365, + "learning_rate": 3.889114541023558e-06, + "loss": 1.7333, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.355262279510498, + "learning_rate": 3.887083671811536e-06, + "loss": 1.6902, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.335932731628418, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8206, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.2893190383911133, + "learning_rate": 3.88302193338749e-06, + "loss": 1.7533, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 1.9290764331817627, + "eval_runtime": 46.9373, + "eval_samples_per_second": 21.305, + "eval_steps_per_second": 0.895, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.739842176437378, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.7318, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.587916374206543, + "learning_rate": 3.878960194963445e-06, + "loss": 1.7979, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.406548500061035, + "learning_rate": 3.876929325751422e-06, + "loss": 1.6899, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.440321922302246, + "learning_rate": 3.874898456539399e-06, + "loss": 1.7546, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.320030450820923, + "learning_rate": 3.872867587327377e-06, + "loss": 1.7254, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.3795316219329834, + "learning_rate": 3.870836718115354e-06, + "loss": 1.728, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.299802541732788, + "learning_rate": 3.868805848903331e-06, + "loss": 1.7464, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.469719409942627, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.765, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.612109899520874, + "learning_rate": 3.864744110479286e-06, + "loss": 1.7827, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.623173952102661, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8019, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 1.9268505573272705, + "eval_runtime": 46.77, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.4146616458892822, + "learning_rate": 3.86068237205524e-06, + "loss": 1.7498, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.361994743347168, + "learning_rate": 3.858651502843217e-06, + "loss": 1.6926, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.493215560913086, + "learning_rate": 3.856620633631195e-06, + "loss": 1.736, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.4341812133789062, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8152, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.391331195831299, + "learning_rate": 3.852558895207149e-06, + "loss": 1.7717, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.419492244720459, + "learning_rate": 3.850528025995126e-06, + "loss": 1.7283, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.349090576171875, + "learning_rate": 3.848497156783104e-06, + "loss": 1.7317, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.298647880554199, + "learning_rate": 3.84646628757108e-06, + "loss": 1.7389, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.4612667560577393, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.7333, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.3139944076538086, + "learning_rate": 3.842404549147035e-06, + "loss": 1.6944, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 1.924438714981079, + "eval_runtime": 45.8276, + "eval_samples_per_second": 21.821, + "eval_steps_per_second": 0.916, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 2.17958664894104, + "learning_rate": 3.840373679935013e-06, + "loss": 1.7474, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.337859869003296, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.7345, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4777672290802, + "learning_rate": 3.836311941510967e-06, + "loss": 1.77, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.280158519744873, + "learning_rate": 3.834281072298944e-06, + "loss": 1.7069, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2387664318084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.7889, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.373771905899048, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8268, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.4950132369995117, + "learning_rate": 3.828188464662876e-06, + "loss": 1.7586, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.3296444416046143, + "learning_rate": 3.826157595450853e-06, + "loss": 1.6991, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.6653530597686768, + "learning_rate": 3.82412672623883e-06, + "loss": 1.7505, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.485128402709961, + "learning_rate": 3.822095857026808e-06, + "loss": 1.7196, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 1.9221889972686768, + "eval_runtime": 46.6581, + "eval_samples_per_second": 21.433, + "eval_steps_per_second": 0.9, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.3726181983947754, + "learning_rate": 3.820064987814785e-06, + "loss": 1.7296, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.233992099761963, + "learning_rate": 3.818034118602762e-06, + "loss": 1.7313, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.5388147830963135, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.7435, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.4787333011627197, + "learning_rate": 3.813972380178717e-06, + "loss": 1.7733, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.232564926147461, + "learning_rate": 3.811941510966694e-06, + "loss": 1.693, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.3882386684417725, + "learning_rate": 3.809910641754671e-06, + "loss": 1.6529, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.4785983562469482, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.7456, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.685098886489868, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.7162, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.353523015975952, + "learning_rate": 3.803818034118603e-06, + "loss": 1.7852, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.342163562774658, + "learning_rate": 3.80178716490658e-06, + "loss": 1.7767, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 1.9235820770263672, + "eval_runtime": 47.1327, + "eval_samples_per_second": 21.217, + "eval_steps_per_second": 0.891, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.550818920135498, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.7328, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.3204798698425293, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.7058, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.4752719402313232, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8051, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.5202794075012207, + "learning_rate": 3.793663688058489e-06, + "loss": 1.7966, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.26521635055542, + "learning_rate": 3.791632818846467e-06, + "loss": 1.7879, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.331515312194824, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.7814, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.37178635597229, + "learning_rate": 3.787571080422421e-06, + "loss": 1.7635, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.321417808532715, + "learning_rate": 3.785540211210398e-06, + "loss": 1.7513, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.485989809036255, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.7306, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.7497379779815674, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8409, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 1.9275448322296143, + "eval_runtime": 46.2088, + "eval_samples_per_second": 21.641, + "eval_steps_per_second": 0.909, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 4.9214019775390625, + "learning_rate": 3.77944760357433e-06, + "loss": 1.7834, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 4.505277156829834, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.7179, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.5772030353546143, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.7766, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.254185438156128, + "learning_rate": 3.773354995938262e-06, + "loss": 1.7697, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 3.632233142852783, + "learning_rate": 3.771324126726239e-06, + "loss": 1.7968, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.174461841583252, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.7346, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.3983266353607178, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.6534, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 3.393791913986206, + "learning_rate": 3.765231519090171e-06, + "loss": 1.7561, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.2853987216949463, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.6963, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.649789810180664, + "learning_rate": 3.761169780666125e-06, + "loss": 1.7395, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 1.9242008924484253, + "eval_runtime": 46.6017, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 3.417872190475464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.784, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.316666841506958, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.7154, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.486161231994629, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.7903, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 4.538443565368652, + "learning_rate": 3.753046303818034e-06, + "loss": 1.7711, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.9772889614105225, + "learning_rate": 3.751015434606012e-06, + "loss": 1.8046, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 7.2353715896606445, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.7189, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.736, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.8026, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 12.551762580871582, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.7788, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 3.0898046493530273, + "learning_rate": 3.742891957757921e-06, + "loss": 1.7878, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 1.9646916389465332, + "eval_runtime": 44.1583, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 8.120022773742676, + "learning_rate": 3.740861088545898e-06, + "loss": 1.768, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 6.485774040222168, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.746, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.85135555267334, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.6968, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 5.8551435470581055, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.6796, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 4.45880126953125, + "learning_rate": 3.732737611697807e-06, + "loss": 1.7219, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 3.0435831546783447, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.6959, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.8194847106933594, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.8152, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 3.1266117095947266, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.6937, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 3.642854690551758, + "learning_rate": 3.724614134849716e-06, + "loss": 1.7992, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 3.344496965408325, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.7583, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 1.9293720722198486, + "eval_runtime": 50.8602, + "eval_samples_per_second": 19.662, + "eval_steps_per_second": 0.826, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.7666447162628174, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7441, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.8819239139556885, + "learning_rate": 3.718521527213648e-06, + "loss": 1.6314, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.6030640602111816, + "learning_rate": 3.716490658001625e-06, + "loss": 1.7389, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.705246925354004, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.7408, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 3.181328535079956, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7092, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 3.0449445247650146, + "learning_rate": 3.710398050365557e-06, + "loss": 1.7399, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.650517225265503, + "learning_rate": 3.708367181153534e-06, + "loss": 1.6861, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.7593672275543213, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.6701, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 3.0486466884613037, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.7191, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.502850294113159, + "learning_rate": 3.702274573517466e-06, + "loss": 1.6982, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 1.9228183031082153, + "eval_runtime": 49.565, + "eval_samples_per_second": 20.176, + "eval_steps_per_second": 0.847, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.400770664215088, + "learning_rate": 3.700243704305443e-06, + "loss": 1.6666, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.5583295822143555, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.7971, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.519505739212036, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.7076, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 3.106309413909912, + "learning_rate": 3.694151096669375e-06, + "loss": 1.7279, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.516906499862671, + "learning_rate": 3.692120227457352e-06, + "loss": 1.6265, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.603166103363037, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.7122, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.842773199081421, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.7312, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.8860342502593994, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7229, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.517254114151001, + "learning_rate": 3.683996750609261e-06, + "loss": 1.7342, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.430025100708008, + "learning_rate": 3.681965881397238e-06, + "loss": 1.6817, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 1.9156773090362549, + "eval_runtime": 44.6903, + "eval_samples_per_second": 22.376, + "eval_steps_per_second": 0.94, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.340075969696045, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.7044, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.4758968353271484, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.6935, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.878903865814209, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.7021, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.388366222381592, + "learning_rate": 3.673842404549147e-06, + "loss": 1.7229, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.5789167881011963, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8112, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.743072509765625, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.7239, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.5485000610351562, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.7641, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.53190541267395, + "learning_rate": 3.665718927701056e-06, + "loss": 1.6996, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.3354969024658203, + "learning_rate": 3.663688058489034e-06, + "loss": 1.6623, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.37371826171875, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.7109, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 1.9122159481048584, + "eval_runtime": 43.8844, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.406137228012085, + "learning_rate": 3.659626320064988e-06, + "loss": 1.6958, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.4220657348632812, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7196, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.6203134059906006, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.6014, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.5388472080230713, + "learning_rate": 3.65353371242892e-06, + "loss": 1.7272, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.672658681869507, + "learning_rate": 3.651502843216897e-06, + "loss": 1.6786, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4366531372070312, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7198, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.4872887134552, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.7082, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.371206283569336, + "learning_rate": 3.645410235580829e-06, + "loss": 1.6964, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.496103286743164, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.716, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.658316135406494, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.7263, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 1.9115047454833984, + "eval_runtime": 43.8788, + "eval_samples_per_second": 22.79, + "eval_steps_per_second": 0.957, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.5931882858276367, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.7586, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.4697139263153076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.7232, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.7274153232574463, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.6917, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.683138132095337, + "learning_rate": 3.633225020308692e-06, + "loss": 1.7518, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.4510817527770996, + "learning_rate": 3.63119415109667e-06, + "loss": 1.6533, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.6939656734466553, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.6545, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.572864532470703, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.6965, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.4816951751708984, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.6851, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.640394926071167, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7336, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.6707329750061035, + "learning_rate": 3.621039805036556e-06, + "loss": 1.7182, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 1.91274094581604, + "eval_runtime": 44.35, + "eval_samples_per_second": 22.548, + "eval_steps_per_second": 0.947, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.6947290897369385, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.7255, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.5496068000793457, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.6209, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.6196374893188477, + "learning_rate": 3.614947197400488e-06, + "loss": 1.7277, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.4534807205200195, + "learning_rate": 3.612916328188465e-06, + "loss": 1.7404, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.5428082942962646, + "learning_rate": 3.610885458976442e-06, + "loss": 1.6618, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.5431482791900635, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.6743, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.707956552505493, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.7483, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.5643982887268066, + "learning_rate": 3.604792851340374e-06, + "loss": 1.7304, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.4612927436828613, + "learning_rate": 3.602761982128351e-06, + "loss": 1.6561, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.3140358924865723, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.6834, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 1.9124692678451538, + "eval_runtime": 44.0522, + "eval_samples_per_second": 22.7, + "eval_steps_per_second": 0.953, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4636526107788086, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.7395, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.541180372238159, + "learning_rate": 3.596669374492283e-06, + "loss": 1.6579, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.621335983276367, + "learning_rate": 3.59463850528026e-06, + "loss": 1.7332, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.363450050354004, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.7047, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.8609447479248047, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7416, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.457338571548462, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.7516, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.3943030834198, + "learning_rate": 3.586515028432169e-06, + "loss": 1.6436, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.4719340801239014, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.6988, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3442330360412598, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.7165, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.3391900062561035, + "learning_rate": 3.580422420796101e-06, + "loss": 1.6263, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 1.9100929498672485, + "eval_runtime": 44.2504, + "eval_samples_per_second": 22.599, + "eval_steps_per_second": 0.949, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.533566474914551, + "learning_rate": 3.578391551584078e-06, + "loss": 1.7141, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.464540481567383, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.7335, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.4789979457855225, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7074, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.8252034187316895, + "learning_rate": 3.57229894394801e-06, + "loss": 1.6808, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.5870537757873535, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.7171, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.349393129348755, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.6687, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.3343207836151123, + "learning_rate": 3.566206336311942e-06, + "loss": 1.6351, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.558645248413086, + "learning_rate": 3.564175467099919e-06, + "loss": 1.708, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.7598931789398193, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.6797, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.6048543453216553, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7117, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 1.906927466392517, + "eval_runtime": 44.5189, + "eval_samples_per_second": 22.462, + "eval_steps_per_second": 0.943, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.6124815940856934, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.7328, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.3410215377807617, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.6932, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.5247764587402344, + "learning_rate": 3.554021121039805e-06, + "loss": 1.6628, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.642958164215088, + "learning_rate": 3.551990251827783e-06, + "loss": 1.7035, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.3978517055511475, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.6796, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.3969287872314453, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.6626, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.5723509788513184, + "learning_rate": 3.545897644191714e-06, + "loss": 1.7135, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.545154571533203, + "learning_rate": 3.543866774979692e-06, + "loss": 1.6626, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.501058340072632, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.7112, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.695483684539795, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.6773, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 1.9063464403152466, + "eval_runtime": 44.2034, + "eval_samples_per_second": 22.623, + "eval_steps_per_second": 0.95, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 3.0584375858306885, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.7242, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.445220947265625, + "learning_rate": 3.535743298131601e-06, + "loss": 1.7349, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.427220106124878, + "learning_rate": 3.533712428919578e-06, + "loss": 1.677, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.5148212909698486, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.7007, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.644319534301758, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.7442, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.6118004322052, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.7509, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.5224156379699707, + "learning_rate": 3.525588952071487e-06, + "loss": 1.7279, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.4668405055999756, + "learning_rate": 3.523558082859464e-06, + "loss": 1.6754, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 3.8330910205841064, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.6815, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.569058895111084, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.6518, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 1.906802773475647, + "eval_runtime": 50.8521, + "eval_samples_per_second": 19.665, + "eval_steps_per_second": 0.826, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.705265522003174, + "learning_rate": 3.517465475223396e-06, + "loss": 1.7172, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 3.2667057514190674, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7641, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 3.5119211673736572, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.7178, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.7348034381866455, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.6815, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 3.019892454147339, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.6417, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 3.2486684322357178, + "learning_rate": 3.507311129163282e-06, + "loss": 1.6888, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.897071361541748, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7121, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.7342352867126465, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.6339, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.691178560256958, + "learning_rate": 3.501218521527214e-06, + "loss": 1.7069, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.548682689666748, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7218, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 1.9029825925827026, + "eval_runtime": 48.287, + "eval_samples_per_second": 20.71, + "eval_steps_per_second": 0.87, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.422056198120117, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.6486, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.4367074966430664, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.759, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.48002290725708, + "learning_rate": 3.493095044679123e-06, + "loss": 1.5902, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.5030295848846436, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.7108, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.5892179012298584, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.6154, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.4282920360565186, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.6807, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.4826700687408447, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7284, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.416597843170166, + "learning_rate": 3.482940698619009e-06, + "loss": 1.6972, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.4664933681488037, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.7119, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.541856288909912, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.6752, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 1.8999215364456177, + "eval_runtime": 46.3528, + "eval_samples_per_second": 21.574, + "eval_steps_per_second": 0.906, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4803109169006348, + "learning_rate": 3.476848090982941e-06, + "loss": 1.776, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.4363205432891846, + "learning_rate": 3.474817221770918e-06, + "loss": 1.7196, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.4377009868621826, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.6937, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.425791025161743, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.6547, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3216712474823, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.6789, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.65527081489563, + "learning_rate": 3.466693744922827e-06, + "loss": 1.7231, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.3415145874023438, + "learning_rate": 3.464662875710805e-06, + "loss": 1.6805, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.3864822387695312, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.7153, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.490896224975586, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.6866, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.339711904525757, + "learning_rate": 3.458570268074736e-06, + "loss": 1.6732, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 1.8985264301300049, + "eval_runtime": 46.7854, + "eval_samples_per_second": 21.374, + "eval_steps_per_second": 0.898, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.393709659576416, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.6549, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.564114809036255, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.7019, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3826403617858887, + "learning_rate": 3.452477660438668e-06, + "loss": 1.6427, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.3185465335845947, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.6916, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4503891468048096, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.6793, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.386446714401245, + "learning_rate": 3.4463850528026e-06, + "loss": 1.7186, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.6404714584350586, + "learning_rate": 3.444354183590577e-06, + "loss": 1.6605, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.342374324798584, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.6219, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.7503273487091064, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.693, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.5416243076324463, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7323, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 1.9016659259796143, + "eval_runtime": 48.5456, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.5787603855133057, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.7057, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.649528980255127, + "learning_rate": 3.434199837530463e-06, + "loss": 1.6348, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.504089832305908, + "learning_rate": 3.432168968318441e-06, + "loss": 1.7202, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.5624678134918213, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7351, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.641331911087036, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.7453, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.599500894546509, + "learning_rate": 3.426076360682372e-06, + "loss": 1.6942, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.668599843978882, + "learning_rate": 3.42404549147035e-06, + "loss": 1.6337, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.75667667388916, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.6988, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.641526937484741, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.7641, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.4972548484802246, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.7269, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 1.8979382514953613, + "eval_runtime": 45.6555, + "eval_samples_per_second": 21.903, + "eval_steps_per_second": 0.92, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.5454585552215576, + "learning_rate": 3.415922014622259e-06, + "loss": 1.7294, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.5921928882598877, + "learning_rate": 3.413891145410236e-06, + "loss": 1.7145, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.4028453826904297, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.6985, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.604522705078125, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.6586, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.528970241546631, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.734, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.565030574798584, + "learning_rate": 3.405767668562145e-06, + "loss": 1.6997, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.7735602855682373, + "learning_rate": 3.403736799350122e-06, + "loss": 1.6543, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.5116474628448486, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.7711, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.4827239513397217, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.6831, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.544309377670288, + "learning_rate": 3.397644191714054e-06, + "loss": 1.6905, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 1.8945367336273193, + "eval_runtime": 46.6027, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.456636905670166, + "learning_rate": 3.395613322502031e-06, + "loss": 1.646, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.5900352001190186, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.6967, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 3.3344573974609375, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.6947, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.5621678829193115, + "learning_rate": 3.389520714865963e-06, + "loss": 1.7084, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.5676684379577637, + "learning_rate": 3.38748984565394e-06, + "loss": 1.6788, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.600691318511963, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.6811, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.5324196815490723, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.661, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.7559397220611572, + "learning_rate": 3.381397238017872e-06, + "loss": 1.7473, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.526397466659546, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.3752379417419434, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.689, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 1.8995447158813477, + "eval_runtime": 48.0083, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.5902793407440186, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.6899, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 3.157573938369751, + "learning_rate": 3.373273761169781e-06, + "loss": 1.6792, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.628035545349121, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.7003, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.4810822010040283, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.6489, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.4859938621520996, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.6768, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.6773345470428467, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7046, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.9695260524749756, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.7341, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.8146679401397705, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.7472, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.534985065460205, + "learning_rate": 3.359057676685622e-06, + "loss": 1.6199, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.9351718425750732, + "learning_rate": 3.357026807473599e-06, + "loss": 1.6656, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 1.8962808847427368, + "eval_runtime": 47.0443, + "eval_samples_per_second": 21.257, + "eval_steps_per_second": 0.893, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.4404256343841553, + "learning_rate": 3.354995938261576e-06, + "loss": 1.6207, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.6691436767578125, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.6627, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.5436172485351562, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.6919, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.7167909145355225, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.6434, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.575446367263794, + "learning_rate": 3.346872461413485e-06, + "loss": 1.6554, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.3874504566192627, + "learning_rate": 3.344841592201463e-06, + "loss": 1.6691, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.66317081451416, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.6864, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.4396398067474365, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.7217, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.5002803802490234, + "learning_rate": 3.338748984565394e-06, + "loss": 1.7556, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.679577350616455, + "learning_rate": 3.336718115353372e-06, + "loss": 1.6822, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 1.8942477703094482, + "eval_runtime": 44.3804, + "eval_samples_per_second": 22.532, + "eval_steps_per_second": 0.946, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.6119370460510254, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.662, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.6024818420410156, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.687, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 4.128854751586914, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.6898, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3910412788391113, + "learning_rate": 3.32859463850528e-06, + "loss": 1.6274, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.8630096912384033, + "learning_rate": 3.326563769293258e-06, + "loss": 1.6554, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 4.201050281524658, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.6676, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.492168664932251, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.6319, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 3.6807360649108887, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.6776, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.7271926403045654, + "learning_rate": 3.318440292445167e-06, + "loss": 1.6383, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.5567307472229004, + "learning_rate": 3.316409423233144e-06, + "loss": 1.7325, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 1.892768383026123, + "eval_runtime": 50.909, + "eval_samples_per_second": 19.643, + "eval_steps_per_second": 0.825, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.5373101234436035, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.689, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.602923631668091, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.6861, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.507110118865967, + "learning_rate": 3.310316815597076e-06, + "loss": 1.7575, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.361786127090454, + "learning_rate": 3.308285946385053e-06, + "loss": 1.6267, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5360615253448486, + "learning_rate": 3.30625507717303e-06, + "loss": 1.6911, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.6238744258880615, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.7193, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.5358500480651855, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.6691, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.7285404205322266, + "learning_rate": 3.300162469536962e-06, + "loss": 1.7249, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.346160650253296, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.6958, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.5415399074554443, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.7544, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 1.8925012350082397, + "eval_runtime": 46.4957, + "eval_samples_per_second": 21.507, + "eval_steps_per_second": 0.903, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.5924015045166016, + "learning_rate": 3.294069861900894e-06, + "loss": 1.6901, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.4649693965911865, + "learning_rate": 3.292038992688871e-06, + "loss": 1.7234, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.3648035526275635, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.7324, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.5146145820617676, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.631, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.5919511318206787, + "learning_rate": 3.285946385052803e-06, + "loss": 1.706, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.299527168273926, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.608, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.480637788772583, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.7415, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.458183526992798, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.6806, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.4864983558654785, + "learning_rate": 3.277822908204712e-06, + "loss": 1.6373, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.3432068824768066, + "learning_rate": 3.275792038992689e-06, + "loss": 1.6832, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 1.8899296522140503, + "eval_runtime": 44.3309, + "eval_samples_per_second": 22.558, + "eval_steps_per_second": 0.947, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.365891218185425, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.6492, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.463261842727661, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.6847, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.3977205753326416, + "learning_rate": 3.269699431356621e-06, + "loss": 1.6789, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.4172558784484863, + "learning_rate": 3.267668562144598e-06, + "loss": 1.6495, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.663381814956665, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.7139, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.3702306747436523, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.6876, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 3.1759073734283447, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.7465, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.6344592571258545, + "learning_rate": 3.259545085296507e-06, + "loss": 1.6339, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.364490032196045, + "learning_rate": 3.257514216084484e-06, + "loss": 1.7394, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.5639748573303223, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.6928, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 1.8937300443649292, + "eval_runtime": 43.9356, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 3.152968645095825, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.6693, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.45776104927063, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.6999, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.5289807319641113, + "learning_rate": 3.249390739236393e-06, + "loss": 1.6523, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.4168646335601807, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.6823, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.5115208625793457, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.6816, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.685535192489624, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.6307, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.253084897994995, + "learning_rate": 3.241267262388302e-06, + "loss": 1.662, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.4825408458709717, + "learning_rate": 3.23923639317628e-06, + "loss": 1.6444, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.691168785095215, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.6912, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.978756904602051, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7536, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 1.8903727531433105, + "eval_runtime": 43.4099, + "eval_samples_per_second": 23.036, + "eval_steps_per_second": 0.968, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.317105531692505, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.6441, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 3.1030166149139404, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.7205, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.595792770385742, + "learning_rate": 3.229082047116166e-06, + "loss": 1.677, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.3315582275390625, + "learning_rate": 3.227051177904143e-06, + "loss": 1.6779, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.4262566566467285, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.7003, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.4438798427581787, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.6081, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.573531150817871, + "learning_rate": 3.220958570268075e-06, + "loss": 1.6489, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.477809429168701, + "learning_rate": 3.218927701056052e-06, + "loss": 1.6025, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.5499629974365234, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.7337, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.4699153900146484, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.6443, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 1.8867121934890747, + "eval_runtime": 42.9343, + "eval_samples_per_second": 23.291, + "eval_steps_per_second": 0.978, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.4200708866119385, + "learning_rate": 3.212835093419984e-06, + "loss": 1.6743, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.418069839477539, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.723, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.452284336090088, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.6358, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.4241347312927246, + "learning_rate": 3.206742485783916e-06, + "loss": 1.638, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.461860418319702, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.7017, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.591986894607544, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7065, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.478470802307129, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.6262, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.49831223487854, + "learning_rate": 3.198619008935825e-06, + "loss": 1.6871, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.485332727432251, + "learning_rate": 3.196588139723802e-06, + "loss": 1.6864, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.533710479736328, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.6794, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 1.88768470287323, + "eval_runtime": 43.5948, + "eval_samples_per_second": 22.938, + "eval_steps_per_second": 0.963, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.368609666824341, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.6877, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.4875223636627197, + "learning_rate": 3.190495532087734e-06, + "loss": 1.7406, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.5075156688690186, + "learning_rate": 3.188464662875711e-06, + "loss": 1.6545, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.428731918334961, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.6213, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.467618465423584, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.6263, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.561612129211426, + "learning_rate": 3.182372055239643e-06, + "loss": 1.6987, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.5035271644592285, + "learning_rate": 3.18034118602762e-06, + "loss": 1.6902, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.538311243057251, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.6523, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.486321210861206, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.6665, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.433173894882202, + "learning_rate": 3.174248578391552e-06, + "loss": 1.6605, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 1.8853652477264404, + "eval_runtime": 43.3417, + "eval_samples_per_second": 23.072, + "eval_steps_per_second": 0.969, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.4658470153808594, + "learning_rate": 3.172217709179529e-06, + "loss": 1.6615, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.452450752258301, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.6332, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.5310161113739014, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.6846, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.4504501819610596, + "learning_rate": 3.166125101543461e-06, + "loss": 1.6996, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.6004488468170166, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.6302, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.5000431537628174, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.6742, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.4869067668914795, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.6797, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.5018959045410156, + "learning_rate": 3.15800162469537e-06, + "loss": 1.673, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.6645195484161377, + "learning_rate": 3.155970755483347e-06, + "loss": 1.6908, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.3708560466766357, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.683, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 1.8829350471496582, + "eval_runtime": 43.5383, + "eval_samples_per_second": 22.968, + "eval_steps_per_second": 0.965, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.574108362197876, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.7155, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.4669814109802246, + "learning_rate": 3.149878147847279e-06, + "loss": 1.7529, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4713149070739746, + "learning_rate": 3.147847278635256e-06, + "loss": 1.6491, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.5128800868988037, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.6276, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.740370035171509, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.6778, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.6038050651550293, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.6527, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.782407283782959, + "learning_rate": 3.139723801787165e-06, + "loss": 1.6989, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.7673134803771973, + "learning_rate": 3.137692932575143e-06, + "loss": 1.6291, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.7472870349884033, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.6776, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.4887819290161133, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7009, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.8792160749435425, + "eval_runtime": 43.7334, + "eval_samples_per_second": 22.866, + "eval_steps_per_second": 0.96, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.4528491497039795, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.7245, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.5177690982818604, + "learning_rate": 3.129569455727051e-06, + "loss": 1.6852, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.9238510131835938, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.6994, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.446660041809082, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.6636, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.644211530685425, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.6983, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.697342872619629, + "learning_rate": 3.12144597887896e-06, + "loss": 1.7172, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.714172124862671, + "learning_rate": 3.119415109666938e-06, + "loss": 1.6702, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.471735715866089, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.6945, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.5089669227600098, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.7181, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.349229335784912, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.7297, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.880064845085144, + "eval_runtime": 47.0636, + "eval_samples_per_second": 21.248, + "eval_steps_per_second": 0.892, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.6843974590301514, + "learning_rate": 3.111291632818847e-06, + "loss": 1.6698, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.4325790405273438, + "learning_rate": 3.109260763606824e-06, + "loss": 1.6973, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.488459825515747, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.6552, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.4992082118988037, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.7114, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.4742603302001953, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.7073, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.556330442428589, + "learning_rate": 3.101137286758733e-06, + "loss": 1.7108, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.5711283683776855, + "learning_rate": 3.09910641754671e-06, + "loss": 1.6811, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.6467158794403076, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.7395, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.465963363647461, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.7006, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.622716188430786, + "learning_rate": 3.093013809910642e-06, + "loss": 1.7385, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.8808013200759888, + "eval_runtime": 45.5126, + "eval_samples_per_second": 21.972, + "eval_steps_per_second": 0.923, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.489701509475708, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.6949, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.4398739337921143, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.6802, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.3654873371124268, + "learning_rate": 3.086921202274574e-06, + "loss": 1.6452, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.518768548965454, + "learning_rate": 3.084890333062551e-06, + "loss": 1.6913, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.63173246383667, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.6611, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.531484603881836, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.611, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.4796738624572754, + "learning_rate": 3.078797725426483e-06, + "loss": 1.6764, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.39241099357605, + "learning_rate": 3.07676685621446e-06, + "loss": 1.6417, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3984735012054443, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.6546, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.5202760696411133, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.6879, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.879974365234375, + "eval_runtime": 44.395, + "eval_samples_per_second": 22.525, + "eval_steps_per_second": 0.946, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.5556983947753906, + "learning_rate": 3.070674248578392e-06, + "loss": 1.6733, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.3853955268859863, + "learning_rate": 3.068643379366369e-06, + "loss": 1.6274, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.4503982067108154, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.6896, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.5618021488189697, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.6592, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.6265292167663574, + "learning_rate": 3.062550771730301e-06, + "loss": 1.6634, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.4617390632629395, + "learning_rate": 3.060519902518278e-06, + "loss": 1.6032, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.4191603660583496, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.6871, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.5509119033813477, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.653, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.5399646759033203, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.7046, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.446967124938965, + "learning_rate": 3.052396425670187e-06, + "loss": 1.6547, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.8803874254226685, + "eval_runtime": 45.361, + "eval_samples_per_second": 22.045, + "eval_steps_per_second": 0.926, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.419206142425537, + "learning_rate": 3.050365556458164e-06, + "loss": 1.6517, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.4592065811157227, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.693, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.5429770946502686, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7124, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.5141642093658447, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.7103, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.362539291381836, + "learning_rate": 3.042242079610073e-06, + "loss": 1.6205, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.401175022125244, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.6479, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.5883190631866455, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.6732, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.5273513793945312, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.6673, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.6569600105285645, + "learning_rate": 3.034118602761982e-06, + "loss": 1.6396, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.8843557834625244, + "learning_rate": 3.03208773354996e-06, + "loss": 1.7136, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.8763896226882935, + "eval_runtime": 46.1886, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.6459810733795166, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.6939, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.4619321823120117, + "learning_rate": 3.028025995125914e-06, + "loss": 1.6197, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.390430450439453, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.6805, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.5494728088378906, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.6116, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.5397655963897705, + "learning_rate": 3.021933387489846e-06, + "loss": 1.6753, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.634716510772705, + "learning_rate": 3.019902518277823e-06, + "loss": 1.6892, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.5844202041625977, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.6401, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.577362060546875, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.6405, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.3554489612579346, + "learning_rate": 3.013809910641755e-06, + "loss": 1.6427, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.893918037414551, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.6739, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.8772739171981812, + "eval_runtime": 44.9972, + "eval_samples_per_second": 22.224, + "eval_steps_per_second": 0.933, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.4818382263183594, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.644, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.4990177154541016, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.6962, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.627502918243408, + "learning_rate": 3.005686433793664e-06, + "loss": 1.639, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.5949175357818604, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.7095, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.422765016555786, + "learning_rate": 3.001624695369618e-06, + "loss": 1.6572, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.708702564239502, + "learning_rate": 2.999593826157596e-06, + "loss": 1.6747, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.624866247177124, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.6643, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.5831358432769775, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.7036, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.5799436569213867, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.6628, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.7047722339630127, + "learning_rate": 2.991470349309505e-06, + "loss": 1.6591, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.8736577033996582, + "eval_runtime": 44.7095, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.411628246307373, + "learning_rate": 2.989439480097482e-06, + "loss": 1.6644, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.8774638175964355, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.7192, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.857028007507324, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7079, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.4619641304016113, + "learning_rate": 2.983346872461414e-06, + "loss": 1.6846, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.6566193103790283, + "learning_rate": 2.981316003249391e-06, + "loss": 1.6768, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.681217670440674, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.7002, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.39505672454834, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.65, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.6062734127044678, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.6659, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.6402504444122314, + "learning_rate": 2.9731925264013e-06, + "loss": 1.6491, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.497014284133911, + "learning_rate": 2.971161657189277e-06, + "loss": 1.6383, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.874670147895813, + "eval_runtime": 42.9696, + "eval_samples_per_second": 23.272, + "eval_steps_per_second": 0.977, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.9070584774017334, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.6388, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.580744981765747, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7078, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.4637644290924072, + "learning_rate": 2.965069049553209e-06, + "loss": 1.6266, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.4489941596984863, + "learning_rate": 2.963038180341186e-06, + "loss": 1.705, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.664654016494751, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.6964, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.7017202377319336, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.6356, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.3691024780273438, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.6566, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.540384292602539, + "learning_rate": 2.954914703493095e-06, + "loss": 1.6334, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.5240535736083984, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.6316, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 3.105684518814087, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.6769, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.8741132020950317, + "eval_runtime": 42.6795, + "eval_samples_per_second": 23.43, + "eval_steps_per_second": 0.984, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.494252920150757, + "learning_rate": 2.948822095857027e-06, + "loss": 1.6359, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 4.069019317626953, + "learning_rate": 2.946791226645004e-06, + "loss": 1.621, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.641447067260742, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.6502, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.6700453758239746, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.6593, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.686068296432495, + "learning_rate": 2.940698619008936e-06, + "loss": 1.67, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.409055471420288, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.6809, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.4707298278808594, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.69, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.4735982418060303, + "learning_rate": 2.934606011372868e-06, + "loss": 1.666, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.633183479309082, + "learning_rate": 2.932575142160845e-06, + "loss": 1.6597, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.4808642864227295, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.6667, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.8731157779693604, + "eval_runtime": 41.3423, + "eval_samples_per_second": 24.188, + "eval_steps_per_second": 1.016, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.4044833183288574, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.7028, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.942699909210205, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.6678, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.505645990371704, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.5972, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.700967788696289, + "learning_rate": 2.922420796100731e-06, + "loss": 1.7064, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.4805755615234375, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.6201, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.753181219100952, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.6704, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.488865613937378, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7057, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.5464885234832764, + "learning_rate": 2.91429731925264e-06, + "loss": 1.6678, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.4997060298919678, + "learning_rate": 2.912266450040618e-06, + "loss": 1.662, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.5162436962127686, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.6501, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.869848370552063, + "eval_runtime": 43.3583, + "eval_samples_per_second": 23.064, + "eval_steps_per_second": 0.969, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.616129159927368, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.6206, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.761080503463745, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7131, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.5994064807891846, + "learning_rate": 2.904142973192527e-06, + "loss": 1.6693, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 3.4927968978881836, + "learning_rate": 2.902112103980504e-06, + "loss": 1.6694, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.6050939559936523, + "learning_rate": 2.900081234768481e-06, + "loss": 1.6419, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.796086311340332, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.6354, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.5658411979675293, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.652, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.3802385330200195, + "learning_rate": 2.893988627132413e-06, + "loss": 1.5902, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.479189872741699, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.6617, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.4367568492889404, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.6691, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.868497371673584, + "eval_runtime": 45.172, + "eval_samples_per_second": 22.138, + "eval_steps_per_second": 0.93, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.51888370513916, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.6872, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.5192768573760986, + "learning_rate": 2.885865150284322e-06, + "loss": 1.6584, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.5898501873016357, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.6179, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.698101043701172, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.663, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.889942169189453, + "learning_rate": 2.879772542648254e-06, + "loss": 1.5954, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.7909929752349854, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.7264, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 3.1931703090667725, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7091, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.7934632301330566, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.7204, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.470545530319214, + "learning_rate": 2.871649065800163e-06, + "loss": 1.644, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 3.011662006378174, + "learning_rate": 2.86961819658814e-06, + "loss": 1.6743, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.8678737878799438, + "eval_runtime": 44.7875, + "eval_samples_per_second": 22.328, + "eval_steps_per_second": 0.938, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.60823655128479, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.6467, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.5347230434417725, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.7044, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.481715440750122, + "learning_rate": 2.863525588952072e-06, + "loss": 1.5969, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.4966487884521484, + "learning_rate": 2.861494719740049e-06, + "loss": 1.6231, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.8809802532196045, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.6562, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.570204734802246, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.6511, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.6073813438415527, + "learning_rate": 2.855402112103981e-06, + "loss": 1.6456, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.8125052452087402, + "learning_rate": 2.853371242891958e-06, + "loss": 1.6722, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 5.341272354125977, + "learning_rate": 2.851340373679935e-06, + "loss": 1.7025, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 3.138002395629883, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.6676, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.8733265399932861, + "eval_runtime": 45.4915, + "eval_samples_per_second": 21.982, + "eval_steps_per_second": 0.923, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 7.264789581298828, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.6335, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 3.8533623218536377, + "learning_rate": 2.845247766043867e-06, + "loss": 1.6029, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 4.1789164543151855, + "learning_rate": 2.843216896831844e-06, + "loss": 1.6368, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 3.117408037185669, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.6924, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.721306085586548, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.6914, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.7099668979644775, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.6508, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 3.7823967933654785, + "learning_rate": 2.835093419983753e-06, + "loss": 1.636, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.696370840072632, + "learning_rate": 2.833062550771731e-06, + "loss": 1.6175, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.7303690910339355, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.6654, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.867727041244507, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.6717, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.8720332384109497, + "eval_runtime": 44.164, + "eval_samples_per_second": 22.643, + "eval_steps_per_second": 0.951, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 3.1209630966186523, + "learning_rate": 2.826969943135662e-06, + "loss": 1.6635, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.5057783126831055, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.6309, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.6650168895721436, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.6365, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.371305465698242, + "learning_rate": 2.820877335499594e-06, + "loss": 1.603, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.387404203414917, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.6189, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.6265947818756104, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.6511, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.6108627319335938, + "learning_rate": 2.814784727863526e-06, + "loss": 1.6727, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.7021450996398926, + "learning_rate": 2.812753858651503e-06, + "loss": 1.6334, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.470118999481201, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.6155, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5722644329071045, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.6788, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.8727209568023682, + "eval_runtime": 44.6253, + "eval_samples_per_second": 22.409, + "eval_steps_per_second": 0.941, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.7143876552581787, + "learning_rate": 2.806661251015435e-06, + "loss": 1.6956, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.8522002696990967, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.7558, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.5352110862731934, + "learning_rate": 2.802599512591389e-06, + "loss": 1.6592, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.7065908908843994, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.6121, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.616450786590576, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.6643, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.528763771057129, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7104, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 3.725569009780884, + "learning_rate": 2.794476035743298e-06, + "loss": 1.6713, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.487497091293335, + "learning_rate": 2.792445166531276e-06, + "loss": 1.692, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.534766912460327, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.6257, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.8553240299224854, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.6147, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.8675988912582397, + "eval_runtime": 46.4445, + "eval_samples_per_second": 21.531, + "eval_steps_per_second": 0.904, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.5054166316986084, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.6082, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.4133870601654053, + "learning_rate": 2.784321689683185e-06, + "loss": 1.6148, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.5863375663757324, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7032, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.58491849899292, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.679, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.5225989818573, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.6688, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.366861343383789, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.7076, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.4159810543060303, + "learning_rate": 2.774167343623071e-06, + "loss": 1.6301, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.5394275188446045, + "learning_rate": 2.772136474411048e-06, + "loss": 1.5758, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.409821033477783, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.664, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.4548892974853516, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.643, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.8672850131988525, + "eval_runtime": 46.3149, + "eval_samples_per_second": 21.591, + "eval_steps_per_second": 0.907, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.7483904361724854, + "learning_rate": 2.76604386677498e-06, + "loss": 1.6898, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.538804769515991, + "learning_rate": 2.764012997562957e-06, + "loss": 1.6019, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.7153313159942627, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.691, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.6892242431640625, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.6386, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.550957679748535, + "learning_rate": 2.757920389926889e-06, + "loss": 1.6237, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.3658559322357178, + "learning_rate": 2.755889520714866e-06, + "loss": 1.6015, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.6524102687835693, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.6644, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.7067437171936035, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.687, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.4044861793518066, + "learning_rate": 2.749796913078798e-06, + "loss": 1.6535, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.4933829307556152, + "learning_rate": 2.747766043866775e-06, + "loss": 1.7017, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.8652688264846802, + "eval_runtime": 46.8502, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.7062129974365234, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7398, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.5704996585845947, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.6561, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.5875790119171143, + "learning_rate": 2.741673436230707e-06, + "loss": 1.643, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.7498223781585693, + "learning_rate": 2.739642567018684e-06, + "loss": 1.6652, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.7275257110595703, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.6559, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.6349947452545166, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.6489, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.7058634757995605, + "learning_rate": 2.733549959382616e-06, + "loss": 1.7176, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.481344223022461, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.685, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.468277931213379, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.6305, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.6183228492736816, + "learning_rate": 2.727457351746548e-06, + "loss": 1.6151, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.864329218864441, + "eval_runtime": 43.0443, + "eval_samples_per_second": 23.232, + "eval_steps_per_second": 0.976, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.50679349899292, + "learning_rate": 2.725426482534525e-06, + "loss": 1.6468, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.6565401554107666, + "learning_rate": 2.723395613322502e-06, + "loss": 1.6496, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.5332367420196533, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.6424, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.6645619869232178, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.8012, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.7359142303466797, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.6025, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.660728693008423, + "learning_rate": 2.715272136474411e-06, + "loss": 1.6601, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.5184221267700195, + "learning_rate": 2.713241267262389e-06, + "loss": 1.6383, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.643712043762207, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.6834, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.504753351211548, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.6292, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.3439948558807373, + "learning_rate": 2.70714865962632e-06, + "loss": 1.6558, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.861581563949585, + "eval_runtime": 43.9224, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4820125102996826, + "learning_rate": 2.705117790414298e-06, + "loss": 1.69, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.6156861782073975, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.6304, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.490123987197876, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.6573, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.4439024925231934, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.6368, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.891603708267212, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7111, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.481175661087036, + "learning_rate": 2.694963444354184e-06, + "loss": 1.6408, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.5378847122192383, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.6782, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.592477321624756, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.6085, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.9336471557617188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.6719, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.3115012645721436, + "learning_rate": 2.686839967506093e-06, + "loss": 1.5797, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.8649041652679443, + "eval_runtime": 43.711, + "eval_samples_per_second": 22.878, + "eval_steps_per_second": 0.961, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.627028465270996, + "learning_rate": 2.68480909829407e-06, + "loss": 1.5784, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.917670488357544, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7007, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.418285608291626, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.6613, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.607773780822754, + "learning_rate": 2.678716490658002e-06, + "loss": 1.6561, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.569714307785034, + "learning_rate": 2.676685621445979e-06, + "loss": 1.6541, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.359494209289551, + "learning_rate": 2.674654752233956e-06, + "loss": 1.6467, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.5155694484710693, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.7003, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.439147710800171, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.5679, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.5044116973876953, + "learning_rate": 2.668562144597888e-06, + "loss": 1.6626, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.4475765228271484, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.656, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.8636877536773682, + "eval_runtime": 43.5138, + "eval_samples_per_second": 22.981, + "eval_steps_per_second": 0.965, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4616832733154297, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.6101, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.4125123023986816, + "learning_rate": 2.66246953696182e-06, + "loss": 1.6021, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.571326494216919, + "learning_rate": 2.660438667749797e-06, + "loss": 1.6482, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.424332618713379, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.653, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.3034703731536865, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.6096, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.4758505821228027, + "learning_rate": 2.654346060113729e-06, + "loss": 1.6274, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 3.800152540206909, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.6217, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.479827880859375, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7202, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.604523181915283, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.6778, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 3.595973491668701, + "learning_rate": 2.646222583265638e-06, + "loss": 1.6678, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.8647090196609497, + "eval_runtime": 43.0415, + "eval_samples_per_second": 23.233, + "eval_steps_per_second": 0.976, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 3.297319173812866, + "learning_rate": 2.644191714053615e-06, + "loss": 1.637, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.423567056655884, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.6018, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 4.4547953605651855, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.6594, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 3.0775582790374756, + "learning_rate": 2.638099106417547e-06, + "loss": 1.6259, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.747938871383667, + "learning_rate": 2.636068237205524e-06, + "loss": 1.7208, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.5350255966186523, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.5705, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 3.0630743503570557, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.6507, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 3.5475759506225586, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7045, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.534303903579712, + "learning_rate": 2.627944760357433e-06, + "loss": 1.7173, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.6964452266693115, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.6823, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.865207552909851, + "eval_runtime": 42.9435, + "eval_samples_per_second": 23.286, + "eval_steps_per_second": 0.978, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.924957752227783, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.6036, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.7058141231536865, + "learning_rate": 2.621852152721365e-06, + "loss": 1.6701, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.735339641571045, + "learning_rate": 2.619821283509342e-06, + "loss": 1.607, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.573958396911621, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.667, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5776426792144775, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.6938, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 4.454444408416748, + "learning_rate": 2.613728675873274e-06, + "loss": 1.7139, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.4883337020874023, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.6524, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.672116994857788, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.6197, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 3.511932611465454, + "learning_rate": 2.607636068237206e-06, + "loss": 1.6739, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.5960047245025635, + "learning_rate": 2.605605199025183e-06, + "loss": 1.6516, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.8604435920715332, + "eval_runtime": 42.9247, + "eval_samples_per_second": 23.297, + "eval_steps_per_second": 0.978, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.565376043319702, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.6884, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.676114797592163, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.6693, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.582751750946045, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.6499, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.748908281326294, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.6837, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.5578067302703857, + "learning_rate": 2.595450852965069e-06, + "loss": 1.6959, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.5702805519104004, + "learning_rate": 2.593419983753047e-06, + "loss": 1.6429, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.4748787879943848, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.6878, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.6157331466674805, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.6246, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.8377959728240967, + "learning_rate": 2.587327376116978e-06, + "loss": 1.6451, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.485398292541504, + "learning_rate": 2.585296506904956e-06, + "loss": 1.6288, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.8593864440917969, + "eval_runtime": 44.6341, + "eval_samples_per_second": 22.404, + "eval_steps_per_second": 0.941, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.46850323677063, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.5766, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.673283576965332, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7383, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.935293197631836, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.5742, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.531733751296997, + "learning_rate": 2.577173030056865e-06, + "loss": 1.6714, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.528425455093384, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6802, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.5644803047180176, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.6201, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.382451295852661, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.5824, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.63320255279541, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.6685, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.488844394683838, + "learning_rate": 2.567018683996751e-06, + "loss": 1.6078, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.7845425605773926, + "learning_rate": 2.564987814784728e-06, + "loss": 1.638, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.8572624921798706, + "eval_runtime": 43.2181, + "eval_samples_per_second": 23.138, + "eval_steps_per_second": 0.972, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.709656000137329, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.6425, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.651277542114258, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7401, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.5424256324768066, + "learning_rate": 2.55889520714866e-06, + "loss": 1.6272, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.6508235931396484, + "learning_rate": 2.556864337936637e-06, + "loss": 1.6384, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.882636308670044, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.6965, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.518367290496826, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.6551, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.7977044582366943, + "learning_rate": 2.550771730300569e-06, + "loss": 1.679, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.353041172027588, + "learning_rate": 2.548740861088546e-06, + "loss": 1.5745, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.5718655586242676, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.5655, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.6955313682556152, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.6764, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.8569272756576538, + "eval_runtime": 45.3844, + "eval_samples_per_second": 22.034, + "eval_steps_per_second": 0.925, + "step": 1500 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 4.67596411404288e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1500/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1700/generation_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1700/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1700/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/tokenizer.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1800/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/generation_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/latest b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/latest new file mode 100644 index 0000000000000000000000000000000000000000..987364c74ed1983839a73116dad31651fadca561 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/latest @@ -0,0 +1 @@ +global_step1900 \ No newline at end of file diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/rng_state_0.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..0b4ae0642e5f92f74c658999b194b5ef1fe11845 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8eeb05b2464265890d03970b06f167c82600ae380c5af274feadb43b529af8da +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/rng_state_1.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..7f17a148b6fb83bbdca37275527834750b348cea --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0ba564af3daac2ee124ab61d7e40b0c6a31626027ac994a15a45b4c2ff14ad78 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/special_tokens_map.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/tokenizer.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..f0f804e30d64b7911f670ef62c04b68282ecbe1d --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/trainer_state.json @@ -0,0 +1,14853 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.0821917808219177, + "eval_steps": 10, + "global_step": 1900, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.5818793773651123, + "learning_rate": 4.569455727051178e-06, + "loss": 1.85, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 3.0734691619873047, + "learning_rate": 4.567424857839156e-06, + "loss": 1.8682, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.496253252029419, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9007, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.360891342163086, + "learning_rate": 4.5633631194151105e-06, + "loss": 1.8262, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.5731117725372314, + "learning_rate": 4.561332250203087e-06, + "loss": 1.8369, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.700558662414551, + "learning_rate": 4.559301380991065e-06, + "loss": 1.7825, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.561306953430176, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.8174, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.618889570236206, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.8812, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.4671308994293213, + "learning_rate": 4.553208773354996e-06, + "loss": 1.7959, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.4571878910064697, + "learning_rate": 4.551177904142974e-06, + "loss": 1.8943, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.010777711868286, + "eval_runtime": 43.7424, + "eval_samples_per_second": 22.861, + "eval_steps_per_second": 0.96, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.4786667823791504, + "learning_rate": 4.5491470349309505e-06, + "loss": 1.8147, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.2955000400543213, + "learning_rate": 4.547116165718929e-06, + "loss": 1.8547, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.3873913288116455, + "learning_rate": 4.545085296506905e-06, + "loss": 1.8156, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.2449734210968018, + "learning_rate": 4.543054427294882e-06, + "loss": 1.8053, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.3969857692718506, + "learning_rate": 4.54102355808286e-06, + "loss": 1.8085, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.3446357250213623, + "learning_rate": 4.538992688870837e-06, + "loss": 1.8355, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.408762216567993, + "learning_rate": 4.536961819658814e-06, + "loss": 1.8319, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.362962245941162, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.8422, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.402984857559204, + "learning_rate": 4.532900081234769e-06, + "loss": 1.8125, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.3964452743530273, + "learning_rate": 4.530869212022746e-06, + "loss": 1.8417, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.006908416748047, + "eval_runtime": 44.2701, + "eval_samples_per_second": 22.589, + "eval_steps_per_second": 0.949, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.595233201980591, + "learning_rate": 4.528838342810723e-06, + "loss": 1.8275, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.326117753982544, + "learning_rate": 4.526807473598701e-06, + "loss": 1.8405, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.4904465675354004, + "learning_rate": 4.524776604386678e-06, + "loss": 1.8172, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.4591610431671143, + "learning_rate": 4.522745735174655e-06, + "loss": 1.876, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.319800853729248, + "learning_rate": 4.520714865962632e-06, + "loss": 1.8294, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.28790020942688, + "learning_rate": 4.51868399675061e-06, + "loss": 1.8537, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.284822702407837, + "learning_rate": 4.516653127538587e-06, + "loss": 1.8087, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.273820638656616, + "learning_rate": 4.514622258326564e-06, + "loss": 1.7992, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.2025654315948486, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.8248, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 2.3218843936920166, + "learning_rate": 4.510560519902519e-06, + "loss": 1.8647, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.0048956871032715, + "eval_runtime": 44.5972, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.515766143798828, + "learning_rate": 4.508529650690496e-06, + "loss": 1.7673, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.4376487731933594, + "learning_rate": 4.506498781478473e-06, + "loss": 1.7755, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.2801272869110107, + "learning_rate": 4.504467912266451e-06, + "loss": 1.8273, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.4707891941070557, + "learning_rate": 4.502437043054428e-06, + "loss": 1.8971, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.301301956176758, + "learning_rate": 4.500406173842405e-06, + "loss": 1.8345, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.2317373752593994, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.8248, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.447373867034912, + "learning_rate": 4.49634443541836e-06, + "loss": 1.8148, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.337245225906372, + "learning_rate": 4.494313566206336e-06, + "loss": 1.8811, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3717031478881836, + "learning_rate": 4.492282696994314e-06, + "loss": 1.8311, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.287677526473999, + "learning_rate": 4.490251827782291e-06, + "loss": 1.7728, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.004889965057373, + "eval_runtime": 43.8132, + "eval_samples_per_second": 22.824, + "eval_steps_per_second": 0.959, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.441445827484131, + "learning_rate": 4.488220958570269e-06, + "loss": 1.7927, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.3191535472869873, + "learning_rate": 4.486190089358245e-06, + "loss": 1.759, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.3118820190429688, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8509, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.4711859226226807, + "learning_rate": 4.4821283509342e-06, + "loss": 1.8402, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.3142173290252686, + "learning_rate": 4.480097481722178e-06, + "loss": 1.8238, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.2665774822235107, + "learning_rate": 4.478066612510154e-06, + "loss": 1.7826, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.1692490577697754, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.8138, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.412783145904541, + "learning_rate": 4.474004874086109e-06, + "loss": 1.9421, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3692686557769775, + "learning_rate": 4.471974004874087e-06, + "loss": 1.7722, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.3224151134490967, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.8753, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.0016305446624756, + "eval_runtime": 45.564, + "eval_samples_per_second": 21.947, + "eval_steps_per_second": 0.922, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.51043438911438, + "learning_rate": 4.467912266450041e-06, + "loss": 1.8325, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.4568657875061035, + "learning_rate": 4.465881397238018e-06, + "loss": 1.8081, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.3279001712799072, + "learning_rate": 4.463850528025995e-06, + "loss": 1.8439, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.6041171550750732, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.8623, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.3780531883239746, + "learning_rate": 4.45978878960195e-06, + "loss": 1.8269, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.511500597000122, + "learning_rate": 4.457757920389927e-06, + "loss": 1.7749, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.707789659500122, + "learning_rate": 4.455727051177904e-06, + "loss": 1.8157, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.5495495796203613, + "learning_rate": 4.453696181965882e-06, + "loss": 1.8534, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.255876302719116, + "learning_rate": 4.451665312753859e-06, + "loss": 1.8377, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.415097951889038, + "learning_rate": 4.449634443541836e-06, + "loss": 1.8764, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 1.9943474531173706, + "eval_runtime": 42.2985, + "eval_samples_per_second": 23.642, + "eval_steps_per_second": 0.993, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.2604713439941406, + "learning_rate": 4.4476035743298135e-06, + "loss": 1.7898, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.5643842220306396, + "learning_rate": 4.445572705117791e-06, + "loss": 1.7953, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.412548780441284, + "learning_rate": 4.443541835905768e-06, + "loss": 1.8611, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.5423877239227295, + "learning_rate": 4.441510966693745e-06, + "loss": 1.8049, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.4986631870269775, + "learning_rate": 4.439480097481723e-06, + "loss": 1.8759, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.8349273204803467, + "learning_rate": 4.4374492282697e-06, + "loss": 1.7416, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.452017068862915, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8268, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.6425273418426514, + "learning_rate": 4.433387489845654e-06, + "loss": 1.8334, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.328740358352661, + "learning_rate": 4.431356620633632e-06, + "loss": 1.7931, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.262209415435791, + "learning_rate": 4.429325751421609e-06, + "loss": 1.8151, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 1.9917123317718506, + "eval_runtime": 42.4309, + "eval_samples_per_second": 23.568, + "eval_steps_per_second": 0.99, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.2984049320220947, + "learning_rate": 4.427294882209586e-06, + "loss": 1.8187, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.3910019397735596, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.7512, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.4494471549987793, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8202, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.3857171535491943, + "learning_rate": 4.421202274573518e-06, + "loss": 1.8279, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.4526500701904297, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.7136, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.2216501235961914, + "learning_rate": 4.417140536149473e-06, + "loss": 1.7574, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.7639758586883545, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8403, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.3177521228790283, + "learning_rate": 4.413078797725427e-06, + "loss": 1.7898, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.527902364730835, + "learning_rate": 4.411047928513404e-06, + "loss": 1.8184, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.34061336517334, + "learning_rate": 4.409017059301382e-06, + "loss": 1.811, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 1.9882216453552246, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.6462934017181396, + "learning_rate": 4.406986190089358e-06, + "loss": 1.8501, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.2725913524627686, + "learning_rate": 4.404955320877336e-06, + "loss": 1.8262, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.390084981918335, + "learning_rate": 4.402924451665313e-06, + "loss": 1.7842, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.314305305480957, + "learning_rate": 4.400893582453291e-06, + "loss": 1.7692, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.2828257083892822, + "learning_rate": 4.398862713241267e-06, + "loss": 1.8031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.3493599891662598, + "learning_rate": 4.396831844029245e-06, + "loss": 1.8183, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.423365831375122, + "learning_rate": 4.394800974817222e-06, + "loss": 1.8012, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.438015937805176, + "learning_rate": 4.392770105605199e-06, + "loss": 1.8217, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.357172727584839, + "learning_rate": 4.390739236393176e-06, + "loss": 1.8382, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.3383474349975586, + "learning_rate": 4.388708367181154e-06, + "loss": 1.841, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 1.9878787994384766, + "eval_runtime": 44.0694, + "eval_samples_per_second": 22.691, + "eval_steps_per_second": 0.953, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.459385633468628, + "learning_rate": 4.386677497969131e-06, + "loss": 1.8377, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.356788158416748, + "learning_rate": 4.384646628757108e-06, + "loss": 1.8718, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.3217222690582275, + "learning_rate": 4.3826157595450855e-06, + "loss": 1.8073, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.3134331703186035, + "learning_rate": 4.380584890333063e-06, + "loss": 1.7675, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.3273653984069824, + "learning_rate": 4.37855402112104e-06, + "loss": 1.7169, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.611034393310547, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8702, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.2077300548553467, + "learning_rate": 4.3744922826969946e-06, + "loss": 1.7951, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.3393311500549316, + "learning_rate": 4.372461413484972e-06, + "loss": 1.8136, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.3306660652160645, + "learning_rate": 4.370430544272949e-06, + "loss": 1.8372, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.4440250396728516, + "learning_rate": 4.368399675060926e-06, + "loss": 1.844, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 1.98439621925354, + "eval_runtime": 44.4595, + "eval_samples_per_second": 22.492, + "eval_steps_per_second": 0.945, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.205542802810669, + "learning_rate": 4.366368805848904e-06, + "loss": 1.8182, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.6030094623565674, + "learning_rate": 4.364337936636881e-06, + "loss": 1.7524, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.4000444412231445, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8269, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.319526433944702, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.7722, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.1624808311462402, + "learning_rate": 4.358245329000813e-06, + "loss": 1.8081, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.26615309715271, + "learning_rate": 4.35621445978879e-06, + "loss": 1.8049, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.4895036220550537, + "learning_rate": 4.354183590576767e-06, + "loss": 1.7932, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.437161445617676, + "learning_rate": 4.352152721364745e-06, + "loss": 1.772, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.344069480895996, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8213, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.360734224319458, + "learning_rate": 4.348090982940699e-06, + "loss": 1.8268, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 1.9846479892730713, + "eval_runtime": 44.947, + "eval_samples_per_second": 22.248, + "eval_steps_per_second": 0.934, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.59487247467041, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.8303, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.6512107849121094, + "learning_rate": 4.344029244516654e-06, + "loss": 1.7979, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.2900476455688477, + "learning_rate": 4.341998375304631e-06, + "loss": 1.8707, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.409494638442993, + "learning_rate": 4.339967506092607e-06, + "loss": 1.8075, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.527205467224121, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.7884, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.33461332321167, + "learning_rate": 4.335905767668562e-06, + "loss": 1.7839, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.3555047512054443, + "learning_rate": 4.33387489845654e-06, + "loss": 1.8245, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.397148609161377, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.8115, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.232139825820923, + "learning_rate": 4.329813160032495e-06, + "loss": 1.7747, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.1684389114379883, + "learning_rate": 4.327782290820471e-06, + "loss": 1.748, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 1.980297565460205, + "eval_runtime": 42.9183, + "eval_samples_per_second": 23.3, + "eval_steps_per_second": 0.979, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.2454638481140137, + "learning_rate": 4.325751421608449e-06, + "loss": 1.83, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.3264827728271484, + "learning_rate": 4.323720552396426e-06, + "loss": 1.7856, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.298804998397827, + "learning_rate": 4.321689683184404e-06, + "loss": 1.78, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2202727794647217, + "learning_rate": 4.31965881397238e-06, + "loss": 1.7875, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.494479179382324, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.8363, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3520524501800537, + "learning_rate": 4.315597075548335e-06, + "loss": 1.8266, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.373878002166748, + "learning_rate": 4.313566206336312e-06, + "loss": 1.8348, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.4232168197631836, + "learning_rate": 4.311535337124289e-06, + "loss": 1.77, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.2572121620178223, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.7914, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.5817952156066895, + "learning_rate": 4.307473598700244e-06, + "loss": 1.8099, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 1.9784427881240845, + "eval_runtime": 43.2981, + "eval_samples_per_second": 23.096, + "eval_steps_per_second": 0.97, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.646040916442871, + "learning_rate": 4.305442729488221e-06, + "loss": 1.7452, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.370220184326172, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8399, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.4599130153656006, + "learning_rate": 4.301380991064176e-06, + "loss": 1.8273, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 2.4135093688964844, + "learning_rate": 4.299350121852153e-06, + "loss": 1.852, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.371021270751953, + "learning_rate": 4.29731925264013e-06, + "loss": 1.8339, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.6858229637145996, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.792, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.4101266860961914, + "learning_rate": 4.293257514216085e-06, + "loss": 1.7556, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.769761800765991, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8223, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.5777804851531982, + "learning_rate": 4.289195775792039e-06, + "loss": 1.8406, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.376864433288574, + "learning_rate": 4.287164906580017e-06, + "loss": 1.7846, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 1.9758412837982178, + "eval_runtime": 43.9353, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.7461960315704346, + "learning_rate": 4.285134037367994e-06, + "loss": 1.776, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.549576759338379, + "learning_rate": 4.283103168155971e-06, + "loss": 1.7644, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.568074941635132, + "learning_rate": 4.281072298943948e-06, + "loss": 1.7571, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.8195722103118896, + "learning_rate": 4.279041429731926e-06, + "loss": 1.7959, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.471623182296753, + "learning_rate": 4.277010560519903e-06, + "loss": 1.7584, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.664689302444458, + "learning_rate": 4.27497969130788e-06, + "loss": 1.7909, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.684504985809326, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.8205, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.5247528553009033, + "learning_rate": 4.270917952883835e-06, + "loss": 1.8365, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.757972478866577, + "learning_rate": 4.268887083671812e-06, + "loss": 1.7665, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.6032614707946777, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8702, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 1.9716291427612305, + "eval_runtime": 43.5482, + "eval_samples_per_second": 22.963, + "eval_steps_per_second": 0.964, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.5324370861053467, + "learning_rate": 4.264825345247766e-06, + "loss": 1.738, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.730712890625, + "learning_rate": 4.262794476035744e-06, + "loss": 1.7918, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.5554966926574707, + "learning_rate": 4.26076360682372e-06, + "loss": 1.7218, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.4774436950683594, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.7823, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.77431058883667, + "learning_rate": 4.256701868399675e-06, + "loss": 1.794, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.258828639984131, + "learning_rate": 4.254670999187653e-06, + "loss": 1.805, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.384399175643921, + "learning_rate": 4.252640129975629e-06, + "loss": 1.818, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3278932571411133, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.7719, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.207348108291626, + "learning_rate": 4.248578391551584e-06, + "loss": 1.7362, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.4525978565216064, + "learning_rate": 4.246547522339562e-06, + "loss": 1.778, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 1.9694374799728394, + "eval_runtime": 42.5977, + "eval_samples_per_second": 23.475, + "eval_steps_per_second": 0.986, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.377835273742676, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.7926, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.4623866081237793, + "learning_rate": 4.242485783915517e-06, + "loss": 1.7366, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.542219400405884, + "learning_rate": 4.240454914703493e-06, + "loss": 1.7409, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.1955819129943848, + "learning_rate": 4.23842404549147e-06, + "loss": 1.7596, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.5834269523620605, + "learning_rate": 4.236393176279448e-06, + "loss": 1.7343, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.219660520553589, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8089, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.283200740814209, + "learning_rate": 4.232331437855402e-06, + "loss": 1.8248, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.274127244949341, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.825, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.3550686836242676, + "learning_rate": 4.228269699431357e-06, + "loss": 1.7957, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.588027238845825, + "learning_rate": 4.226238830219334e-06, + "loss": 1.7865, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 1.9662212133407593, + "eval_runtime": 43.2661, + "eval_samples_per_second": 23.113, + "eval_steps_per_second": 0.971, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.2445170879364014, + "learning_rate": 4.224207961007311e-06, + "loss": 1.791, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.399174690246582, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.7944, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.586184501647949, + "learning_rate": 4.220146222583266e-06, + "loss": 1.7856, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.4174695014953613, + "learning_rate": 4.218115353371243e-06, + "loss": 1.7842, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2684640884399414, + "learning_rate": 4.21608448415922e-06, + "loss": 1.7482, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 3.1782288551330566, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8286, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.4046335220336914, + "learning_rate": 4.212022745735175e-06, + "loss": 1.7228, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 3.3722083568573, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8044, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.293466329574585, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.7415, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.28663969039917, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8282, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 1.9627619981765747, + "eval_runtime": 43.6092, + "eval_samples_per_second": 22.931, + "eval_steps_per_second": 0.963, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.255870819091797, + "learning_rate": 4.203899268887084e-06, + "loss": 1.7983, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.4964773654937744, + "learning_rate": 4.201868399675061e-06, + "loss": 1.8141, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.3372740745544434, + "learning_rate": 4.199837530463039e-06, + "loss": 1.7142, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.3506968021392822, + "learning_rate": 4.197806661251016e-06, + "loss": 1.7898, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.242410898208618, + "learning_rate": 4.195775792038993e-06, + "loss": 1.7677, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.3864502906799316, + "learning_rate": 4.19374492282697e-06, + "loss": 1.7097, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.482349157333374, + "learning_rate": 4.191714053614948e-06, + "loss": 1.8052, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.317652940750122, + "learning_rate": 4.189683184402924e-06, + "loss": 1.7692, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.2777392864227295, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8246, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.3506035804748535, + "learning_rate": 4.185621445978879e-06, + "loss": 1.7576, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 1.9610412120819092, + "eval_runtime": 43.3119, + "eval_samples_per_second": 23.088, + "eval_steps_per_second": 0.97, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.3011536598205566, + "learning_rate": 4.183590576766857e-06, + "loss": 1.7891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.2614853382110596, + "learning_rate": 4.181559707554833e-06, + "loss": 1.8258, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.219175100326538, + "learning_rate": 4.179528838342811e-06, + "loss": 1.7494, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.4239423274993896, + "learning_rate": 4.177497969130788e-06, + "loss": 1.7611, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.470712423324585, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8281, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.504060745239258, + "learning_rate": 4.173436230706742e-06, + "loss": 1.7641, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.326601982116699, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.7671, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.4280030727386475, + "learning_rate": 4.169374492282697e-06, + "loss": 1.7573, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.455716371536255, + "learning_rate": 4.167343623070675e-06, + "loss": 1.809, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.3029844760894775, + "learning_rate": 4.165312753858652e-06, + "loss": 1.77, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 1.9604709148406982, + "eval_runtime": 44.4627, + "eval_samples_per_second": 22.491, + "eval_steps_per_second": 0.945, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.3833982944488525, + "learning_rate": 4.163281884646629e-06, + "loss": 1.7652, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.2392966747283936, + "learning_rate": 4.161251015434607e-06, + "loss": 1.7904, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.4546353816986084, + "learning_rate": 4.159220146222583e-06, + "loss": 1.7324, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.13592529296875, + "learning_rate": 4.157189277010561e-06, + "loss": 1.759, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.381657361984253, + "learning_rate": 4.155158407798538e-06, + "loss": 1.7871, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.4142940044403076, + "learning_rate": 4.153127538586516e-06, + "loss": 1.7832, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.305697441101074, + "learning_rate": 4.151096669374492e-06, + "loss": 1.7619, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.193429708480835, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.7738, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.3007619380950928, + "learning_rate": 4.147034930950447e-06, + "loss": 1.7839, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.6337807178497314, + "learning_rate": 4.145004061738425e-06, + "loss": 1.7748, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 1.9597856998443604, + "eval_runtime": 43.8644, + "eval_samples_per_second": 22.798, + "eval_steps_per_second": 0.957, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.4294793605804443, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.7001, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.512155532836914, + "learning_rate": 4.140942323314379e-06, + "loss": 1.7334, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.290808916091919, + "learning_rate": 4.138911454102356e-06, + "loss": 1.7522, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.483161211013794, + "learning_rate": 4.136880584890333e-06, + "loss": 1.798, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2973921298980713, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.7847, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.253084659576416, + "learning_rate": 4.132818846466288e-06, + "loss": 1.7812, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.306342840194702, + "learning_rate": 4.130787977254265e-06, + "loss": 1.7886, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.5153164863586426, + "learning_rate": 4.128757108042242e-06, + "loss": 1.7954, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.4122278690338135, + "learning_rate": 4.12672623883022e-06, + "loss": 1.8525, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.4520716667175293, + "learning_rate": 4.124695369618197e-06, + "loss": 1.7648, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 1.9547661542892456, + "eval_runtime": 45.3774, + "eval_samples_per_second": 22.037, + "eval_steps_per_second": 0.926, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2959611415863037, + "learning_rate": 4.122664500406174e-06, + "loss": 1.7626, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.2280471324920654, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.7954, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.3870065212249756, + "learning_rate": 4.118602761982129e-06, + "loss": 1.7377, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.292612075805664, + "learning_rate": 4.116571892770106e-06, + "loss": 1.7, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.4607021808624268, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8338, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.532824993133545, + "learning_rate": 4.112510154346061e-06, + "loss": 1.7549, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2139623165130615, + "learning_rate": 4.110479285134038e-06, + "loss": 1.78, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.4749462604522705, + "learning_rate": 4.108448415922015e-06, + "loss": 1.8034, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.609265089035034, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.7941, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.367753028869629, + "learning_rate": 4.10438667749797e-06, + "loss": 1.6653, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 1.9539440870285034, + "eval_runtime": 48.8438, + "eval_samples_per_second": 20.473, + "eval_steps_per_second": 0.86, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.9723567962646484, + "learning_rate": 4.102355808285947e-06, + "loss": 1.8317, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.388806104660034, + "learning_rate": 4.100324939073924e-06, + "loss": 1.7597, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4882986545562744, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.7708, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.7586395740509033, + "learning_rate": 4.096263200649879e-06, + "loss": 1.7804, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.237025260925293, + "learning_rate": 4.094232331437856e-06, + "loss": 1.7415, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.421330213546753, + "learning_rate": 4.092201462225833e-06, + "loss": 1.734, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.7281360626220703, + "learning_rate": 4.090170593013811e-06, + "loss": 1.7716, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.2200379371643066, + "learning_rate": 4.088139723801787e-06, + "loss": 1.7266, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.465440034866333, + "learning_rate": 4.086108854589765e-06, + "loss": 1.741, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.39835262298584, + "learning_rate": 4.084077985377742e-06, + "loss": 1.757, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 1.9514222145080566, + "eval_runtime": 52.8068, + "eval_samples_per_second": 18.937, + "eval_steps_per_second": 0.795, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.526684522628784, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8195, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.341954469680786, + "learning_rate": 4.080016246953696e-06, + "loss": 1.7196, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.3755977153778076, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8115, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.3952431678771973, + "learning_rate": 4.075954508529651e-06, + "loss": 1.812, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.554089069366455, + "learning_rate": 4.073923639317629e-06, + "loss": 1.8184, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.2319302558898926, + "learning_rate": 4.071892770105605e-06, + "loss": 1.7448, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.284856081008911, + "learning_rate": 4.069861900893583e-06, + "loss": 1.7154, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.4081835746765137, + "learning_rate": 4.06783103168156e-06, + "loss": 1.7311, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.6115968227386475, + "learning_rate": 4.065800162469538e-06, + "loss": 1.7436, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.368004322052002, + "learning_rate": 4.063769293257514e-06, + "loss": 1.7811, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 1.9508955478668213, + "eval_runtime": 46.8711, + "eval_samples_per_second": 21.335, + "eval_steps_per_second": 0.896, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.4607419967651367, + "learning_rate": 4.061738424045492e-06, + "loss": 1.7563, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.2978763580322266, + "learning_rate": 4.059707554833469e-06, + "loss": 1.7755, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.3316237926483154, + "learning_rate": 4.057676685621446e-06, + "loss": 1.8226, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.3699169158935547, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.7755, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.195596218109131, + "learning_rate": 4.053614947197401e-06, + "loss": 1.6916, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.4524662494659424, + "learning_rate": 4.051584077985378e-06, + "loss": 1.774, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.573888063430786, + "learning_rate": 4.049553208773355e-06, + "loss": 1.7953, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.5157253742218018, + "learning_rate": 4.047522339561333e-06, + "loss": 1.7597, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.5588185787200928, + "learning_rate": 4.04549147034931e-06, + "loss": 1.7553, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.8950047492980957, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8647, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 1.945882797241211, + "eval_runtime": 45.481, + "eval_samples_per_second": 21.987, + "eval_steps_per_second": 0.923, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.217116117477417, + "learning_rate": 4.041429731925264e-06, + "loss": 1.7465, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.5601565837860107, + "learning_rate": 4.039398862713242e-06, + "loss": 1.7575, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.4540457725524902, + "learning_rate": 4.037367993501219e-06, + "loss": 1.8291, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.327181100845337, + "learning_rate": 4.035337124289196e-06, + "loss": 1.7409, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.485457181930542, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.7192, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.4613442420959473, + "learning_rate": 4.031275385865151e-06, + "loss": 1.7613, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.303217649459839, + "learning_rate": 4.029244516653128e-06, + "loss": 1.7799, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.3830621242523193, + "learning_rate": 4.027213647441105e-06, + "loss": 1.7045, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.342332601547241, + "learning_rate": 4.025182778229083e-06, + "loss": 1.7246, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.4014711380004883, + "learning_rate": 4.02315190901706e-06, + "loss": 1.7621, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 1.95199453830719, + "eval_runtime": 45.4754, + "eval_samples_per_second": 21.99, + "eval_steps_per_second": 0.924, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.6015400886535645, + "learning_rate": 4.021121039805037e-06, + "loss": 1.7536, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.866978406906128, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8405, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.389810085296631, + "learning_rate": 4.017059301380992e-06, + "loss": 1.7365, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.280442953109741, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8273, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.509330987930298, + "learning_rate": 4.012997562956945e-06, + "loss": 1.7127, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.452078104019165, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.7457, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.368337631225586, + "learning_rate": 4.0089358245329e-06, + "loss": 1.7412, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.4529192447662354, + "learning_rate": 4.006904955320878e-06, + "loss": 1.7434, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.647761821746826, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8013, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.387424945831299, + "learning_rate": 4.002843216896833e-06, + "loss": 1.7948, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 1.946029543876648, + "eval_runtime": 43.9666, + "eval_samples_per_second": 22.745, + "eval_steps_per_second": 0.955, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.9217846393585205, + "learning_rate": 4.000812347684809e-06, + "loss": 1.7384, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.400040864944458, + "learning_rate": 3.998781478472787e-06, + "loss": 1.7496, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.4050440788269043, + "learning_rate": 3.996750609260764e-06, + "loss": 1.7165, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 3.6001529693603516, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8679, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 2.4040098190307617, + "learning_rate": 3.992688870836718e-06, + "loss": 1.7671, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.5645527839660645, + "learning_rate": 3.990658001624696e-06, + "loss": 1.7374, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.799394130706787, + "learning_rate": 3.988627132412673e-06, + "loss": 1.796, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.5481607913970947, + "learning_rate": 3.98659626320065e-06, + "loss": 1.753, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.8851330280303955, + "learning_rate": 3.984565393988627e-06, + "loss": 1.7505, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.6668694019317627, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.7604, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 1.939759612083435, + "eval_runtime": 44.5019, + "eval_samples_per_second": 22.471, + "eval_steps_per_second": 0.944, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.320935010910034, + "learning_rate": 3.980503655564582e-06, + "loss": 1.7749, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.829897880554199, + "learning_rate": 3.978472786352559e-06, + "loss": 1.7763, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.8420908451080322, + "learning_rate": 3.976441917140536e-06, + "loss": 1.7598, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.8183, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.781, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 3.9271891117095947, + "learning_rate": 3.972380178716491e-06, + "loss": 1.7949, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.277897834777832, + "learning_rate": 3.970349309504468e-06, + "loss": 1.707, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 3.241063117980957, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.7175, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.514458417892456, + "learning_rate": 3.966287571080423e-06, + "loss": 1.7764, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.420212745666504, + "learning_rate": 3.9642567018684e-06, + "loss": 1.7618, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 1.9401386976242065, + "eval_runtime": 44.1672, + "eval_samples_per_second": 22.641, + "eval_steps_per_second": 0.951, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.3980538845062256, + "learning_rate": 3.962225832656377e-06, + "loss": 1.6922, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.892646312713623, + "learning_rate": 3.960194963444355e-06, + "loss": 1.7571, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.499260902404785, + "learning_rate": 3.958164094232332e-06, + "loss": 1.7501, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.409695625305176, + "learning_rate": 3.956133225020309e-06, + "loss": 1.7553, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.640069007873535, + "learning_rate": 3.954102355808286e-06, + "loss": 1.7526, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.9101760387420654, + "learning_rate": 3.952071486596264e-06, + "loss": 1.7483, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.2579474449157715, + "learning_rate": 3.950040617384241e-06, + "loss": 1.7379, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 4.039298057556152, + "learning_rate": 3.948009748172218e-06, + "loss": 1.7988, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.3232030868530273, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.7592, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.697164297103882, + "learning_rate": 3.943948009748173e-06, + "loss": 1.7622, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 1.9367746114730835, + "eval_runtime": 43.7013, + "eval_samples_per_second": 22.883, + "eval_steps_per_second": 0.961, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.480297803878784, + "learning_rate": 3.94191714053615e-06, + "loss": 1.8006, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.3624603748321533, + "learning_rate": 3.939886271324127e-06, + "loss": 1.7152, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.417187213897705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.7077, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.6575164794921875, + "learning_rate": 3.935824532900082e-06, + "loss": 1.7744, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.4896557331085205, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8052, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.475693941116333, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.7008, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.3210372924804688, + "learning_rate": 3.929731925264013e-06, + "loss": 1.7339, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.433785915374756, + "learning_rate": 3.927701056051991e-06, + "loss": 1.7544, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 3.181574583053589, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.7229, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.704399347305298, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.7086, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 1.9335075616836548, + "eval_runtime": 47.0248, + "eval_samples_per_second": 21.265, + "eval_steps_per_second": 0.893, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 3.5027623176574707, + "learning_rate": 3.921608448415922e-06, + "loss": 1.7622, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.450943946838379, + "learning_rate": 3.9195775792039e-06, + "loss": 1.7286, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.4603374004364014, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.7725, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.8217649459838867, + "learning_rate": 3.915515840779855e-06, + "loss": 1.7859, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.3308372497558594, + "learning_rate": 3.913484971567831e-06, + "loss": 1.6874, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.624776840209961, + "learning_rate": 3.911454102355808e-06, + "loss": 1.7632, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.4097838401794434, + "learning_rate": 3.909423233143786e-06, + "loss": 1.7576, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.487410545349121, + "learning_rate": 3.907392363931763e-06, + "loss": 1.831, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.647245168685913, + "learning_rate": 3.90536149471974e-06, + "loss": 1.8448, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.397664785385132, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.868, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 1.930379033088684, + "eval_runtime": 51.1662, + "eval_samples_per_second": 19.544, + "eval_steps_per_second": 0.821, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.277433395385742, + "learning_rate": 3.901299756295695e-06, + "loss": 1.7394, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 3.1312496662139893, + "learning_rate": 3.899268887083672e-06, + "loss": 1.7422, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.5011534690856934, + "learning_rate": 3.897238017871649e-06, + "loss": 1.7484, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.8716166019439697, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.7253, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.4346914291381836, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8197, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.348475933074951, + "learning_rate": 3.891145410235581e-06, + "loss": 1.7737, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.3996007442474365, + "learning_rate": 3.889114541023558e-06, + "loss": 1.7333, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.355262279510498, + "learning_rate": 3.887083671811536e-06, + "loss": 1.6902, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.335932731628418, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8206, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.2893190383911133, + "learning_rate": 3.88302193338749e-06, + "loss": 1.7533, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 1.9290764331817627, + "eval_runtime": 46.9373, + "eval_samples_per_second": 21.305, + "eval_steps_per_second": 0.895, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.739842176437378, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.7318, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.587916374206543, + "learning_rate": 3.878960194963445e-06, + "loss": 1.7979, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.406548500061035, + "learning_rate": 3.876929325751422e-06, + "loss": 1.6899, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.440321922302246, + "learning_rate": 3.874898456539399e-06, + "loss": 1.7546, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.320030450820923, + "learning_rate": 3.872867587327377e-06, + "loss": 1.7254, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.3795316219329834, + "learning_rate": 3.870836718115354e-06, + "loss": 1.728, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.299802541732788, + "learning_rate": 3.868805848903331e-06, + "loss": 1.7464, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.469719409942627, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.765, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.612109899520874, + "learning_rate": 3.864744110479286e-06, + "loss": 1.7827, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.623173952102661, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8019, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 1.9268505573272705, + "eval_runtime": 46.77, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.4146616458892822, + "learning_rate": 3.86068237205524e-06, + "loss": 1.7498, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.361994743347168, + "learning_rate": 3.858651502843217e-06, + "loss": 1.6926, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.493215560913086, + "learning_rate": 3.856620633631195e-06, + "loss": 1.736, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.4341812133789062, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8152, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.391331195831299, + "learning_rate": 3.852558895207149e-06, + "loss": 1.7717, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.419492244720459, + "learning_rate": 3.850528025995126e-06, + "loss": 1.7283, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.349090576171875, + "learning_rate": 3.848497156783104e-06, + "loss": 1.7317, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.298647880554199, + "learning_rate": 3.84646628757108e-06, + "loss": 1.7389, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.4612667560577393, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.7333, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.3139944076538086, + "learning_rate": 3.842404549147035e-06, + "loss": 1.6944, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 1.924438714981079, + "eval_runtime": 45.8276, + "eval_samples_per_second": 21.821, + "eval_steps_per_second": 0.916, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 2.17958664894104, + "learning_rate": 3.840373679935013e-06, + "loss": 1.7474, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.337859869003296, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.7345, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4777672290802, + "learning_rate": 3.836311941510967e-06, + "loss": 1.77, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.280158519744873, + "learning_rate": 3.834281072298944e-06, + "loss": 1.7069, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2387664318084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.7889, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.373771905899048, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8268, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.4950132369995117, + "learning_rate": 3.828188464662876e-06, + "loss": 1.7586, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.3296444416046143, + "learning_rate": 3.826157595450853e-06, + "loss": 1.6991, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.6653530597686768, + "learning_rate": 3.82412672623883e-06, + "loss": 1.7505, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.485128402709961, + "learning_rate": 3.822095857026808e-06, + "loss": 1.7196, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 1.9221889972686768, + "eval_runtime": 46.6581, + "eval_samples_per_second": 21.433, + "eval_steps_per_second": 0.9, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.3726181983947754, + "learning_rate": 3.820064987814785e-06, + "loss": 1.7296, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.233992099761963, + "learning_rate": 3.818034118602762e-06, + "loss": 1.7313, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.5388147830963135, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.7435, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.4787333011627197, + "learning_rate": 3.813972380178717e-06, + "loss": 1.7733, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.232564926147461, + "learning_rate": 3.811941510966694e-06, + "loss": 1.693, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.3882386684417725, + "learning_rate": 3.809910641754671e-06, + "loss": 1.6529, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.4785983562469482, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.7456, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.685098886489868, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.7162, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.353523015975952, + "learning_rate": 3.803818034118603e-06, + "loss": 1.7852, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.342163562774658, + "learning_rate": 3.80178716490658e-06, + "loss": 1.7767, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 1.9235820770263672, + "eval_runtime": 47.1327, + "eval_samples_per_second": 21.217, + "eval_steps_per_second": 0.891, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.550818920135498, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.7328, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.3204798698425293, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.7058, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.4752719402313232, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8051, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.5202794075012207, + "learning_rate": 3.793663688058489e-06, + "loss": 1.7966, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.26521635055542, + "learning_rate": 3.791632818846467e-06, + "loss": 1.7879, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.331515312194824, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.7814, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.37178635597229, + "learning_rate": 3.787571080422421e-06, + "loss": 1.7635, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.321417808532715, + "learning_rate": 3.785540211210398e-06, + "loss": 1.7513, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.485989809036255, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.7306, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.7497379779815674, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8409, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 1.9275448322296143, + "eval_runtime": 46.2088, + "eval_samples_per_second": 21.641, + "eval_steps_per_second": 0.909, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 4.9214019775390625, + "learning_rate": 3.77944760357433e-06, + "loss": 1.7834, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 4.505277156829834, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.7179, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.5772030353546143, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.7766, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.254185438156128, + "learning_rate": 3.773354995938262e-06, + "loss": 1.7697, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 3.632233142852783, + "learning_rate": 3.771324126726239e-06, + "loss": 1.7968, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.174461841583252, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.7346, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.3983266353607178, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.6534, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 3.393791913986206, + "learning_rate": 3.765231519090171e-06, + "loss": 1.7561, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.2853987216949463, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.6963, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.649789810180664, + "learning_rate": 3.761169780666125e-06, + "loss": 1.7395, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 1.9242008924484253, + "eval_runtime": 46.6017, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 3.417872190475464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.784, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.316666841506958, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.7154, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.486161231994629, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.7903, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 4.538443565368652, + "learning_rate": 3.753046303818034e-06, + "loss": 1.7711, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.9772889614105225, + "learning_rate": 3.751015434606012e-06, + "loss": 1.8046, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 7.2353715896606445, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.7189, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.736, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.8026, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 12.551762580871582, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.7788, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 3.0898046493530273, + "learning_rate": 3.742891957757921e-06, + "loss": 1.7878, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 1.9646916389465332, + "eval_runtime": 44.1583, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 8.120022773742676, + "learning_rate": 3.740861088545898e-06, + "loss": 1.768, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 6.485774040222168, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.746, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.85135555267334, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.6968, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 5.8551435470581055, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.6796, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 4.45880126953125, + "learning_rate": 3.732737611697807e-06, + "loss": 1.7219, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 3.0435831546783447, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.6959, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.8194847106933594, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.8152, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 3.1266117095947266, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.6937, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 3.642854690551758, + "learning_rate": 3.724614134849716e-06, + "loss": 1.7992, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 3.344496965408325, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.7583, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 1.9293720722198486, + "eval_runtime": 50.8602, + "eval_samples_per_second": 19.662, + "eval_steps_per_second": 0.826, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.7666447162628174, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7441, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.8819239139556885, + "learning_rate": 3.718521527213648e-06, + "loss": 1.6314, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.6030640602111816, + "learning_rate": 3.716490658001625e-06, + "loss": 1.7389, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.705246925354004, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.7408, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 3.181328535079956, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7092, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 3.0449445247650146, + "learning_rate": 3.710398050365557e-06, + "loss": 1.7399, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.650517225265503, + "learning_rate": 3.708367181153534e-06, + "loss": 1.6861, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.7593672275543213, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.6701, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 3.0486466884613037, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.7191, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.502850294113159, + "learning_rate": 3.702274573517466e-06, + "loss": 1.6982, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 1.9228183031082153, + "eval_runtime": 49.565, + "eval_samples_per_second": 20.176, + "eval_steps_per_second": 0.847, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.400770664215088, + "learning_rate": 3.700243704305443e-06, + "loss": 1.6666, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.5583295822143555, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.7971, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.519505739212036, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.7076, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 3.106309413909912, + "learning_rate": 3.694151096669375e-06, + "loss": 1.7279, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.516906499862671, + "learning_rate": 3.692120227457352e-06, + "loss": 1.6265, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.603166103363037, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.7122, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.842773199081421, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.7312, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.8860342502593994, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7229, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.517254114151001, + "learning_rate": 3.683996750609261e-06, + "loss": 1.7342, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.430025100708008, + "learning_rate": 3.681965881397238e-06, + "loss": 1.6817, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 1.9156773090362549, + "eval_runtime": 44.6903, + "eval_samples_per_second": 22.376, + "eval_steps_per_second": 0.94, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.340075969696045, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.7044, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.4758968353271484, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.6935, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.878903865814209, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.7021, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.388366222381592, + "learning_rate": 3.673842404549147e-06, + "loss": 1.7229, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.5789167881011963, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8112, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.743072509765625, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.7239, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.5485000610351562, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.7641, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.53190541267395, + "learning_rate": 3.665718927701056e-06, + "loss": 1.6996, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.3354969024658203, + "learning_rate": 3.663688058489034e-06, + "loss": 1.6623, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.37371826171875, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.7109, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 1.9122159481048584, + "eval_runtime": 43.8844, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.406137228012085, + "learning_rate": 3.659626320064988e-06, + "loss": 1.6958, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.4220657348632812, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7196, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.6203134059906006, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.6014, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.5388472080230713, + "learning_rate": 3.65353371242892e-06, + "loss": 1.7272, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.672658681869507, + "learning_rate": 3.651502843216897e-06, + "loss": 1.6786, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4366531372070312, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7198, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.4872887134552, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.7082, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.371206283569336, + "learning_rate": 3.645410235580829e-06, + "loss": 1.6964, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.496103286743164, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.716, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.658316135406494, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.7263, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 1.9115047454833984, + "eval_runtime": 43.8788, + "eval_samples_per_second": 22.79, + "eval_steps_per_second": 0.957, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.5931882858276367, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.7586, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.4697139263153076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.7232, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.7274153232574463, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.6917, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.683138132095337, + "learning_rate": 3.633225020308692e-06, + "loss": 1.7518, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.4510817527770996, + "learning_rate": 3.63119415109667e-06, + "loss": 1.6533, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.6939656734466553, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.6545, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.572864532470703, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.6965, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.4816951751708984, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.6851, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.640394926071167, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7336, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.6707329750061035, + "learning_rate": 3.621039805036556e-06, + "loss": 1.7182, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 1.91274094581604, + "eval_runtime": 44.35, + "eval_samples_per_second": 22.548, + "eval_steps_per_second": 0.947, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.6947290897369385, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.7255, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.5496068000793457, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.6209, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.6196374893188477, + "learning_rate": 3.614947197400488e-06, + "loss": 1.7277, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.4534807205200195, + "learning_rate": 3.612916328188465e-06, + "loss": 1.7404, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.5428082942962646, + "learning_rate": 3.610885458976442e-06, + "loss": 1.6618, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.5431482791900635, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.6743, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.707956552505493, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.7483, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.5643982887268066, + "learning_rate": 3.604792851340374e-06, + "loss": 1.7304, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.4612927436828613, + "learning_rate": 3.602761982128351e-06, + "loss": 1.6561, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.3140358924865723, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.6834, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 1.9124692678451538, + "eval_runtime": 44.0522, + "eval_samples_per_second": 22.7, + "eval_steps_per_second": 0.953, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4636526107788086, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.7395, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.541180372238159, + "learning_rate": 3.596669374492283e-06, + "loss": 1.6579, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.621335983276367, + "learning_rate": 3.59463850528026e-06, + "loss": 1.7332, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.363450050354004, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.7047, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.8609447479248047, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7416, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.457338571548462, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.7516, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.3943030834198, + "learning_rate": 3.586515028432169e-06, + "loss": 1.6436, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.4719340801239014, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.6988, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3442330360412598, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.7165, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.3391900062561035, + "learning_rate": 3.580422420796101e-06, + "loss": 1.6263, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 1.9100929498672485, + "eval_runtime": 44.2504, + "eval_samples_per_second": 22.599, + "eval_steps_per_second": 0.949, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.533566474914551, + "learning_rate": 3.578391551584078e-06, + "loss": 1.7141, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.464540481567383, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.7335, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.4789979457855225, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7074, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.8252034187316895, + "learning_rate": 3.57229894394801e-06, + "loss": 1.6808, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.5870537757873535, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.7171, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.349393129348755, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.6687, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.3343207836151123, + "learning_rate": 3.566206336311942e-06, + "loss": 1.6351, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.558645248413086, + "learning_rate": 3.564175467099919e-06, + "loss": 1.708, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.7598931789398193, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.6797, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.6048543453216553, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7117, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 1.906927466392517, + "eval_runtime": 44.5189, + "eval_samples_per_second": 22.462, + "eval_steps_per_second": 0.943, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.6124815940856934, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.7328, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.3410215377807617, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.6932, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.5247764587402344, + "learning_rate": 3.554021121039805e-06, + "loss": 1.6628, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.642958164215088, + "learning_rate": 3.551990251827783e-06, + "loss": 1.7035, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.3978517055511475, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.6796, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.3969287872314453, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.6626, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.5723509788513184, + "learning_rate": 3.545897644191714e-06, + "loss": 1.7135, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.545154571533203, + "learning_rate": 3.543866774979692e-06, + "loss": 1.6626, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.501058340072632, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.7112, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.695483684539795, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.6773, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 1.9063464403152466, + "eval_runtime": 44.2034, + "eval_samples_per_second": 22.623, + "eval_steps_per_second": 0.95, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 3.0584375858306885, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.7242, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.445220947265625, + "learning_rate": 3.535743298131601e-06, + "loss": 1.7349, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.427220106124878, + "learning_rate": 3.533712428919578e-06, + "loss": 1.677, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.5148212909698486, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.7007, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.644319534301758, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.7442, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.6118004322052, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.7509, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.5224156379699707, + "learning_rate": 3.525588952071487e-06, + "loss": 1.7279, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.4668405055999756, + "learning_rate": 3.523558082859464e-06, + "loss": 1.6754, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 3.8330910205841064, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.6815, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.569058895111084, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.6518, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 1.906802773475647, + "eval_runtime": 50.8521, + "eval_samples_per_second": 19.665, + "eval_steps_per_second": 0.826, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.705265522003174, + "learning_rate": 3.517465475223396e-06, + "loss": 1.7172, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 3.2667057514190674, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7641, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 3.5119211673736572, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.7178, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.7348034381866455, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.6815, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 3.019892454147339, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.6417, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 3.2486684322357178, + "learning_rate": 3.507311129163282e-06, + "loss": 1.6888, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.897071361541748, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7121, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.7342352867126465, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.6339, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.691178560256958, + "learning_rate": 3.501218521527214e-06, + "loss": 1.7069, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.548682689666748, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7218, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 1.9029825925827026, + "eval_runtime": 48.287, + "eval_samples_per_second": 20.71, + "eval_steps_per_second": 0.87, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.422056198120117, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.6486, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.4367074966430664, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.759, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.48002290725708, + "learning_rate": 3.493095044679123e-06, + "loss": 1.5902, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.5030295848846436, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.7108, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.5892179012298584, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.6154, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.4282920360565186, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.6807, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.4826700687408447, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7284, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.416597843170166, + "learning_rate": 3.482940698619009e-06, + "loss": 1.6972, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.4664933681488037, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.7119, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.541856288909912, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.6752, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 1.8999215364456177, + "eval_runtime": 46.3528, + "eval_samples_per_second": 21.574, + "eval_steps_per_second": 0.906, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4803109169006348, + "learning_rate": 3.476848090982941e-06, + "loss": 1.776, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.4363205432891846, + "learning_rate": 3.474817221770918e-06, + "loss": 1.7196, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.4377009868621826, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.6937, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.425791025161743, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.6547, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3216712474823, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.6789, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.65527081489563, + "learning_rate": 3.466693744922827e-06, + "loss": 1.7231, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.3415145874023438, + "learning_rate": 3.464662875710805e-06, + "loss": 1.6805, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.3864822387695312, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.7153, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.490896224975586, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.6866, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.339711904525757, + "learning_rate": 3.458570268074736e-06, + "loss": 1.6732, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 1.8985264301300049, + "eval_runtime": 46.7854, + "eval_samples_per_second": 21.374, + "eval_steps_per_second": 0.898, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.393709659576416, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.6549, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.564114809036255, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.7019, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3826403617858887, + "learning_rate": 3.452477660438668e-06, + "loss": 1.6427, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.3185465335845947, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.6916, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4503891468048096, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.6793, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.386446714401245, + "learning_rate": 3.4463850528026e-06, + "loss": 1.7186, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.6404714584350586, + "learning_rate": 3.444354183590577e-06, + "loss": 1.6605, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.342374324798584, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.6219, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.7503273487091064, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.693, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.5416243076324463, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7323, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 1.9016659259796143, + "eval_runtime": 48.5456, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.5787603855133057, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.7057, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.649528980255127, + "learning_rate": 3.434199837530463e-06, + "loss": 1.6348, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.504089832305908, + "learning_rate": 3.432168968318441e-06, + "loss": 1.7202, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.5624678134918213, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7351, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.641331911087036, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.7453, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.599500894546509, + "learning_rate": 3.426076360682372e-06, + "loss": 1.6942, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.668599843978882, + "learning_rate": 3.42404549147035e-06, + "loss": 1.6337, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.75667667388916, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.6988, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.641526937484741, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.7641, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.4972548484802246, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.7269, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 1.8979382514953613, + "eval_runtime": 45.6555, + "eval_samples_per_second": 21.903, + "eval_steps_per_second": 0.92, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.5454585552215576, + "learning_rate": 3.415922014622259e-06, + "loss": 1.7294, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.5921928882598877, + "learning_rate": 3.413891145410236e-06, + "loss": 1.7145, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.4028453826904297, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.6985, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.604522705078125, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.6586, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.528970241546631, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.734, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.565030574798584, + "learning_rate": 3.405767668562145e-06, + "loss": 1.6997, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.7735602855682373, + "learning_rate": 3.403736799350122e-06, + "loss": 1.6543, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.5116474628448486, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.7711, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.4827239513397217, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.6831, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.544309377670288, + "learning_rate": 3.397644191714054e-06, + "loss": 1.6905, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 1.8945367336273193, + "eval_runtime": 46.6027, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.456636905670166, + "learning_rate": 3.395613322502031e-06, + "loss": 1.646, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.5900352001190186, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.6967, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 3.3344573974609375, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.6947, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.5621678829193115, + "learning_rate": 3.389520714865963e-06, + "loss": 1.7084, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.5676684379577637, + "learning_rate": 3.38748984565394e-06, + "loss": 1.6788, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.600691318511963, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.6811, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.5324196815490723, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.661, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.7559397220611572, + "learning_rate": 3.381397238017872e-06, + "loss": 1.7473, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.526397466659546, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.3752379417419434, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.689, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 1.8995447158813477, + "eval_runtime": 48.0083, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.5902793407440186, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.6899, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 3.157573938369751, + "learning_rate": 3.373273761169781e-06, + "loss": 1.6792, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.628035545349121, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.7003, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.4810822010040283, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.6489, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.4859938621520996, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.6768, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.6773345470428467, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7046, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.9695260524749756, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.7341, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.8146679401397705, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.7472, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.534985065460205, + "learning_rate": 3.359057676685622e-06, + "loss": 1.6199, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.9351718425750732, + "learning_rate": 3.357026807473599e-06, + "loss": 1.6656, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 1.8962808847427368, + "eval_runtime": 47.0443, + "eval_samples_per_second": 21.257, + "eval_steps_per_second": 0.893, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.4404256343841553, + "learning_rate": 3.354995938261576e-06, + "loss": 1.6207, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.6691436767578125, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.6627, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.5436172485351562, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.6919, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.7167909145355225, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.6434, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.575446367263794, + "learning_rate": 3.346872461413485e-06, + "loss": 1.6554, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.3874504566192627, + "learning_rate": 3.344841592201463e-06, + "loss": 1.6691, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.66317081451416, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.6864, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.4396398067474365, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.7217, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.5002803802490234, + "learning_rate": 3.338748984565394e-06, + "loss": 1.7556, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.679577350616455, + "learning_rate": 3.336718115353372e-06, + "loss": 1.6822, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 1.8942477703094482, + "eval_runtime": 44.3804, + "eval_samples_per_second": 22.532, + "eval_steps_per_second": 0.946, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.6119370460510254, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.662, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.6024818420410156, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.687, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 4.128854751586914, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.6898, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3910412788391113, + "learning_rate": 3.32859463850528e-06, + "loss": 1.6274, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.8630096912384033, + "learning_rate": 3.326563769293258e-06, + "loss": 1.6554, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 4.201050281524658, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.6676, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.492168664932251, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.6319, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 3.6807360649108887, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.6776, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.7271926403045654, + "learning_rate": 3.318440292445167e-06, + "loss": 1.6383, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.5567307472229004, + "learning_rate": 3.316409423233144e-06, + "loss": 1.7325, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 1.892768383026123, + "eval_runtime": 50.909, + "eval_samples_per_second": 19.643, + "eval_steps_per_second": 0.825, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.5373101234436035, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.689, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.602923631668091, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.6861, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.507110118865967, + "learning_rate": 3.310316815597076e-06, + "loss": 1.7575, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.361786127090454, + "learning_rate": 3.308285946385053e-06, + "loss": 1.6267, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5360615253448486, + "learning_rate": 3.30625507717303e-06, + "loss": 1.6911, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.6238744258880615, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.7193, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.5358500480651855, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.6691, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.7285404205322266, + "learning_rate": 3.300162469536962e-06, + "loss": 1.7249, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.346160650253296, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.6958, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.5415399074554443, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.7544, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 1.8925012350082397, + "eval_runtime": 46.4957, + "eval_samples_per_second": 21.507, + "eval_steps_per_second": 0.903, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.5924015045166016, + "learning_rate": 3.294069861900894e-06, + "loss": 1.6901, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.4649693965911865, + "learning_rate": 3.292038992688871e-06, + "loss": 1.7234, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.3648035526275635, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.7324, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.5146145820617676, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.631, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.5919511318206787, + "learning_rate": 3.285946385052803e-06, + "loss": 1.706, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.299527168273926, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.608, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.480637788772583, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.7415, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.458183526992798, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.6806, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.4864983558654785, + "learning_rate": 3.277822908204712e-06, + "loss": 1.6373, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.3432068824768066, + "learning_rate": 3.275792038992689e-06, + "loss": 1.6832, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 1.8899296522140503, + "eval_runtime": 44.3309, + "eval_samples_per_second": 22.558, + "eval_steps_per_second": 0.947, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.365891218185425, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.6492, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.463261842727661, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.6847, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.3977205753326416, + "learning_rate": 3.269699431356621e-06, + "loss": 1.6789, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.4172558784484863, + "learning_rate": 3.267668562144598e-06, + "loss": 1.6495, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.663381814956665, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.7139, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.3702306747436523, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.6876, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 3.1759073734283447, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.7465, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.6344592571258545, + "learning_rate": 3.259545085296507e-06, + "loss": 1.6339, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.364490032196045, + "learning_rate": 3.257514216084484e-06, + "loss": 1.7394, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.5639748573303223, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.6928, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 1.8937300443649292, + "eval_runtime": 43.9356, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 3.152968645095825, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.6693, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.45776104927063, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.6999, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.5289807319641113, + "learning_rate": 3.249390739236393e-06, + "loss": 1.6523, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.4168646335601807, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.6823, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.5115208625793457, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.6816, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.685535192489624, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.6307, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.253084897994995, + "learning_rate": 3.241267262388302e-06, + "loss": 1.662, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.4825408458709717, + "learning_rate": 3.23923639317628e-06, + "loss": 1.6444, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.691168785095215, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.6912, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.978756904602051, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7536, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 1.8903727531433105, + "eval_runtime": 43.4099, + "eval_samples_per_second": 23.036, + "eval_steps_per_second": 0.968, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.317105531692505, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.6441, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 3.1030166149139404, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.7205, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.595792770385742, + "learning_rate": 3.229082047116166e-06, + "loss": 1.677, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.3315582275390625, + "learning_rate": 3.227051177904143e-06, + "loss": 1.6779, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.4262566566467285, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.7003, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.4438798427581787, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.6081, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.573531150817871, + "learning_rate": 3.220958570268075e-06, + "loss": 1.6489, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.477809429168701, + "learning_rate": 3.218927701056052e-06, + "loss": 1.6025, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.5499629974365234, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.7337, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.4699153900146484, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.6443, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 1.8867121934890747, + "eval_runtime": 42.9343, + "eval_samples_per_second": 23.291, + "eval_steps_per_second": 0.978, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.4200708866119385, + "learning_rate": 3.212835093419984e-06, + "loss": 1.6743, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.418069839477539, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.723, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.452284336090088, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.6358, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.4241347312927246, + "learning_rate": 3.206742485783916e-06, + "loss": 1.638, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.461860418319702, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.7017, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.591986894607544, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7065, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.478470802307129, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.6262, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.49831223487854, + "learning_rate": 3.198619008935825e-06, + "loss": 1.6871, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.485332727432251, + "learning_rate": 3.196588139723802e-06, + "loss": 1.6864, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.533710479736328, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.6794, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 1.88768470287323, + "eval_runtime": 43.5948, + "eval_samples_per_second": 22.938, + "eval_steps_per_second": 0.963, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.368609666824341, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.6877, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.4875223636627197, + "learning_rate": 3.190495532087734e-06, + "loss": 1.7406, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.5075156688690186, + "learning_rate": 3.188464662875711e-06, + "loss": 1.6545, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.428731918334961, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.6213, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.467618465423584, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.6263, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.561612129211426, + "learning_rate": 3.182372055239643e-06, + "loss": 1.6987, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.5035271644592285, + "learning_rate": 3.18034118602762e-06, + "loss": 1.6902, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.538311243057251, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.6523, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.486321210861206, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.6665, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.433173894882202, + "learning_rate": 3.174248578391552e-06, + "loss": 1.6605, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 1.8853652477264404, + "eval_runtime": 43.3417, + "eval_samples_per_second": 23.072, + "eval_steps_per_second": 0.969, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.4658470153808594, + "learning_rate": 3.172217709179529e-06, + "loss": 1.6615, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.452450752258301, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.6332, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.5310161113739014, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.6846, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.4504501819610596, + "learning_rate": 3.166125101543461e-06, + "loss": 1.6996, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.6004488468170166, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.6302, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.5000431537628174, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.6742, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.4869067668914795, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.6797, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.5018959045410156, + "learning_rate": 3.15800162469537e-06, + "loss": 1.673, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.6645195484161377, + "learning_rate": 3.155970755483347e-06, + "loss": 1.6908, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.3708560466766357, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.683, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 1.8829350471496582, + "eval_runtime": 43.5383, + "eval_samples_per_second": 22.968, + "eval_steps_per_second": 0.965, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.574108362197876, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.7155, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.4669814109802246, + "learning_rate": 3.149878147847279e-06, + "loss": 1.7529, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4713149070739746, + "learning_rate": 3.147847278635256e-06, + "loss": 1.6491, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.5128800868988037, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.6276, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.740370035171509, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.6778, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.6038050651550293, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.6527, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.782407283782959, + "learning_rate": 3.139723801787165e-06, + "loss": 1.6989, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.7673134803771973, + "learning_rate": 3.137692932575143e-06, + "loss": 1.6291, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.7472870349884033, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.6776, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.4887819290161133, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7009, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.8792160749435425, + "eval_runtime": 43.7334, + "eval_samples_per_second": 22.866, + "eval_steps_per_second": 0.96, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.4528491497039795, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.7245, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.5177690982818604, + "learning_rate": 3.129569455727051e-06, + "loss": 1.6852, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.9238510131835938, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.6994, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.446660041809082, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.6636, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.644211530685425, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.6983, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.697342872619629, + "learning_rate": 3.12144597887896e-06, + "loss": 1.7172, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.714172124862671, + "learning_rate": 3.119415109666938e-06, + "loss": 1.6702, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.471735715866089, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.6945, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.5089669227600098, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.7181, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.349229335784912, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.7297, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.880064845085144, + "eval_runtime": 47.0636, + "eval_samples_per_second": 21.248, + "eval_steps_per_second": 0.892, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.6843974590301514, + "learning_rate": 3.111291632818847e-06, + "loss": 1.6698, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.4325790405273438, + "learning_rate": 3.109260763606824e-06, + "loss": 1.6973, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.488459825515747, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.6552, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.4992082118988037, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.7114, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.4742603302001953, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.7073, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.556330442428589, + "learning_rate": 3.101137286758733e-06, + "loss": 1.7108, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.5711283683776855, + "learning_rate": 3.09910641754671e-06, + "loss": 1.6811, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.6467158794403076, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.7395, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.465963363647461, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.7006, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.622716188430786, + "learning_rate": 3.093013809910642e-06, + "loss": 1.7385, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.8808013200759888, + "eval_runtime": 45.5126, + "eval_samples_per_second": 21.972, + "eval_steps_per_second": 0.923, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.489701509475708, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.6949, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.4398739337921143, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.6802, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.3654873371124268, + "learning_rate": 3.086921202274574e-06, + "loss": 1.6452, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.518768548965454, + "learning_rate": 3.084890333062551e-06, + "loss": 1.6913, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.63173246383667, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.6611, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.531484603881836, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.611, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.4796738624572754, + "learning_rate": 3.078797725426483e-06, + "loss": 1.6764, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.39241099357605, + "learning_rate": 3.07676685621446e-06, + "loss": 1.6417, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3984735012054443, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.6546, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.5202760696411133, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.6879, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.879974365234375, + "eval_runtime": 44.395, + "eval_samples_per_second": 22.525, + "eval_steps_per_second": 0.946, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.5556983947753906, + "learning_rate": 3.070674248578392e-06, + "loss": 1.6733, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.3853955268859863, + "learning_rate": 3.068643379366369e-06, + "loss": 1.6274, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.4503982067108154, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.6896, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.5618021488189697, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.6592, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.6265292167663574, + "learning_rate": 3.062550771730301e-06, + "loss": 1.6634, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.4617390632629395, + "learning_rate": 3.060519902518278e-06, + "loss": 1.6032, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.4191603660583496, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.6871, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.5509119033813477, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.653, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.5399646759033203, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.7046, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.446967124938965, + "learning_rate": 3.052396425670187e-06, + "loss": 1.6547, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.8803874254226685, + "eval_runtime": 45.361, + "eval_samples_per_second": 22.045, + "eval_steps_per_second": 0.926, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.419206142425537, + "learning_rate": 3.050365556458164e-06, + "loss": 1.6517, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.4592065811157227, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.693, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.5429770946502686, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7124, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.5141642093658447, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.7103, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.362539291381836, + "learning_rate": 3.042242079610073e-06, + "loss": 1.6205, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.401175022125244, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.6479, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.5883190631866455, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.6732, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.5273513793945312, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.6673, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.6569600105285645, + "learning_rate": 3.034118602761982e-06, + "loss": 1.6396, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.8843557834625244, + "learning_rate": 3.03208773354996e-06, + "loss": 1.7136, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.8763896226882935, + "eval_runtime": 46.1886, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.6459810733795166, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.6939, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.4619321823120117, + "learning_rate": 3.028025995125914e-06, + "loss": 1.6197, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.390430450439453, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.6805, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.5494728088378906, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.6116, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.5397655963897705, + "learning_rate": 3.021933387489846e-06, + "loss": 1.6753, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.634716510772705, + "learning_rate": 3.019902518277823e-06, + "loss": 1.6892, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.5844202041625977, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.6401, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.577362060546875, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.6405, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.3554489612579346, + "learning_rate": 3.013809910641755e-06, + "loss": 1.6427, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.893918037414551, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.6739, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.8772739171981812, + "eval_runtime": 44.9972, + "eval_samples_per_second": 22.224, + "eval_steps_per_second": 0.933, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.4818382263183594, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.644, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.4990177154541016, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.6962, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.627502918243408, + "learning_rate": 3.005686433793664e-06, + "loss": 1.639, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.5949175357818604, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.7095, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.422765016555786, + "learning_rate": 3.001624695369618e-06, + "loss": 1.6572, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.708702564239502, + "learning_rate": 2.999593826157596e-06, + "loss": 1.6747, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.624866247177124, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.6643, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.5831358432769775, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.7036, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.5799436569213867, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.6628, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.7047722339630127, + "learning_rate": 2.991470349309505e-06, + "loss": 1.6591, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.8736577033996582, + "eval_runtime": 44.7095, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.411628246307373, + "learning_rate": 2.989439480097482e-06, + "loss": 1.6644, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.8774638175964355, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.7192, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.857028007507324, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7079, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.4619641304016113, + "learning_rate": 2.983346872461414e-06, + "loss": 1.6846, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.6566193103790283, + "learning_rate": 2.981316003249391e-06, + "loss": 1.6768, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.681217670440674, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.7002, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.39505672454834, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.65, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.6062734127044678, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.6659, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.6402504444122314, + "learning_rate": 2.9731925264013e-06, + "loss": 1.6491, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.497014284133911, + "learning_rate": 2.971161657189277e-06, + "loss": 1.6383, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.874670147895813, + "eval_runtime": 42.9696, + "eval_samples_per_second": 23.272, + "eval_steps_per_second": 0.977, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.9070584774017334, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.6388, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.580744981765747, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7078, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.4637644290924072, + "learning_rate": 2.965069049553209e-06, + "loss": 1.6266, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.4489941596984863, + "learning_rate": 2.963038180341186e-06, + "loss": 1.705, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.664654016494751, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.6964, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.7017202377319336, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.6356, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.3691024780273438, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.6566, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.540384292602539, + "learning_rate": 2.954914703493095e-06, + "loss": 1.6334, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.5240535736083984, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.6316, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 3.105684518814087, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.6769, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.8741132020950317, + "eval_runtime": 42.6795, + "eval_samples_per_second": 23.43, + "eval_steps_per_second": 0.984, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.494252920150757, + "learning_rate": 2.948822095857027e-06, + "loss": 1.6359, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 4.069019317626953, + "learning_rate": 2.946791226645004e-06, + "loss": 1.621, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.641447067260742, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.6502, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.6700453758239746, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.6593, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.686068296432495, + "learning_rate": 2.940698619008936e-06, + "loss": 1.67, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.409055471420288, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.6809, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.4707298278808594, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.69, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.4735982418060303, + "learning_rate": 2.934606011372868e-06, + "loss": 1.666, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.633183479309082, + "learning_rate": 2.932575142160845e-06, + "loss": 1.6597, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.4808642864227295, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.6667, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.8731157779693604, + "eval_runtime": 41.3423, + "eval_samples_per_second": 24.188, + "eval_steps_per_second": 1.016, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.4044833183288574, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.7028, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.942699909210205, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.6678, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.505645990371704, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.5972, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.700967788696289, + "learning_rate": 2.922420796100731e-06, + "loss": 1.7064, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.4805755615234375, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.6201, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.753181219100952, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.6704, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.488865613937378, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7057, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.5464885234832764, + "learning_rate": 2.91429731925264e-06, + "loss": 1.6678, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.4997060298919678, + "learning_rate": 2.912266450040618e-06, + "loss": 1.662, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.5162436962127686, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.6501, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.869848370552063, + "eval_runtime": 43.3583, + "eval_samples_per_second": 23.064, + "eval_steps_per_second": 0.969, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.616129159927368, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.6206, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.761080503463745, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7131, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.5994064807891846, + "learning_rate": 2.904142973192527e-06, + "loss": 1.6693, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 3.4927968978881836, + "learning_rate": 2.902112103980504e-06, + "loss": 1.6694, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.6050939559936523, + "learning_rate": 2.900081234768481e-06, + "loss": 1.6419, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.796086311340332, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.6354, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.5658411979675293, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.652, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.3802385330200195, + "learning_rate": 2.893988627132413e-06, + "loss": 1.5902, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.479189872741699, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.6617, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.4367568492889404, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.6691, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.868497371673584, + "eval_runtime": 45.172, + "eval_samples_per_second": 22.138, + "eval_steps_per_second": 0.93, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.51888370513916, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.6872, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.5192768573760986, + "learning_rate": 2.885865150284322e-06, + "loss": 1.6584, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.5898501873016357, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.6179, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.698101043701172, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.663, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.889942169189453, + "learning_rate": 2.879772542648254e-06, + "loss": 1.5954, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.7909929752349854, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.7264, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 3.1931703090667725, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7091, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.7934632301330566, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.7204, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.470545530319214, + "learning_rate": 2.871649065800163e-06, + "loss": 1.644, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 3.011662006378174, + "learning_rate": 2.86961819658814e-06, + "loss": 1.6743, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.8678737878799438, + "eval_runtime": 44.7875, + "eval_samples_per_second": 22.328, + "eval_steps_per_second": 0.938, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.60823655128479, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.6467, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.5347230434417725, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.7044, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.481715440750122, + "learning_rate": 2.863525588952072e-06, + "loss": 1.5969, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.4966487884521484, + "learning_rate": 2.861494719740049e-06, + "loss": 1.6231, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.8809802532196045, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.6562, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.570204734802246, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.6511, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.6073813438415527, + "learning_rate": 2.855402112103981e-06, + "loss": 1.6456, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.8125052452087402, + "learning_rate": 2.853371242891958e-06, + "loss": 1.6722, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 5.341272354125977, + "learning_rate": 2.851340373679935e-06, + "loss": 1.7025, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 3.138002395629883, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.6676, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.8733265399932861, + "eval_runtime": 45.4915, + "eval_samples_per_second": 21.982, + "eval_steps_per_second": 0.923, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 7.264789581298828, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.6335, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 3.8533623218536377, + "learning_rate": 2.845247766043867e-06, + "loss": 1.6029, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 4.1789164543151855, + "learning_rate": 2.843216896831844e-06, + "loss": 1.6368, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 3.117408037185669, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.6924, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.721306085586548, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.6914, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.7099668979644775, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.6508, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 3.7823967933654785, + "learning_rate": 2.835093419983753e-06, + "loss": 1.636, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.696370840072632, + "learning_rate": 2.833062550771731e-06, + "loss": 1.6175, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.7303690910339355, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.6654, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.867727041244507, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.6717, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.8720332384109497, + "eval_runtime": 44.164, + "eval_samples_per_second": 22.643, + "eval_steps_per_second": 0.951, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 3.1209630966186523, + "learning_rate": 2.826969943135662e-06, + "loss": 1.6635, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.5057783126831055, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.6309, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.6650168895721436, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.6365, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.371305465698242, + "learning_rate": 2.820877335499594e-06, + "loss": 1.603, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.387404203414917, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.6189, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.6265947818756104, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.6511, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.6108627319335938, + "learning_rate": 2.814784727863526e-06, + "loss": 1.6727, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.7021450996398926, + "learning_rate": 2.812753858651503e-06, + "loss": 1.6334, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.470118999481201, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.6155, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5722644329071045, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.6788, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.8727209568023682, + "eval_runtime": 44.6253, + "eval_samples_per_second": 22.409, + "eval_steps_per_second": 0.941, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.7143876552581787, + "learning_rate": 2.806661251015435e-06, + "loss": 1.6956, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.8522002696990967, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.7558, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.5352110862731934, + "learning_rate": 2.802599512591389e-06, + "loss": 1.6592, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.7065908908843994, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.6121, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.616450786590576, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.6643, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.528763771057129, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7104, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 3.725569009780884, + "learning_rate": 2.794476035743298e-06, + "loss": 1.6713, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.487497091293335, + "learning_rate": 2.792445166531276e-06, + "loss": 1.692, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.534766912460327, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.6257, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.8553240299224854, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.6147, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.8675988912582397, + "eval_runtime": 46.4445, + "eval_samples_per_second": 21.531, + "eval_steps_per_second": 0.904, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.5054166316986084, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.6082, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.4133870601654053, + "learning_rate": 2.784321689683185e-06, + "loss": 1.6148, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.5863375663757324, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7032, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.58491849899292, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.679, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.5225989818573, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.6688, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.366861343383789, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.7076, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.4159810543060303, + "learning_rate": 2.774167343623071e-06, + "loss": 1.6301, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.5394275188446045, + "learning_rate": 2.772136474411048e-06, + "loss": 1.5758, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.409821033477783, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.664, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.4548892974853516, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.643, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.8672850131988525, + "eval_runtime": 46.3149, + "eval_samples_per_second": 21.591, + "eval_steps_per_second": 0.907, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.7483904361724854, + "learning_rate": 2.76604386677498e-06, + "loss": 1.6898, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.538804769515991, + "learning_rate": 2.764012997562957e-06, + "loss": 1.6019, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.7153313159942627, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.691, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.6892242431640625, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.6386, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.550957679748535, + "learning_rate": 2.757920389926889e-06, + "loss": 1.6237, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.3658559322357178, + "learning_rate": 2.755889520714866e-06, + "loss": 1.6015, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.6524102687835693, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.6644, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.7067437171936035, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.687, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.4044861793518066, + "learning_rate": 2.749796913078798e-06, + "loss": 1.6535, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.4933829307556152, + "learning_rate": 2.747766043866775e-06, + "loss": 1.7017, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.8652688264846802, + "eval_runtime": 46.8502, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.7062129974365234, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7398, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.5704996585845947, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.6561, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.5875790119171143, + "learning_rate": 2.741673436230707e-06, + "loss": 1.643, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.7498223781585693, + "learning_rate": 2.739642567018684e-06, + "loss": 1.6652, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.7275257110595703, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.6559, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.6349947452545166, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.6489, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.7058634757995605, + "learning_rate": 2.733549959382616e-06, + "loss": 1.7176, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.481344223022461, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.685, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.468277931213379, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.6305, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.6183228492736816, + "learning_rate": 2.727457351746548e-06, + "loss": 1.6151, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.864329218864441, + "eval_runtime": 43.0443, + "eval_samples_per_second": 23.232, + "eval_steps_per_second": 0.976, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.50679349899292, + "learning_rate": 2.725426482534525e-06, + "loss": 1.6468, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.6565401554107666, + "learning_rate": 2.723395613322502e-06, + "loss": 1.6496, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.5332367420196533, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.6424, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.6645619869232178, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.8012, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.7359142303466797, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.6025, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.660728693008423, + "learning_rate": 2.715272136474411e-06, + "loss": 1.6601, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.5184221267700195, + "learning_rate": 2.713241267262389e-06, + "loss": 1.6383, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.643712043762207, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.6834, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.504753351211548, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.6292, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.3439948558807373, + "learning_rate": 2.70714865962632e-06, + "loss": 1.6558, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.861581563949585, + "eval_runtime": 43.9224, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4820125102996826, + "learning_rate": 2.705117790414298e-06, + "loss": 1.69, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.6156861782073975, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.6304, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.490123987197876, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.6573, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.4439024925231934, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.6368, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.891603708267212, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7111, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.481175661087036, + "learning_rate": 2.694963444354184e-06, + "loss": 1.6408, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.5378847122192383, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.6782, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.592477321624756, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.6085, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.9336471557617188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.6719, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.3115012645721436, + "learning_rate": 2.686839967506093e-06, + "loss": 1.5797, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.8649041652679443, + "eval_runtime": 43.711, + "eval_samples_per_second": 22.878, + "eval_steps_per_second": 0.961, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.627028465270996, + "learning_rate": 2.68480909829407e-06, + "loss": 1.5784, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.917670488357544, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7007, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.418285608291626, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.6613, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.607773780822754, + "learning_rate": 2.678716490658002e-06, + "loss": 1.6561, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.569714307785034, + "learning_rate": 2.676685621445979e-06, + "loss": 1.6541, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.359494209289551, + "learning_rate": 2.674654752233956e-06, + "loss": 1.6467, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.5155694484710693, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.7003, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.439147710800171, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.5679, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.5044116973876953, + "learning_rate": 2.668562144597888e-06, + "loss": 1.6626, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.4475765228271484, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.656, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.8636877536773682, + "eval_runtime": 43.5138, + "eval_samples_per_second": 22.981, + "eval_steps_per_second": 0.965, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4616832733154297, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.6101, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.4125123023986816, + "learning_rate": 2.66246953696182e-06, + "loss": 1.6021, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.571326494216919, + "learning_rate": 2.660438667749797e-06, + "loss": 1.6482, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.424332618713379, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.653, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.3034703731536865, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.6096, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.4758505821228027, + "learning_rate": 2.654346060113729e-06, + "loss": 1.6274, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 3.800152540206909, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.6217, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.479827880859375, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7202, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.604523181915283, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.6778, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 3.595973491668701, + "learning_rate": 2.646222583265638e-06, + "loss": 1.6678, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.8647090196609497, + "eval_runtime": 43.0415, + "eval_samples_per_second": 23.233, + "eval_steps_per_second": 0.976, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 3.297319173812866, + "learning_rate": 2.644191714053615e-06, + "loss": 1.637, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.423567056655884, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.6018, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 4.4547953605651855, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.6594, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 3.0775582790374756, + "learning_rate": 2.638099106417547e-06, + "loss": 1.6259, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.747938871383667, + "learning_rate": 2.636068237205524e-06, + "loss": 1.7208, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.5350255966186523, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.5705, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 3.0630743503570557, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.6507, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 3.5475759506225586, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7045, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.534303903579712, + "learning_rate": 2.627944760357433e-06, + "loss": 1.7173, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.6964452266693115, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.6823, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.865207552909851, + "eval_runtime": 42.9435, + "eval_samples_per_second": 23.286, + "eval_steps_per_second": 0.978, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.924957752227783, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.6036, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.7058141231536865, + "learning_rate": 2.621852152721365e-06, + "loss": 1.6701, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.735339641571045, + "learning_rate": 2.619821283509342e-06, + "loss": 1.607, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.573958396911621, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.667, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5776426792144775, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.6938, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 4.454444408416748, + "learning_rate": 2.613728675873274e-06, + "loss": 1.7139, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.4883337020874023, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.6524, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.672116994857788, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.6197, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 3.511932611465454, + "learning_rate": 2.607636068237206e-06, + "loss": 1.6739, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.5960047245025635, + "learning_rate": 2.605605199025183e-06, + "loss": 1.6516, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.8604435920715332, + "eval_runtime": 42.9247, + "eval_samples_per_second": 23.297, + "eval_steps_per_second": 0.978, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.565376043319702, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.6884, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.676114797592163, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.6693, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.582751750946045, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.6499, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.748908281326294, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.6837, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.5578067302703857, + "learning_rate": 2.595450852965069e-06, + "loss": 1.6959, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.5702805519104004, + "learning_rate": 2.593419983753047e-06, + "loss": 1.6429, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.4748787879943848, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.6878, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.6157331466674805, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.6246, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.8377959728240967, + "learning_rate": 2.587327376116978e-06, + "loss": 1.6451, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.485398292541504, + "learning_rate": 2.585296506904956e-06, + "loss": 1.6288, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.8593864440917969, + "eval_runtime": 44.6341, + "eval_samples_per_second": 22.404, + "eval_steps_per_second": 0.941, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.46850323677063, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.5766, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.673283576965332, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7383, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.935293197631836, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.5742, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.531733751296997, + "learning_rate": 2.577173030056865e-06, + "loss": 1.6714, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.528425455093384, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6802, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.5644803047180176, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.6201, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.382451295852661, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.5824, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.63320255279541, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.6685, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.488844394683838, + "learning_rate": 2.567018683996751e-06, + "loss": 1.6078, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.7845425605773926, + "learning_rate": 2.564987814784728e-06, + "loss": 1.638, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.8572624921798706, + "eval_runtime": 43.2181, + "eval_samples_per_second": 23.138, + "eval_steps_per_second": 0.972, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.709656000137329, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.6425, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.651277542114258, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7401, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.5424256324768066, + "learning_rate": 2.55889520714866e-06, + "loss": 1.6272, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.6508235931396484, + "learning_rate": 2.556864337936637e-06, + "loss": 1.6384, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.882636308670044, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.6965, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.518367290496826, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.6551, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.7977044582366943, + "learning_rate": 2.550771730300569e-06, + "loss": 1.679, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.353041172027588, + "learning_rate": 2.548740861088546e-06, + "loss": 1.5745, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.5718655586242676, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.5655, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.6955313682556152, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.6764, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.8569272756576538, + "eval_runtime": 45.3844, + "eval_samples_per_second": 22.034, + "eval_steps_per_second": 0.925, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4756696224212646, + "learning_rate": 2.542648253452478e-06, + "loss": 1.6522, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.573193311691284, + "learning_rate": 2.540617384240455e-06, + "loss": 1.6079, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.5967671871185303, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.6341, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.7896342277526855, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.6449, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.519728183746338, + "learning_rate": 2.534524776604387e-06, + "loss": 1.549, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.543139934539795, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.6335, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.5580177307128906, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6402, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.443547010421753, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.6421, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.7075212001800537, + "learning_rate": 2.526401299756296e-06, + "loss": 1.6283, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.6505324840545654, + "learning_rate": 2.524370430544273e-06, + "loss": 1.6028, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.8556184768676758, + "eval_runtime": 43.731, + "eval_samples_per_second": 22.867, + "eval_steps_per_second": 0.96, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.597172260284424, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.6631, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 3.1529037952423096, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.6219, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.5374975204467773, + "learning_rate": 2.518277822908205e-06, + "loss": 1.6476, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 3.0557401180267334, + "learning_rate": 2.516246953696182e-06, + "loss": 1.6627, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 6.759350776672363, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.6532, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.670427083969116, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.658, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 4.269686698913574, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.5873, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 13.37558650970459, + "learning_rate": 2.508123476848091e-06, + "loss": 1.6872, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 7.060877323150635, + "learning_rate": 2.506092607636069e-06, + "loss": 1.6406, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 11.259824752807617, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.6421, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.861360788345337, + "eval_runtime": 45.4438, + "eval_samples_per_second": 22.005, + "eval_steps_per_second": 0.924, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 5.861843585968018, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.701, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 3.9136719703674316, + "learning_rate": 2.5e-06, + "loss": 1.6289, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 12.720739364624023, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.6532, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 5.970798492431641, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7035, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 3.490950584411621, + "learning_rate": 2.493907392363932e-06, + "loss": 1.6565, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 5.826146602630615, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.655, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 9.610876083374023, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.7295, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 3.754833221435547, + "learning_rate": 2.487814784727864e-06, + "loss": 1.6406, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.7952232360839844, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.6463, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 6.3178558349609375, + "learning_rate": 2.483753046303818e-06, + "loss": 1.6474, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.8672106266021729, + "eval_runtime": 50.0163, + "eval_samples_per_second": 19.993, + "eval_steps_per_second": 0.84, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 7.4854960441589355, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.6798, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.8903026580810547, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.667, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.481947898864746, + "learning_rate": 2.47766043866775e-06, + "loss": 1.6677, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.8459818363189697, + "learning_rate": 2.475629569455727e-06, + "loss": 1.6129, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 5.00568962097168, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.6233, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 3.7508153915405273, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.6414, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.7895889282226562, + "learning_rate": 2.469536961819659e-06, + "loss": 1.65, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.532818555831909, + "learning_rate": 2.467506092607636e-06, + "loss": 1.6174, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.699639320373535, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7128, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.5636072158813477, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.6416, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.8608473539352417, + "eval_runtime": 43.8315, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.7466976642608643, + "learning_rate": 2.461413484971568e-06, + "loss": 1.7246, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.690990686416626, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.596, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.868340492248535, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.622, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.4456427097320557, + "learning_rate": 2.4553208773355e-06, + "loss": 1.6446, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.5499041080474854, + "learning_rate": 2.453290008123477e-06, + "loss": 1.6587, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.68770170211792, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.6612, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.589557647705078, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.6428, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.7014052867889404, + "learning_rate": 2.447197400487409e-06, + "loss": 1.6605, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.6242196559906006, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.6897, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.5285110473632812, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.6636, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.8544306755065918, + "eval_runtime": 45.6109, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.484952449798584, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.6636, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.5793070793151855, + "learning_rate": 2.439073923639318e-06, + "loss": 1.6044, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.5558605194091797, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.5885, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.6365678310394287, + "learning_rate": 2.435012185215272e-06, + "loss": 1.6236, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.428422451019287, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.6446, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.535776138305664, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.6252, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.743934154510498, + "learning_rate": 2.428919577579204e-06, + "loss": 1.729, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.4823739528656006, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.6251, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.4650003910064697, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.6191, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.4338109493255615, + "learning_rate": 2.422826969943136e-06, + "loss": 1.5919, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.8501240015029907, + "eval_runtime": 43.8558, + "eval_samples_per_second": 22.802, + "eval_steps_per_second": 0.958, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.6054813861846924, + "learning_rate": 2.420796100731113e-06, + "loss": 1.6485, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.390143394470215, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.661, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.467470645904541, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.6308, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.360908269882202, + "learning_rate": 2.414703493095045e-06, + "loss": 1.6394, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.5966176986694336, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.6138, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.528402328491211, + "learning_rate": 2.410641754670999e-06, + "loss": 1.6817, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.5319254398345947, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.6189, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.559260129928589, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.6476, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.5595016479492188, + "learning_rate": 2.404549147034931e-06, + "loss": 1.6878, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.488196849822998, + "learning_rate": 2.402518277822908e-06, + "loss": 1.6095, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.8484610319137573, + "eval_runtime": 43.4829, + "eval_samples_per_second": 22.998, + "eval_steps_per_second": 0.966, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6928117275238037, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.6533, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.485538959503174, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.6174, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.4725208282470703, + "learning_rate": 2.39642567018684e-06, + "loss": 1.6306, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.456028461456299, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.6586, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3731250762939453, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.6009, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.4567086696624756, + "learning_rate": 2.390333062550772e-06, + "loss": 1.5257, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.577141284942627, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7127, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.4612233638763428, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.638, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.463212728500366, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.5933, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.5121495723724365, + "learning_rate": 2.382209585702681e-06, + "loss": 1.68, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.8499811887741089, + "eval_runtime": 45.066, + "eval_samples_per_second": 22.19, + "eval_steps_per_second": 0.932, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.658245801925659, + "learning_rate": 2.380178716490658e-06, + "loss": 1.6354, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.6265509128570557, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.6755, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 3.016807794570923, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.6764, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.6552720069885254, + "learning_rate": 2.37408610885459e-06, + "loss": 1.6281, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.520078659057617, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.7077, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.528329849243164, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.6494, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.623776912689209, + "learning_rate": 2.367993501218522e-06, + "loss": 1.6436, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.56669020652771, + "learning_rate": 2.365962632006499e-06, + "loss": 1.6762, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.5415968894958496, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.6711, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.5164198875427246, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.6725, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.8490949869155884, + "eval_runtime": 43.612, + "eval_samples_per_second": 22.929, + "eval_steps_per_second": 0.963, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.626594066619873, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.6637, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3480069637298584, + "learning_rate": 2.357839155158408e-06, + "loss": 1.6561, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.644103527069092, + "learning_rate": 2.355808285946385e-06, + "loss": 1.6763, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.5539023876190186, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.653, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.5055317878723145, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.6643, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.4204092025756836, + "learning_rate": 2.349715678310317e-06, + "loss": 1.6133, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.5867843627929688, + "learning_rate": 2.347684809098294e-06, + "loss": 1.6804, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.5993685722351074, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.6101, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.565439462661743, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.6221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.5296084880828857, + "learning_rate": 2.341592201462226e-06, + "loss": 1.6234, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.8482695817947388, + "eval_runtime": 45.1972, + "eval_samples_per_second": 22.125, + "eval_steps_per_second": 0.929, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.6142499446868896, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.5886, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.486616849899292, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.6097, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.5778701305389404, + "learning_rate": 2.335499593826158e-06, + "loss": 1.6585, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.6165313720703125, + "learning_rate": 2.333468724614135e-06, + "loss": 1.663, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.4976565837860107, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.6083, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.485017776489258, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.6006, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.620819330215454, + "learning_rate": 2.327376116978067e-06, + "loss": 1.6641, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.528599262237549, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.6128, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.541442632675171, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.6545, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.4722139835357666, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.6146, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.848878264427185, + "eval_runtime": 43.7961, + "eval_samples_per_second": 22.833, + "eval_steps_per_second": 0.959, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.6106903553009033, + "learning_rate": 2.319252640129976e-06, + "loss": 1.7086, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.5255045890808105, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.6738, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.43829083442688, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.5861, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.527411460876465, + "learning_rate": 2.313160032493908e-06, + "loss": 1.6634, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.4998528957366943, + "learning_rate": 2.311129163281885e-06, + "loss": 1.6187, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.4838368892669678, + "learning_rate": 2.309098294069862e-06, + "loss": 1.5999, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.466900110244751, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.6446, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.583103895187378, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.6658, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.478386163711548, + "learning_rate": 2.303005686433794e-06, + "loss": 1.6261, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.5724942684173584, + "learning_rate": 2.300974817221771e-06, + "loss": 1.5955, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.8444501161575317, + "eval_runtime": 50.0263, + "eval_samples_per_second": 19.989, + "eval_steps_per_second": 0.84, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.601644277572632, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.6493, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.5638768672943115, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.6702, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.4802567958831787, + "learning_rate": 2.294882209585703e-06, + "loss": 1.6036, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.5114285945892334, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.642, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.487563133239746, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.673, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.6014463901519775, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.6261, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.53680157661438, + "learning_rate": 2.286758732737612e-06, + "loss": 1.6364, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.764408588409424, + "learning_rate": 2.284727863525589e-06, + "loss": 1.6791, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.8998825550079346, + "learning_rate": 2.282696994313566e-06, + "loss": 1.644, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.5797383785247803, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.6267, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.8467868566513062, + "eval_runtime": 49.4778, + "eval_samples_per_second": 20.211, + "eval_steps_per_second": 0.849, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.447178602218628, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.5955, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.6275408267974854, + "learning_rate": 2.276604386677498e-06, + "loss": 1.6117, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.6952037811279297, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.6894, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.5299718379974365, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.6361, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.837455987930298, + "learning_rate": 2.27051177904143e-06, + "loss": 1.6269, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.5406227111816406, + "learning_rate": 2.268480909829407e-06, + "loss": 1.6538, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.716038227081299, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.6691, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.883439064025879, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.6298, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.7631118297576904, + "learning_rate": 2.262388302193339e-06, + "loss": 1.6245, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.6777117252349854, + "learning_rate": 2.260357432981316e-06, + "loss": 1.6446, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.845224142074585, + "eval_runtime": 46.8476, + "eval_samples_per_second": 21.346, + "eval_steps_per_second": 0.897, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.6280243396759033, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.6398, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.471925973892212, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.6403, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.6293203830718994, + "learning_rate": 2.254264825345248e-06, + "loss": 1.642, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.651144027709961, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.6445, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.6489877700805664, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.6761, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.5869088172912598, + "learning_rate": 2.24817221770918e-06, + "loss": 1.6593, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.411802053451538, + "learning_rate": 2.246141348497157e-06, + "loss": 1.6232, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.522610902786255, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.6358, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.5430991649627686, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.6605, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.4440951347351074, + "learning_rate": 2.240048740861089e-06, + "loss": 1.5354, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.8440407514572144, + "eval_runtime": 48.0239, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.5492775440216064, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.668, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.7511909008026123, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6424, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.5176775455474854, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.6336, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.5408577919006348, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.5843, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.4791393280029297, + "learning_rate": 2.229894394800975e-06, + "loss": 1.555, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.383148670196533, + "learning_rate": 2.227863525588952e-06, + "loss": 1.598, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.429348945617676, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.5943, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.708796501159668, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.6382, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 3.0029661655426025, + "learning_rate": 2.221770917952884e-06, + "loss": 1.6699, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.5410728454589844, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.5965, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.8443207740783691, + "eval_runtime": 47.8671, + "eval_samples_per_second": 20.891, + "eval_steps_per_second": 0.877, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.5718178749084473, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.6214, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.474177837371826, + "learning_rate": 2.215678310316816e-06, + "loss": 1.6498, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.5960052013397217, + "learning_rate": 2.213647441104793e-06, + "loss": 1.6156, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.6217992305755615, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.6311, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.5601158142089844, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.6008, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.7786355018615723, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.62, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.548356771469116, + "learning_rate": 2.205523964256702e-06, + "loss": 1.6473, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.6265289783477783, + "learning_rate": 2.203493095044679e-06, + "loss": 1.6565, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.522379159927368, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.6764, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.5479297637939453, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.6194, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.8434110879898071, + "eval_runtime": 46.6043, + "eval_samples_per_second": 21.457, + "eval_steps_per_second": 0.901, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.5401806831359863, + "learning_rate": 2.197400487408611e-06, + "loss": 1.5911, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.466893434524536, + "learning_rate": 2.195369618196588e-06, + "loss": 1.6145, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.566653251647949, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.6558, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.664053201675415, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7089, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 3.6594645977020264, + "learning_rate": 2.18927701056052e-06, + "loss": 1.6549, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.68450927734375, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.6306, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.7094614505767822, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.6147, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.481332540512085, + "learning_rate": 2.183184402924452e-06, + "loss": 1.6612, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5651051998138428, + "learning_rate": 2.181153533712429e-06, + "loss": 1.6394, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.5570919513702393, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.6162, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.8433003425598145, + "eval_runtime": 46.8673, + "eval_samples_per_second": 21.337, + "eval_steps_per_second": 0.896, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.6206037998199463, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.6869, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.8962907791137695, + "learning_rate": 2.175060926076361e-06, + "loss": 1.7022, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.6211588382720947, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.6249, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.7770566940307617, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.5821, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.543658494949341, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.6587, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.508188486099243, + "learning_rate": 2.16693744922827e-06, + "loss": 1.6114, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.703335762023926, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.6152, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.469688892364502, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.6121, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.4836199283599854, + "learning_rate": 2.160844841592202e-06, + "loss": 1.5918, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.521851062774658, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.6228, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.845065712928772, + "eval_runtime": 45.3003, + "eval_samples_per_second": 22.075, + "eval_steps_per_second": 0.927, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.7808425426483154, + "learning_rate": 2.156783103168156e-06, + "loss": 1.5887, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.696708917617798, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.6517, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.7222893238067627, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.6817, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.504417896270752, + "learning_rate": 2.150690495532088e-06, + "loss": 1.647, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.574836015701294, + "learning_rate": 2.148659626320065e-06, + "loss": 1.6554, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.5004355907440186, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.6089, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.6979293823242188, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.6226, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.5239524841308594, + "learning_rate": 2.142567018683997e-06, + "loss": 1.6875, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.5218467712402344, + "learning_rate": 2.140536149471974e-06, + "loss": 1.6095, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.5778310298919678, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.6155, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.8443745374679565, + "eval_runtime": 46.6273, + "eval_samples_per_second": 21.447, + "eval_steps_per_second": 0.901, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.671973705291748, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.6696, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.5520615577697754, + "learning_rate": 2.134443541835906e-06, + "loss": 1.6291, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.39286208152771, + "learning_rate": 2.132412672623883e-06, + "loss": 1.6012, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.8277461528778076, + "learning_rate": 2.13038180341186e-06, + "loss": 1.6398, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.7147881984710693, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.6685, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.499044418334961, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.6128, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.514784574508667, + "learning_rate": 2.124289195775792e-06, + "loss": 1.6263, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.505342483520508, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.5532, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.608139753341675, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.5946, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.5578057765960693, + "learning_rate": 2.118196588139724e-06, + "loss": 1.6103, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.8425958156585693, + "eval_runtime": 44.206, + "eval_samples_per_second": 22.621, + "eval_steps_per_second": 0.95, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.4881014823913574, + "learning_rate": 2.116165718927701e-06, + "loss": 1.6364, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.4141550064086914, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.5943, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.649827718734741, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.5994, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.547238826751709, + "learning_rate": 2.110073111291633e-06, + "loss": 1.6247, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.6515395641326904, + "learning_rate": 2.10804224207961e-06, + "loss": 1.6543, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.5575809478759766, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.6618, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.466344118118286, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.5437, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4897878170013428, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6199, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.6578829288482666, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.5989, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.6173501014709473, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.6392, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.838721752166748, + "eval_runtime": 49.4357, + "eval_samples_per_second": 20.228, + "eval_steps_per_second": 0.85, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.5655064582824707, + "learning_rate": 2.095857026807474e-06, + "loss": 1.6276, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.685790777206421, + "learning_rate": 2.093826157595451e-06, + "loss": 1.6142, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.7278990745544434, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.6292, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.5769667625427246, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.6774, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.5120608806610107, + "learning_rate": 2.087733549959383e-06, + "loss": 1.5972, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.4658212661743164, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.5683, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.526524782180786, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.6687, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.3989250659942627, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.6184, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.589115858078003, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.6153, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.4985275268554688, + "learning_rate": 2.077579203899269e-06, + "loss": 1.5682, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.8374594449996948, + "eval_runtime": 47.0329, + "eval_samples_per_second": 21.262, + "eval_steps_per_second": 0.893, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5770835876464844, + "learning_rate": 2.075548334687246e-06, + "loss": 1.6437, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.5111560821533203, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.5711, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.6110589504241943, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.6125, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.6229705810546875, + "learning_rate": 2.069455727051178e-06, + "loss": 1.595, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.648716926574707, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.5908, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.673701763153076, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.6652, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.667478561401367, + "learning_rate": 2.06336311941511e-06, + "loss": 1.6088, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 3.5688586235046387, + "learning_rate": 2.061332250203087e-06, + "loss": 1.5852, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.512603282928467, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.6218, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.5296075344085693, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.6382, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.8406264781951904, + "eval_runtime": 43.8351, + "eval_samples_per_second": 22.813, + "eval_steps_per_second": 0.958, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.598994255065918, + "learning_rate": 2.055239642567019e-06, + "loss": 1.6263, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.93025279045105, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.6502, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 3.3655412197113037, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.5872, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.955120325088501, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.6404, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.5226638317108154, + "learning_rate": 2.047116165718928e-06, + "loss": 1.6102, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.5462093353271484, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.6757, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 4.066190242767334, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.6072, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.8548827171325684, + "learning_rate": 2.04102355808286e-06, + "loss": 1.5717, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.6206634044647217, + "learning_rate": 2.038992688870837e-06, + "loss": 1.5943, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.8793396949768066, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.6121, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.83925199508667, + "eval_runtime": 44.747, + "eval_samples_per_second": 22.348, + "eval_steps_per_second": 0.939, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.6215157508850098, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.6492, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.6103456020355225, + "learning_rate": 2.032900081234769e-06, + "loss": 1.6531, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.485775947570801, + "learning_rate": 2.030869212022746e-06, + "loss": 1.6086, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.399557113647461, + "learning_rate": 2.028838342810723e-06, + "loss": 1.6126, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.4711368083953857, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.5985, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.654580593109131, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.6557, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.5091657638549805, + "learning_rate": 2.022745735174655e-06, + "loss": 1.6181, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.5723490715026855, + "learning_rate": 2.020714865962632e-06, + "loss": 1.6421, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4211535453796387, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.6226, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.5550055503845215, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.5729, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.8397995233535767, + "eval_runtime": 44.4762, + "eval_samples_per_second": 22.484, + "eval_steps_per_second": 0.944, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.618758201599121, + "learning_rate": 2.014622258326564e-06, + "loss": 1.6356, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.6717913150787354, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.6246, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.5464835166931152, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.6027, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.557669162750244, + "learning_rate": 2.008529650690496e-06, + "loss": 1.6531, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.581970453262329, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.6076, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.609236478805542, + "learning_rate": 2.00446791226645e-06, + "loss": 1.5932, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.4249160289764404, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.6372, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.5093257427215576, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.6627, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.3989412784576416, + "learning_rate": 1.998375304630382e-06, + "loss": 1.6292, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4299519062042236, + "learning_rate": 1.996344435418359e-06, + "loss": 1.6385, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.8367400169372559, + "eval_runtime": 45.921, + "eval_samples_per_second": 21.777, + "eval_steps_per_second": 0.915, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.509530782699585, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.6441, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.591426134109497, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.6382, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.6569297313690186, + "learning_rate": 1.990251827782291e-06, + "loss": 1.6469, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.6453206539154053, + "learning_rate": 1.988220958570268e-06, + "loss": 1.6348, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.5522994995117188, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.6311, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.5981009006500244, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.585, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.617323637008667, + "learning_rate": 1.9821283509342e-06, + "loss": 1.584, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.6743686199188232, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.6334, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.418383836746216, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.5934, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.7952582836151123, + "learning_rate": 1.976035743298132e-06, + "loss": 1.6779, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.8379877805709839, + "eval_runtime": 46.545, + "eval_samples_per_second": 21.485, + "eval_steps_per_second": 0.902, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.5364768505096436, + "learning_rate": 1.974004874086109e-06, + "loss": 1.5802, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.446056365966797, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6141, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.527529239654541, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.6042, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.5953001976013184, + "learning_rate": 1.967912266450041e-06, + "loss": 1.5959, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.545994758605957, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.7091, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.3900482654571533, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.5946, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.51353120803833, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.6166, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.550389528274536, + "learning_rate": 1.95978878960195e-06, + "loss": 1.6817, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.6815762519836426, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.6594, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.591909885406494, + "learning_rate": 1.955727051177904e-06, + "loss": 1.5991, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.8390737771987915, + "eval_runtime": 46.8495, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.5111982822418213, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.5993, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.655891180038452, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.5915, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.471090316772461, + "learning_rate": 1.949634443541836e-06, + "loss": 1.5867, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.8599209785461426, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.6485, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4986093044281006, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.6084, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.6345789432525635, + "learning_rate": 1.943541835905768e-06, + "loss": 1.6719, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.5636799335479736, + "learning_rate": 1.941510966693745e-06, + "loss": 1.5817, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.5867199897766113, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.6498, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.6369481086730957, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.6791, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.4554147720336914, + "learning_rate": 1.935418359057677e-06, + "loss": 1.6319, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.8408559560775757, + "eval_runtime": 49.3515, + "eval_samples_per_second": 20.263, + "eval_steps_per_second": 0.851, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.5412509441375732, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.6673, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.5755672454833984, + "learning_rate": 1.931356620633631e-06, + "loss": 1.5636, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.602530002593994, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.6261, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.755247116088867, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.5817, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.528991460800171, + "learning_rate": 1.925264012997563e-06, + "loss": 1.6173, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.473057508468628, + "learning_rate": 1.92323314378554e-06, + "loss": 1.648, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.62630033493042, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.5864, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.7876687049865723, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.5804, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.491473436355591, + "learning_rate": 1.917140536149472e-06, + "loss": 1.5716, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.656989336013794, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.6595, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.839701771736145, + "eval_runtime": 44.5976, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.6800942420959473, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.6415, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.4725751876831055, + "learning_rate": 1.911047928513404e-06, + "loss": 1.5901, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.552624464035034, + "learning_rate": 1.909017059301381e-06, + "loss": 1.5956, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.6042184829711914, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.6328, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.5159223079681396, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.604, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.488041639328003, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.5539, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.659151077270508, + "learning_rate": 1.90089358245329e-06, + "loss": 1.5999, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.700378179550171, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.6485, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.3943607807159424, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.5736, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.8171095848083496, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6259, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.8379307985305786, + "eval_runtime": 48.6918, + "eval_samples_per_second": 20.537, + "eval_steps_per_second": 0.863, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.562655210494995, + "learning_rate": 1.892770105605199e-06, + "loss": 1.6439, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.5640337467193604, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.6113, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.5324809551239014, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.6331, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.393224000930786, + "learning_rate": 1.886677497969131e-06, + "loss": 1.5228, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 3.073945999145508, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.632, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.7066094875335693, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.5508, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.981933832168579, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6973, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.903444528579712, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.6036, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.8616178035736084, + "learning_rate": 1.876523151909017e-06, + "loss": 1.5723, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.8243541717529297, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.555, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.8437565565109253, + "eval_runtime": 48.5449, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.803142547607422, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.5861, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 3.146677017211914, + "learning_rate": 1.870430544272949e-06, + "loss": 1.6791, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 3.2554781436920166, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.5626, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.955395460128784, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.5893, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.941377878189087, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.5962, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.765620708465576, + "learning_rate": 1.862307067424858e-06, + "loss": 1.595, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.8529839515686035, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.6296, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.7770590782165527, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.5711, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.827406883239746, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.5966, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.927119016647339, + "learning_rate": 1.854183590576767e-06, + "loss": 1.6463, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.841707468032837, + "eval_runtime": 49.0965, + "eval_samples_per_second": 20.368, + "eval_steps_per_second": 0.855, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.8223392963409424, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.559, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.862473726272583, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.5189, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.9060091972351074, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.5631, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.8727498054504395, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6013, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.862298011779785, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.549, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.8221330642700195, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.6341, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.938336133956909, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.602, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.719489812850952, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.5626, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.6615874767303467, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.5535, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.835413932800293, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.5742, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.8375873565673828, + "eval_runtime": 45.731, + "eval_samples_per_second": 21.867, + "eval_steps_per_second": 0.918, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.773707389831543, + "learning_rate": 1.831844029244517e-06, + "loss": 1.5499, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 3.0959930419921875, + "learning_rate": 1.829813160032494e-06, + "loss": 1.5656, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.8521792888641357, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.5982, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.970618486404419, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.5731, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 3.0716512203216553, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.5824, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.8776299953460693, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.6283, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.72183895111084, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.5422, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.853234052658081, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.5533, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.7193424701690674, + "learning_rate": 1.815597075548335e-06, + "loss": 1.5923, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.9768450260162354, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.6277, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.8401803970336914, + "eval_runtime": 45.8062, + "eval_samples_per_second": 21.831, + "eval_steps_per_second": 0.917, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.752772569656372, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.5241, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.784980297088623, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.5657, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.7485857009887695, + "learning_rate": 1.807473598700244e-06, + "loss": 1.5501, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.7191033363342285, + "learning_rate": 1.805442729488221e-06, + "loss": 1.5718, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.697540044784546, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.5804, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.7175710201263428, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.5809, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.764824151992798, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.545, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.946669816970825, + "learning_rate": 1.79731925264013e-06, + "loss": 1.595, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.689326524734497, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.5676, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.759997606277466, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.6032, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.8379274606704712, + "eval_runtime": 45.5941, + "eval_samples_per_second": 21.933, + "eval_steps_per_second": 0.921, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.8656060695648193, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.6086, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.7155873775482178, + "learning_rate": 1.789195775792039e-06, + "loss": 1.5218, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.8487389087677, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6285, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.8070144653320312, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.5479, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.856282949447632, + "learning_rate": 1.783103168155971e-06, + "loss": 1.5689, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.7465922832489014, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.5733, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.824997901916504, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.5583, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.8030786514282227, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.5383, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.911377429962158, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.586, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.7452991008758545, + "learning_rate": 1.772948822095857e-06, + "loss": 1.6409, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.8399248123168945, + "eval_runtime": 46.1241, + "eval_samples_per_second": 21.681, + "eval_steps_per_second": 0.911, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.8255481719970703, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.6134, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.865100622177124, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.5791, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.740133047103882, + "learning_rate": 1.766856214459789e-06, + "loss": 1.5651, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.7219033241271973, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.5453, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.7266135215759277, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.6375, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.8476688861846924, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.5877, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.810105085372925, + "learning_rate": 1.758732737611698e-06, + "loss": 1.5796, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.78798246383667, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.5708, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.750525951385498, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.5567, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.818673849105835, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.5661, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.8374614715576172, + "eval_runtime": 43.5618, + "eval_samples_per_second": 22.956, + "eval_steps_per_second": 0.964, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.7967586517333984, + "learning_rate": 1.750609260763607e-06, + "loss": 1.5846, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.766272783279419, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.5721, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.8006162643432617, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.5261, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 3.714587688446045, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.6599, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 3.143052339553833, + "learning_rate": 1.742485783915516e-06, + "loss": 1.659, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.8256232738494873, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.5076, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.8843302726745605, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.5862, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 6.692182540893555, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.6144, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.928053617477417, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.571, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.8071024417877197, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.5861, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.837381362915039, + "eval_runtime": 44.7089, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1900 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 5.922887877787648e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-1900/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/rng_state_0.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..e39746492b970dcc6a87f52bb29c56fc5ba161fc --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ce9e99fc568ea7d69c95f8da5622cf2e24538055df1accfbeaccff9c07072311 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/rng_state_1.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..abccc0fe9164aa3960835284734191dc786ae981 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cf507f5973fce5bbcb6429ce7c7164e2e033221c79824bee0dcc77248e530ab9 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..6fc0da9c514e1a86118ed6e753db92d26ba258c9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3bd2607e58c5b6d795789aee02c9a0a2d71e807483894bf38b21ea6bacd6acfa +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/scheduler.pt b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..c7720f79f77e472101504db0a1d78506e26ff573 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:61286e2ca15e31a69f9ce385a984f8bdcd5c89d5ad65e5bcef3cb297b4937542 +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/tokenizer.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-200/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/generation_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/rng_state_1.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..b32943dd52f24c79c4a09bfcf59fbfbb94eea725 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0fe86e559cde3904e483c35c110a6ec5956237290f913253e911d4bf45d20674 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..028025132ce3c773ea373f1d990200a8d2f08a37 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c41d639def0ca99b5b4aedc81d4d3f65c59ee96f4fe0b4cb39a3f8d1ba5d4c0e +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/tokenizer.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..e6b55f715b0049dc7388516b17fed8d9382d50b4 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/trainer_state.json @@ -0,0 +1,15633 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.191780821917808, + "eval_steps": 10, + "global_step": 2000, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.5818793773651123, + "learning_rate": 4.569455727051178e-06, + "loss": 1.85, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 3.0734691619873047, + "learning_rate": 4.567424857839156e-06, + "loss": 1.8682, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.496253252029419, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9007, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.360891342163086, + "learning_rate": 4.5633631194151105e-06, + "loss": 1.8262, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.5731117725372314, + "learning_rate": 4.561332250203087e-06, + "loss": 1.8369, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.700558662414551, + "learning_rate": 4.559301380991065e-06, + "loss": 1.7825, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.561306953430176, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.8174, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.618889570236206, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.8812, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.4671308994293213, + "learning_rate": 4.553208773354996e-06, + "loss": 1.7959, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.4571878910064697, + "learning_rate": 4.551177904142974e-06, + "loss": 1.8943, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.010777711868286, + "eval_runtime": 43.7424, + "eval_samples_per_second": 22.861, + "eval_steps_per_second": 0.96, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.4786667823791504, + "learning_rate": 4.5491470349309505e-06, + "loss": 1.8147, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.2955000400543213, + "learning_rate": 4.547116165718929e-06, + "loss": 1.8547, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.3873913288116455, + "learning_rate": 4.545085296506905e-06, + "loss": 1.8156, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.2449734210968018, + "learning_rate": 4.543054427294882e-06, + "loss": 1.8053, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.3969857692718506, + "learning_rate": 4.54102355808286e-06, + "loss": 1.8085, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.3446357250213623, + "learning_rate": 4.538992688870837e-06, + "loss": 1.8355, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.408762216567993, + "learning_rate": 4.536961819658814e-06, + "loss": 1.8319, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.362962245941162, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.8422, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.402984857559204, + "learning_rate": 4.532900081234769e-06, + "loss": 1.8125, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.3964452743530273, + "learning_rate": 4.530869212022746e-06, + "loss": 1.8417, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.006908416748047, + "eval_runtime": 44.2701, + "eval_samples_per_second": 22.589, + "eval_steps_per_second": 0.949, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.595233201980591, + "learning_rate": 4.528838342810723e-06, + "loss": 1.8275, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.326117753982544, + "learning_rate": 4.526807473598701e-06, + "loss": 1.8405, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.4904465675354004, + "learning_rate": 4.524776604386678e-06, + "loss": 1.8172, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.4591610431671143, + "learning_rate": 4.522745735174655e-06, + "loss": 1.876, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.319800853729248, + "learning_rate": 4.520714865962632e-06, + "loss": 1.8294, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.28790020942688, + "learning_rate": 4.51868399675061e-06, + "loss": 1.8537, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.284822702407837, + "learning_rate": 4.516653127538587e-06, + "loss": 1.8087, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.273820638656616, + "learning_rate": 4.514622258326564e-06, + "loss": 1.7992, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.2025654315948486, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.8248, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 2.3218843936920166, + "learning_rate": 4.510560519902519e-06, + "loss": 1.8647, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.0048956871032715, + "eval_runtime": 44.5972, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.515766143798828, + "learning_rate": 4.508529650690496e-06, + "loss": 1.7673, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.4376487731933594, + "learning_rate": 4.506498781478473e-06, + "loss": 1.7755, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.2801272869110107, + "learning_rate": 4.504467912266451e-06, + "loss": 1.8273, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.4707891941070557, + "learning_rate": 4.502437043054428e-06, + "loss": 1.8971, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.301301956176758, + "learning_rate": 4.500406173842405e-06, + "loss": 1.8345, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.2317373752593994, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.8248, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.447373867034912, + "learning_rate": 4.49634443541836e-06, + "loss": 1.8148, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.337245225906372, + "learning_rate": 4.494313566206336e-06, + "loss": 1.8811, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3717031478881836, + "learning_rate": 4.492282696994314e-06, + "loss": 1.8311, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.287677526473999, + "learning_rate": 4.490251827782291e-06, + "loss": 1.7728, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.004889965057373, + "eval_runtime": 43.8132, + "eval_samples_per_second": 22.824, + "eval_steps_per_second": 0.959, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.441445827484131, + "learning_rate": 4.488220958570269e-06, + "loss": 1.7927, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.3191535472869873, + "learning_rate": 4.486190089358245e-06, + "loss": 1.759, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.3118820190429688, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8509, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.4711859226226807, + "learning_rate": 4.4821283509342e-06, + "loss": 1.8402, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.3142173290252686, + "learning_rate": 4.480097481722178e-06, + "loss": 1.8238, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.2665774822235107, + "learning_rate": 4.478066612510154e-06, + "loss": 1.7826, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.1692490577697754, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.8138, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.412783145904541, + "learning_rate": 4.474004874086109e-06, + "loss": 1.9421, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3692686557769775, + "learning_rate": 4.471974004874087e-06, + "loss": 1.7722, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.3224151134490967, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.8753, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.0016305446624756, + "eval_runtime": 45.564, + "eval_samples_per_second": 21.947, + "eval_steps_per_second": 0.922, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.51043438911438, + "learning_rate": 4.467912266450041e-06, + "loss": 1.8325, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.4568657875061035, + "learning_rate": 4.465881397238018e-06, + "loss": 1.8081, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.3279001712799072, + "learning_rate": 4.463850528025995e-06, + "loss": 1.8439, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.6041171550750732, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.8623, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.3780531883239746, + "learning_rate": 4.45978878960195e-06, + "loss": 1.8269, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.511500597000122, + "learning_rate": 4.457757920389927e-06, + "loss": 1.7749, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.707789659500122, + "learning_rate": 4.455727051177904e-06, + "loss": 1.8157, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.5495495796203613, + "learning_rate": 4.453696181965882e-06, + "loss": 1.8534, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.255876302719116, + "learning_rate": 4.451665312753859e-06, + "loss": 1.8377, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.415097951889038, + "learning_rate": 4.449634443541836e-06, + "loss": 1.8764, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 1.9943474531173706, + "eval_runtime": 42.2985, + "eval_samples_per_second": 23.642, + "eval_steps_per_second": 0.993, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.2604713439941406, + "learning_rate": 4.4476035743298135e-06, + "loss": 1.7898, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.5643842220306396, + "learning_rate": 4.445572705117791e-06, + "loss": 1.7953, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.412548780441284, + "learning_rate": 4.443541835905768e-06, + "loss": 1.8611, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.5423877239227295, + "learning_rate": 4.441510966693745e-06, + "loss": 1.8049, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.4986631870269775, + "learning_rate": 4.439480097481723e-06, + "loss": 1.8759, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.8349273204803467, + "learning_rate": 4.4374492282697e-06, + "loss": 1.7416, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.452017068862915, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8268, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.6425273418426514, + "learning_rate": 4.433387489845654e-06, + "loss": 1.8334, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.328740358352661, + "learning_rate": 4.431356620633632e-06, + "loss": 1.7931, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.262209415435791, + "learning_rate": 4.429325751421609e-06, + "loss": 1.8151, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 1.9917123317718506, + "eval_runtime": 42.4309, + "eval_samples_per_second": 23.568, + "eval_steps_per_second": 0.99, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.2984049320220947, + "learning_rate": 4.427294882209586e-06, + "loss": 1.8187, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.3910019397735596, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.7512, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.4494471549987793, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8202, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.3857171535491943, + "learning_rate": 4.421202274573518e-06, + "loss": 1.8279, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.4526500701904297, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.7136, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.2216501235961914, + "learning_rate": 4.417140536149473e-06, + "loss": 1.7574, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.7639758586883545, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8403, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.3177521228790283, + "learning_rate": 4.413078797725427e-06, + "loss": 1.7898, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.527902364730835, + "learning_rate": 4.411047928513404e-06, + "loss": 1.8184, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.34061336517334, + "learning_rate": 4.409017059301382e-06, + "loss": 1.811, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 1.9882216453552246, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.6462934017181396, + "learning_rate": 4.406986190089358e-06, + "loss": 1.8501, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.2725913524627686, + "learning_rate": 4.404955320877336e-06, + "loss": 1.8262, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.390084981918335, + "learning_rate": 4.402924451665313e-06, + "loss": 1.7842, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.314305305480957, + "learning_rate": 4.400893582453291e-06, + "loss": 1.7692, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.2828257083892822, + "learning_rate": 4.398862713241267e-06, + "loss": 1.8031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.3493599891662598, + "learning_rate": 4.396831844029245e-06, + "loss": 1.8183, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.423365831375122, + "learning_rate": 4.394800974817222e-06, + "loss": 1.8012, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.438015937805176, + "learning_rate": 4.392770105605199e-06, + "loss": 1.8217, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.357172727584839, + "learning_rate": 4.390739236393176e-06, + "loss": 1.8382, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.3383474349975586, + "learning_rate": 4.388708367181154e-06, + "loss": 1.841, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 1.9878787994384766, + "eval_runtime": 44.0694, + "eval_samples_per_second": 22.691, + "eval_steps_per_second": 0.953, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.459385633468628, + "learning_rate": 4.386677497969131e-06, + "loss": 1.8377, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.356788158416748, + "learning_rate": 4.384646628757108e-06, + "loss": 1.8718, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.3217222690582275, + "learning_rate": 4.3826157595450855e-06, + "loss": 1.8073, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.3134331703186035, + "learning_rate": 4.380584890333063e-06, + "loss": 1.7675, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.3273653984069824, + "learning_rate": 4.37855402112104e-06, + "loss": 1.7169, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.611034393310547, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8702, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.2077300548553467, + "learning_rate": 4.3744922826969946e-06, + "loss": 1.7951, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.3393311500549316, + "learning_rate": 4.372461413484972e-06, + "loss": 1.8136, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.3306660652160645, + "learning_rate": 4.370430544272949e-06, + "loss": 1.8372, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.4440250396728516, + "learning_rate": 4.368399675060926e-06, + "loss": 1.844, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 1.98439621925354, + "eval_runtime": 44.4595, + "eval_samples_per_second": 22.492, + "eval_steps_per_second": 0.945, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.205542802810669, + "learning_rate": 4.366368805848904e-06, + "loss": 1.8182, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.6030094623565674, + "learning_rate": 4.364337936636881e-06, + "loss": 1.7524, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.4000444412231445, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8269, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.319526433944702, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.7722, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.1624808311462402, + "learning_rate": 4.358245329000813e-06, + "loss": 1.8081, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.26615309715271, + "learning_rate": 4.35621445978879e-06, + "loss": 1.8049, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.4895036220550537, + "learning_rate": 4.354183590576767e-06, + "loss": 1.7932, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.437161445617676, + "learning_rate": 4.352152721364745e-06, + "loss": 1.772, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.344069480895996, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8213, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.360734224319458, + "learning_rate": 4.348090982940699e-06, + "loss": 1.8268, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 1.9846479892730713, + "eval_runtime": 44.947, + "eval_samples_per_second": 22.248, + "eval_steps_per_second": 0.934, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.59487247467041, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.8303, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.6512107849121094, + "learning_rate": 4.344029244516654e-06, + "loss": 1.7979, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.2900476455688477, + "learning_rate": 4.341998375304631e-06, + "loss": 1.8707, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.409494638442993, + "learning_rate": 4.339967506092607e-06, + "loss": 1.8075, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.527205467224121, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.7884, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.33461332321167, + "learning_rate": 4.335905767668562e-06, + "loss": 1.7839, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.3555047512054443, + "learning_rate": 4.33387489845654e-06, + "loss": 1.8245, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.397148609161377, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.8115, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.232139825820923, + "learning_rate": 4.329813160032495e-06, + "loss": 1.7747, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.1684389114379883, + "learning_rate": 4.327782290820471e-06, + "loss": 1.748, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 1.980297565460205, + "eval_runtime": 42.9183, + "eval_samples_per_second": 23.3, + "eval_steps_per_second": 0.979, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.2454638481140137, + "learning_rate": 4.325751421608449e-06, + "loss": 1.83, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.3264827728271484, + "learning_rate": 4.323720552396426e-06, + "loss": 1.7856, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.298804998397827, + "learning_rate": 4.321689683184404e-06, + "loss": 1.78, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2202727794647217, + "learning_rate": 4.31965881397238e-06, + "loss": 1.7875, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.494479179382324, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.8363, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3520524501800537, + "learning_rate": 4.315597075548335e-06, + "loss": 1.8266, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.373878002166748, + "learning_rate": 4.313566206336312e-06, + "loss": 1.8348, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.4232168197631836, + "learning_rate": 4.311535337124289e-06, + "loss": 1.77, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.2572121620178223, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.7914, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.5817952156066895, + "learning_rate": 4.307473598700244e-06, + "loss": 1.8099, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 1.9784427881240845, + "eval_runtime": 43.2981, + "eval_samples_per_second": 23.096, + "eval_steps_per_second": 0.97, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.646040916442871, + "learning_rate": 4.305442729488221e-06, + "loss": 1.7452, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.370220184326172, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8399, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.4599130153656006, + "learning_rate": 4.301380991064176e-06, + "loss": 1.8273, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 2.4135093688964844, + "learning_rate": 4.299350121852153e-06, + "loss": 1.852, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.371021270751953, + "learning_rate": 4.29731925264013e-06, + "loss": 1.8339, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.6858229637145996, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.792, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.4101266860961914, + "learning_rate": 4.293257514216085e-06, + "loss": 1.7556, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.769761800765991, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8223, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.5777804851531982, + "learning_rate": 4.289195775792039e-06, + "loss": 1.8406, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.376864433288574, + "learning_rate": 4.287164906580017e-06, + "loss": 1.7846, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 1.9758412837982178, + "eval_runtime": 43.9353, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.7461960315704346, + "learning_rate": 4.285134037367994e-06, + "loss": 1.776, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.549576759338379, + "learning_rate": 4.283103168155971e-06, + "loss": 1.7644, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.568074941635132, + "learning_rate": 4.281072298943948e-06, + "loss": 1.7571, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.8195722103118896, + "learning_rate": 4.279041429731926e-06, + "loss": 1.7959, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.471623182296753, + "learning_rate": 4.277010560519903e-06, + "loss": 1.7584, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.664689302444458, + "learning_rate": 4.27497969130788e-06, + "loss": 1.7909, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.684504985809326, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.8205, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.5247528553009033, + "learning_rate": 4.270917952883835e-06, + "loss": 1.8365, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.757972478866577, + "learning_rate": 4.268887083671812e-06, + "loss": 1.7665, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.6032614707946777, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8702, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 1.9716291427612305, + "eval_runtime": 43.5482, + "eval_samples_per_second": 22.963, + "eval_steps_per_second": 0.964, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.5324370861053467, + "learning_rate": 4.264825345247766e-06, + "loss": 1.738, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.730712890625, + "learning_rate": 4.262794476035744e-06, + "loss": 1.7918, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.5554966926574707, + "learning_rate": 4.26076360682372e-06, + "loss": 1.7218, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.4774436950683594, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.7823, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.77431058883667, + "learning_rate": 4.256701868399675e-06, + "loss": 1.794, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.258828639984131, + "learning_rate": 4.254670999187653e-06, + "loss": 1.805, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.384399175643921, + "learning_rate": 4.252640129975629e-06, + "loss": 1.818, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3278932571411133, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.7719, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.207348108291626, + "learning_rate": 4.248578391551584e-06, + "loss": 1.7362, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.4525978565216064, + "learning_rate": 4.246547522339562e-06, + "loss": 1.778, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 1.9694374799728394, + "eval_runtime": 42.5977, + "eval_samples_per_second": 23.475, + "eval_steps_per_second": 0.986, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.377835273742676, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.7926, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.4623866081237793, + "learning_rate": 4.242485783915517e-06, + "loss": 1.7366, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.542219400405884, + "learning_rate": 4.240454914703493e-06, + "loss": 1.7409, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.1955819129943848, + "learning_rate": 4.23842404549147e-06, + "loss": 1.7596, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.5834269523620605, + "learning_rate": 4.236393176279448e-06, + "loss": 1.7343, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.219660520553589, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8089, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.283200740814209, + "learning_rate": 4.232331437855402e-06, + "loss": 1.8248, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.274127244949341, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.825, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.3550686836242676, + "learning_rate": 4.228269699431357e-06, + "loss": 1.7957, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.588027238845825, + "learning_rate": 4.226238830219334e-06, + "loss": 1.7865, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 1.9662212133407593, + "eval_runtime": 43.2661, + "eval_samples_per_second": 23.113, + "eval_steps_per_second": 0.971, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.2445170879364014, + "learning_rate": 4.224207961007311e-06, + "loss": 1.791, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.399174690246582, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.7944, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.586184501647949, + "learning_rate": 4.220146222583266e-06, + "loss": 1.7856, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.4174695014953613, + "learning_rate": 4.218115353371243e-06, + "loss": 1.7842, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2684640884399414, + "learning_rate": 4.21608448415922e-06, + "loss": 1.7482, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 3.1782288551330566, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8286, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.4046335220336914, + "learning_rate": 4.212022745735175e-06, + "loss": 1.7228, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 3.3722083568573, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8044, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.293466329574585, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.7415, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.28663969039917, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8282, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 1.9627619981765747, + "eval_runtime": 43.6092, + "eval_samples_per_second": 22.931, + "eval_steps_per_second": 0.963, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.255870819091797, + "learning_rate": 4.203899268887084e-06, + "loss": 1.7983, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.4964773654937744, + "learning_rate": 4.201868399675061e-06, + "loss": 1.8141, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.3372740745544434, + "learning_rate": 4.199837530463039e-06, + "loss": 1.7142, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.3506968021392822, + "learning_rate": 4.197806661251016e-06, + "loss": 1.7898, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.242410898208618, + "learning_rate": 4.195775792038993e-06, + "loss": 1.7677, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.3864502906799316, + "learning_rate": 4.19374492282697e-06, + "loss": 1.7097, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.482349157333374, + "learning_rate": 4.191714053614948e-06, + "loss": 1.8052, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.317652940750122, + "learning_rate": 4.189683184402924e-06, + "loss": 1.7692, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.2777392864227295, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8246, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.3506035804748535, + "learning_rate": 4.185621445978879e-06, + "loss": 1.7576, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 1.9610412120819092, + "eval_runtime": 43.3119, + "eval_samples_per_second": 23.088, + "eval_steps_per_second": 0.97, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.3011536598205566, + "learning_rate": 4.183590576766857e-06, + "loss": 1.7891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.2614853382110596, + "learning_rate": 4.181559707554833e-06, + "loss": 1.8258, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.219175100326538, + "learning_rate": 4.179528838342811e-06, + "loss": 1.7494, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.4239423274993896, + "learning_rate": 4.177497969130788e-06, + "loss": 1.7611, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.470712423324585, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8281, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.504060745239258, + "learning_rate": 4.173436230706742e-06, + "loss": 1.7641, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.326601982116699, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.7671, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.4280030727386475, + "learning_rate": 4.169374492282697e-06, + "loss": 1.7573, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.455716371536255, + "learning_rate": 4.167343623070675e-06, + "loss": 1.809, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.3029844760894775, + "learning_rate": 4.165312753858652e-06, + "loss": 1.77, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 1.9604709148406982, + "eval_runtime": 44.4627, + "eval_samples_per_second": 22.491, + "eval_steps_per_second": 0.945, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.3833982944488525, + "learning_rate": 4.163281884646629e-06, + "loss": 1.7652, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.2392966747283936, + "learning_rate": 4.161251015434607e-06, + "loss": 1.7904, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.4546353816986084, + "learning_rate": 4.159220146222583e-06, + "loss": 1.7324, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.13592529296875, + "learning_rate": 4.157189277010561e-06, + "loss": 1.759, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.381657361984253, + "learning_rate": 4.155158407798538e-06, + "loss": 1.7871, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.4142940044403076, + "learning_rate": 4.153127538586516e-06, + "loss": 1.7832, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.305697441101074, + "learning_rate": 4.151096669374492e-06, + "loss": 1.7619, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.193429708480835, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.7738, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.3007619380950928, + "learning_rate": 4.147034930950447e-06, + "loss": 1.7839, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.6337807178497314, + "learning_rate": 4.145004061738425e-06, + "loss": 1.7748, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 1.9597856998443604, + "eval_runtime": 43.8644, + "eval_samples_per_second": 22.798, + "eval_steps_per_second": 0.957, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.4294793605804443, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.7001, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.512155532836914, + "learning_rate": 4.140942323314379e-06, + "loss": 1.7334, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.290808916091919, + "learning_rate": 4.138911454102356e-06, + "loss": 1.7522, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.483161211013794, + "learning_rate": 4.136880584890333e-06, + "loss": 1.798, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2973921298980713, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.7847, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.253084659576416, + "learning_rate": 4.132818846466288e-06, + "loss": 1.7812, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.306342840194702, + "learning_rate": 4.130787977254265e-06, + "loss": 1.7886, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.5153164863586426, + "learning_rate": 4.128757108042242e-06, + "loss": 1.7954, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.4122278690338135, + "learning_rate": 4.12672623883022e-06, + "loss": 1.8525, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.4520716667175293, + "learning_rate": 4.124695369618197e-06, + "loss": 1.7648, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 1.9547661542892456, + "eval_runtime": 45.3774, + "eval_samples_per_second": 22.037, + "eval_steps_per_second": 0.926, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2959611415863037, + "learning_rate": 4.122664500406174e-06, + "loss": 1.7626, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.2280471324920654, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.7954, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.3870065212249756, + "learning_rate": 4.118602761982129e-06, + "loss": 1.7377, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.292612075805664, + "learning_rate": 4.116571892770106e-06, + "loss": 1.7, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.4607021808624268, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8338, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.532824993133545, + "learning_rate": 4.112510154346061e-06, + "loss": 1.7549, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2139623165130615, + "learning_rate": 4.110479285134038e-06, + "loss": 1.78, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.4749462604522705, + "learning_rate": 4.108448415922015e-06, + "loss": 1.8034, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.609265089035034, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.7941, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.367753028869629, + "learning_rate": 4.10438667749797e-06, + "loss": 1.6653, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 1.9539440870285034, + "eval_runtime": 48.8438, + "eval_samples_per_second": 20.473, + "eval_steps_per_second": 0.86, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.9723567962646484, + "learning_rate": 4.102355808285947e-06, + "loss": 1.8317, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.388806104660034, + "learning_rate": 4.100324939073924e-06, + "loss": 1.7597, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4882986545562744, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.7708, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.7586395740509033, + "learning_rate": 4.096263200649879e-06, + "loss": 1.7804, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.237025260925293, + "learning_rate": 4.094232331437856e-06, + "loss": 1.7415, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.421330213546753, + "learning_rate": 4.092201462225833e-06, + "loss": 1.734, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.7281360626220703, + "learning_rate": 4.090170593013811e-06, + "loss": 1.7716, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.2200379371643066, + "learning_rate": 4.088139723801787e-06, + "loss": 1.7266, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.465440034866333, + "learning_rate": 4.086108854589765e-06, + "loss": 1.741, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.39835262298584, + "learning_rate": 4.084077985377742e-06, + "loss": 1.757, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 1.9514222145080566, + "eval_runtime": 52.8068, + "eval_samples_per_second": 18.937, + "eval_steps_per_second": 0.795, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.526684522628784, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8195, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.341954469680786, + "learning_rate": 4.080016246953696e-06, + "loss": 1.7196, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.3755977153778076, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8115, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.3952431678771973, + "learning_rate": 4.075954508529651e-06, + "loss": 1.812, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.554089069366455, + "learning_rate": 4.073923639317629e-06, + "loss": 1.8184, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.2319302558898926, + "learning_rate": 4.071892770105605e-06, + "loss": 1.7448, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.284856081008911, + "learning_rate": 4.069861900893583e-06, + "loss": 1.7154, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.4081835746765137, + "learning_rate": 4.06783103168156e-06, + "loss": 1.7311, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.6115968227386475, + "learning_rate": 4.065800162469538e-06, + "loss": 1.7436, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.368004322052002, + "learning_rate": 4.063769293257514e-06, + "loss": 1.7811, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 1.9508955478668213, + "eval_runtime": 46.8711, + "eval_samples_per_second": 21.335, + "eval_steps_per_second": 0.896, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.4607419967651367, + "learning_rate": 4.061738424045492e-06, + "loss": 1.7563, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.2978763580322266, + "learning_rate": 4.059707554833469e-06, + "loss": 1.7755, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.3316237926483154, + "learning_rate": 4.057676685621446e-06, + "loss": 1.8226, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.3699169158935547, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.7755, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.195596218109131, + "learning_rate": 4.053614947197401e-06, + "loss": 1.6916, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.4524662494659424, + "learning_rate": 4.051584077985378e-06, + "loss": 1.774, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.573888063430786, + "learning_rate": 4.049553208773355e-06, + "loss": 1.7953, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.5157253742218018, + "learning_rate": 4.047522339561333e-06, + "loss": 1.7597, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.5588185787200928, + "learning_rate": 4.04549147034931e-06, + "loss": 1.7553, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.8950047492980957, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8647, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 1.945882797241211, + "eval_runtime": 45.481, + "eval_samples_per_second": 21.987, + "eval_steps_per_second": 0.923, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.217116117477417, + "learning_rate": 4.041429731925264e-06, + "loss": 1.7465, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.5601565837860107, + "learning_rate": 4.039398862713242e-06, + "loss": 1.7575, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.4540457725524902, + "learning_rate": 4.037367993501219e-06, + "loss": 1.8291, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.327181100845337, + "learning_rate": 4.035337124289196e-06, + "loss": 1.7409, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.485457181930542, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.7192, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.4613442420959473, + "learning_rate": 4.031275385865151e-06, + "loss": 1.7613, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.303217649459839, + "learning_rate": 4.029244516653128e-06, + "loss": 1.7799, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.3830621242523193, + "learning_rate": 4.027213647441105e-06, + "loss": 1.7045, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.342332601547241, + "learning_rate": 4.025182778229083e-06, + "loss": 1.7246, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.4014711380004883, + "learning_rate": 4.02315190901706e-06, + "loss": 1.7621, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 1.95199453830719, + "eval_runtime": 45.4754, + "eval_samples_per_second": 21.99, + "eval_steps_per_second": 0.924, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.6015400886535645, + "learning_rate": 4.021121039805037e-06, + "loss": 1.7536, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.866978406906128, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8405, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.389810085296631, + "learning_rate": 4.017059301380992e-06, + "loss": 1.7365, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.280442953109741, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8273, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.509330987930298, + "learning_rate": 4.012997562956945e-06, + "loss": 1.7127, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.452078104019165, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.7457, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.368337631225586, + "learning_rate": 4.0089358245329e-06, + "loss": 1.7412, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.4529192447662354, + "learning_rate": 4.006904955320878e-06, + "loss": 1.7434, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.647761821746826, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8013, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.387424945831299, + "learning_rate": 4.002843216896833e-06, + "loss": 1.7948, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 1.946029543876648, + "eval_runtime": 43.9666, + "eval_samples_per_second": 22.745, + "eval_steps_per_second": 0.955, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.9217846393585205, + "learning_rate": 4.000812347684809e-06, + "loss": 1.7384, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.400040864944458, + "learning_rate": 3.998781478472787e-06, + "loss": 1.7496, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.4050440788269043, + "learning_rate": 3.996750609260764e-06, + "loss": 1.7165, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 3.6001529693603516, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8679, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 2.4040098190307617, + "learning_rate": 3.992688870836718e-06, + "loss": 1.7671, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.5645527839660645, + "learning_rate": 3.990658001624696e-06, + "loss": 1.7374, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.799394130706787, + "learning_rate": 3.988627132412673e-06, + "loss": 1.796, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.5481607913970947, + "learning_rate": 3.98659626320065e-06, + "loss": 1.753, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.8851330280303955, + "learning_rate": 3.984565393988627e-06, + "loss": 1.7505, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.6668694019317627, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.7604, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 1.939759612083435, + "eval_runtime": 44.5019, + "eval_samples_per_second": 22.471, + "eval_steps_per_second": 0.944, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.320935010910034, + "learning_rate": 3.980503655564582e-06, + "loss": 1.7749, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.829897880554199, + "learning_rate": 3.978472786352559e-06, + "loss": 1.7763, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.8420908451080322, + "learning_rate": 3.976441917140536e-06, + "loss": 1.7598, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.8183, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.781, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 3.9271891117095947, + "learning_rate": 3.972380178716491e-06, + "loss": 1.7949, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.277897834777832, + "learning_rate": 3.970349309504468e-06, + "loss": 1.707, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 3.241063117980957, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.7175, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.514458417892456, + "learning_rate": 3.966287571080423e-06, + "loss": 1.7764, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.420212745666504, + "learning_rate": 3.9642567018684e-06, + "loss": 1.7618, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 1.9401386976242065, + "eval_runtime": 44.1672, + "eval_samples_per_second": 22.641, + "eval_steps_per_second": 0.951, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.3980538845062256, + "learning_rate": 3.962225832656377e-06, + "loss": 1.6922, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.892646312713623, + "learning_rate": 3.960194963444355e-06, + "loss": 1.7571, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.499260902404785, + "learning_rate": 3.958164094232332e-06, + "loss": 1.7501, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.409695625305176, + "learning_rate": 3.956133225020309e-06, + "loss": 1.7553, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.640069007873535, + "learning_rate": 3.954102355808286e-06, + "loss": 1.7526, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.9101760387420654, + "learning_rate": 3.952071486596264e-06, + "loss": 1.7483, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.2579474449157715, + "learning_rate": 3.950040617384241e-06, + "loss": 1.7379, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 4.039298057556152, + "learning_rate": 3.948009748172218e-06, + "loss": 1.7988, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.3232030868530273, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.7592, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.697164297103882, + "learning_rate": 3.943948009748173e-06, + "loss": 1.7622, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 1.9367746114730835, + "eval_runtime": 43.7013, + "eval_samples_per_second": 22.883, + "eval_steps_per_second": 0.961, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.480297803878784, + "learning_rate": 3.94191714053615e-06, + "loss": 1.8006, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.3624603748321533, + "learning_rate": 3.939886271324127e-06, + "loss": 1.7152, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.417187213897705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.7077, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.6575164794921875, + "learning_rate": 3.935824532900082e-06, + "loss": 1.7744, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.4896557331085205, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8052, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.475693941116333, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.7008, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.3210372924804688, + "learning_rate": 3.929731925264013e-06, + "loss": 1.7339, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.433785915374756, + "learning_rate": 3.927701056051991e-06, + "loss": 1.7544, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 3.181574583053589, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.7229, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.704399347305298, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.7086, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 1.9335075616836548, + "eval_runtime": 47.0248, + "eval_samples_per_second": 21.265, + "eval_steps_per_second": 0.893, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 3.5027623176574707, + "learning_rate": 3.921608448415922e-06, + "loss": 1.7622, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.450943946838379, + "learning_rate": 3.9195775792039e-06, + "loss": 1.7286, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.4603374004364014, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.7725, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.8217649459838867, + "learning_rate": 3.915515840779855e-06, + "loss": 1.7859, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.3308372497558594, + "learning_rate": 3.913484971567831e-06, + "loss": 1.6874, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.624776840209961, + "learning_rate": 3.911454102355808e-06, + "loss": 1.7632, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.4097838401794434, + "learning_rate": 3.909423233143786e-06, + "loss": 1.7576, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.487410545349121, + "learning_rate": 3.907392363931763e-06, + "loss": 1.831, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.647245168685913, + "learning_rate": 3.90536149471974e-06, + "loss": 1.8448, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.397664785385132, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.868, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 1.930379033088684, + "eval_runtime": 51.1662, + "eval_samples_per_second": 19.544, + "eval_steps_per_second": 0.821, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.277433395385742, + "learning_rate": 3.901299756295695e-06, + "loss": 1.7394, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 3.1312496662139893, + "learning_rate": 3.899268887083672e-06, + "loss": 1.7422, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.5011534690856934, + "learning_rate": 3.897238017871649e-06, + "loss": 1.7484, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.8716166019439697, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.7253, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.4346914291381836, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8197, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.348475933074951, + "learning_rate": 3.891145410235581e-06, + "loss": 1.7737, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.3996007442474365, + "learning_rate": 3.889114541023558e-06, + "loss": 1.7333, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.355262279510498, + "learning_rate": 3.887083671811536e-06, + "loss": 1.6902, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.335932731628418, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8206, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.2893190383911133, + "learning_rate": 3.88302193338749e-06, + "loss": 1.7533, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 1.9290764331817627, + "eval_runtime": 46.9373, + "eval_samples_per_second": 21.305, + "eval_steps_per_second": 0.895, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.739842176437378, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.7318, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.587916374206543, + "learning_rate": 3.878960194963445e-06, + "loss": 1.7979, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.406548500061035, + "learning_rate": 3.876929325751422e-06, + "loss": 1.6899, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.440321922302246, + "learning_rate": 3.874898456539399e-06, + "loss": 1.7546, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.320030450820923, + "learning_rate": 3.872867587327377e-06, + "loss": 1.7254, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.3795316219329834, + "learning_rate": 3.870836718115354e-06, + "loss": 1.728, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.299802541732788, + "learning_rate": 3.868805848903331e-06, + "loss": 1.7464, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.469719409942627, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.765, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.612109899520874, + "learning_rate": 3.864744110479286e-06, + "loss": 1.7827, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.623173952102661, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8019, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 1.9268505573272705, + "eval_runtime": 46.77, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.4146616458892822, + "learning_rate": 3.86068237205524e-06, + "loss": 1.7498, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.361994743347168, + "learning_rate": 3.858651502843217e-06, + "loss": 1.6926, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.493215560913086, + "learning_rate": 3.856620633631195e-06, + "loss": 1.736, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.4341812133789062, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8152, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.391331195831299, + "learning_rate": 3.852558895207149e-06, + "loss": 1.7717, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.419492244720459, + "learning_rate": 3.850528025995126e-06, + "loss": 1.7283, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.349090576171875, + "learning_rate": 3.848497156783104e-06, + "loss": 1.7317, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.298647880554199, + "learning_rate": 3.84646628757108e-06, + "loss": 1.7389, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.4612667560577393, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.7333, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.3139944076538086, + "learning_rate": 3.842404549147035e-06, + "loss": 1.6944, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 1.924438714981079, + "eval_runtime": 45.8276, + "eval_samples_per_second": 21.821, + "eval_steps_per_second": 0.916, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 2.17958664894104, + "learning_rate": 3.840373679935013e-06, + "loss": 1.7474, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.337859869003296, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.7345, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4777672290802, + "learning_rate": 3.836311941510967e-06, + "loss": 1.77, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.280158519744873, + "learning_rate": 3.834281072298944e-06, + "loss": 1.7069, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2387664318084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.7889, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.373771905899048, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8268, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.4950132369995117, + "learning_rate": 3.828188464662876e-06, + "loss": 1.7586, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.3296444416046143, + "learning_rate": 3.826157595450853e-06, + "loss": 1.6991, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.6653530597686768, + "learning_rate": 3.82412672623883e-06, + "loss": 1.7505, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.485128402709961, + "learning_rate": 3.822095857026808e-06, + "loss": 1.7196, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 1.9221889972686768, + "eval_runtime": 46.6581, + "eval_samples_per_second": 21.433, + "eval_steps_per_second": 0.9, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.3726181983947754, + "learning_rate": 3.820064987814785e-06, + "loss": 1.7296, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.233992099761963, + "learning_rate": 3.818034118602762e-06, + "loss": 1.7313, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.5388147830963135, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.7435, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.4787333011627197, + "learning_rate": 3.813972380178717e-06, + "loss": 1.7733, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.232564926147461, + "learning_rate": 3.811941510966694e-06, + "loss": 1.693, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.3882386684417725, + "learning_rate": 3.809910641754671e-06, + "loss": 1.6529, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.4785983562469482, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.7456, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.685098886489868, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.7162, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.353523015975952, + "learning_rate": 3.803818034118603e-06, + "loss": 1.7852, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.342163562774658, + "learning_rate": 3.80178716490658e-06, + "loss": 1.7767, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 1.9235820770263672, + "eval_runtime": 47.1327, + "eval_samples_per_second": 21.217, + "eval_steps_per_second": 0.891, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.550818920135498, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.7328, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.3204798698425293, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.7058, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.4752719402313232, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8051, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.5202794075012207, + "learning_rate": 3.793663688058489e-06, + "loss": 1.7966, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.26521635055542, + "learning_rate": 3.791632818846467e-06, + "loss": 1.7879, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.331515312194824, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.7814, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.37178635597229, + "learning_rate": 3.787571080422421e-06, + "loss": 1.7635, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.321417808532715, + "learning_rate": 3.785540211210398e-06, + "loss": 1.7513, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.485989809036255, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.7306, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.7497379779815674, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8409, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 1.9275448322296143, + "eval_runtime": 46.2088, + "eval_samples_per_second": 21.641, + "eval_steps_per_second": 0.909, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 4.9214019775390625, + "learning_rate": 3.77944760357433e-06, + "loss": 1.7834, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 4.505277156829834, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.7179, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.5772030353546143, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.7766, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.254185438156128, + "learning_rate": 3.773354995938262e-06, + "loss": 1.7697, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 3.632233142852783, + "learning_rate": 3.771324126726239e-06, + "loss": 1.7968, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.174461841583252, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.7346, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.3983266353607178, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.6534, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 3.393791913986206, + "learning_rate": 3.765231519090171e-06, + "loss": 1.7561, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.2853987216949463, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.6963, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.649789810180664, + "learning_rate": 3.761169780666125e-06, + "loss": 1.7395, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 1.9242008924484253, + "eval_runtime": 46.6017, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 3.417872190475464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.784, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.316666841506958, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.7154, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.486161231994629, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.7903, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 4.538443565368652, + "learning_rate": 3.753046303818034e-06, + "loss": 1.7711, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.9772889614105225, + "learning_rate": 3.751015434606012e-06, + "loss": 1.8046, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 7.2353715896606445, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.7189, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.736, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.8026, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 12.551762580871582, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.7788, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 3.0898046493530273, + "learning_rate": 3.742891957757921e-06, + "loss": 1.7878, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 1.9646916389465332, + "eval_runtime": 44.1583, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 8.120022773742676, + "learning_rate": 3.740861088545898e-06, + "loss": 1.768, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 6.485774040222168, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.746, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.85135555267334, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.6968, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 5.8551435470581055, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.6796, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 4.45880126953125, + "learning_rate": 3.732737611697807e-06, + "loss": 1.7219, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 3.0435831546783447, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.6959, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.8194847106933594, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.8152, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 3.1266117095947266, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.6937, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 3.642854690551758, + "learning_rate": 3.724614134849716e-06, + "loss": 1.7992, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 3.344496965408325, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.7583, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 1.9293720722198486, + "eval_runtime": 50.8602, + "eval_samples_per_second": 19.662, + "eval_steps_per_second": 0.826, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.7666447162628174, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7441, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.8819239139556885, + "learning_rate": 3.718521527213648e-06, + "loss": 1.6314, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.6030640602111816, + "learning_rate": 3.716490658001625e-06, + "loss": 1.7389, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.705246925354004, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.7408, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 3.181328535079956, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7092, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 3.0449445247650146, + "learning_rate": 3.710398050365557e-06, + "loss": 1.7399, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.650517225265503, + "learning_rate": 3.708367181153534e-06, + "loss": 1.6861, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.7593672275543213, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.6701, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 3.0486466884613037, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.7191, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.502850294113159, + "learning_rate": 3.702274573517466e-06, + "loss": 1.6982, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 1.9228183031082153, + "eval_runtime": 49.565, + "eval_samples_per_second": 20.176, + "eval_steps_per_second": 0.847, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.400770664215088, + "learning_rate": 3.700243704305443e-06, + "loss": 1.6666, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.5583295822143555, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.7971, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.519505739212036, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.7076, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 3.106309413909912, + "learning_rate": 3.694151096669375e-06, + "loss": 1.7279, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.516906499862671, + "learning_rate": 3.692120227457352e-06, + "loss": 1.6265, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.603166103363037, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.7122, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.842773199081421, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.7312, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.8860342502593994, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7229, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.517254114151001, + "learning_rate": 3.683996750609261e-06, + "loss": 1.7342, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.430025100708008, + "learning_rate": 3.681965881397238e-06, + "loss": 1.6817, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 1.9156773090362549, + "eval_runtime": 44.6903, + "eval_samples_per_second": 22.376, + "eval_steps_per_second": 0.94, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.340075969696045, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.7044, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.4758968353271484, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.6935, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.878903865814209, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.7021, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.388366222381592, + "learning_rate": 3.673842404549147e-06, + "loss": 1.7229, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.5789167881011963, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8112, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.743072509765625, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.7239, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.5485000610351562, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.7641, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.53190541267395, + "learning_rate": 3.665718927701056e-06, + "loss": 1.6996, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.3354969024658203, + "learning_rate": 3.663688058489034e-06, + "loss": 1.6623, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.37371826171875, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.7109, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 1.9122159481048584, + "eval_runtime": 43.8844, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.406137228012085, + "learning_rate": 3.659626320064988e-06, + "loss": 1.6958, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.4220657348632812, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7196, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.6203134059906006, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.6014, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.5388472080230713, + "learning_rate": 3.65353371242892e-06, + "loss": 1.7272, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.672658681869507, + "learning_rate": 3.651502843216897e-06, + "loss": 1.6786, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4366531372070312, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7198, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.4872887134552, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.7082, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.371206283569336, + "learning_rate": 3.645410235580829e-06, + "loss": 1.6964, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.496103286743164, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.716, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.658316135406494, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.7263, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 1.9115047454833984, + "eval_runtime": 43.8788, + "eval_samples_per_second": 22.79, + "eval_steps_per_second": 0.957, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.5931882858276367, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.7586, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.4697139263153076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.7232, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.7274153232574463, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.6917, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.683138132095337, + "learning_rate": 3.633225020308692e-06, + "loss": 1.7518, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.4510817527770996, + "learning_rate": 3.63119415109667e-06, + "loss": 1.6533, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.6939656734466553, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.6545, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.572864532470703, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.6965, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.4816951751708984, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.6851, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.640394926071167, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7336, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.6707329750061035, + "learning_rate": 3.621039805036556e-06, + "loss": 1.7182, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 1.91274094581604, + "eval_runtime": 44.35, + "eval_samples_per_second": 22.548, + "eval_steps_per_second": 0.947, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.6947290897369385, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.7255, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.5496068000793457, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.6209, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.6196374893188477, + "learning_rate": 3.614947197400488e-06, + "loss": 1.7277, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.4534807205200195, + "learning_rate": 3.612916328188465e-06, + "loss": 1.7404, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.5428082942962646, + "learning_rate": 3.610885458976442e-06, + "loss": 1.6618, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.5431482791900635, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.6743, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.707956552505493, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.7483, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.5643982887268066, + "learning_rate": 3.604792851340374e-06, + "loss": 1.7304, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.4612927436828613, + "learning_rate": 3.602761982128351e-06, + "loss": 1.6561, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.3140358924865723, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.6834, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 1.9124692678451538, + "eval_runtime": 44.0522, + "eval_samples_per_second": 22.7, + "eval_steps_per_second": 0.953, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4636526107788086, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.7395, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.541180372238159, + "learning_rate": 3.596669374492283e-06, + "loss": 1.6579, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.621335983276367, + "learning_rate": 3.59463850528026e-06, + "loss": 1.7332, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.363450050354004, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.7047, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.8609447479248047, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7416, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.457338571548462, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.7516, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.3943030834198, + "learning_rate": 3.586515028432169e-06, + "loss": 1.6436, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.4719340801239014, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.6988, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3442330360412598, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.7165, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.3391900062561035, + "learning_rate": 3.580422420796101e-06, + "loss": 1.6263, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 1.9100929498672485, + "eval_runtime": 44.2504, + "eval_samples_per_second": 22.599, + "eval_steps_per_second": 0.949, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.533566474914551, + "learning_rate": 3.578391551584078e-06, + "loss": 1.7141, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.464540481567383, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.7335, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.4789979457855225, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7074, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.8252034187316895, + "learning_rate": 3.57229894394801e-06, + "loss": 1.6808, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.5870537757873535, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.7171, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.349393129348755, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.6687, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.3343207836151123, + "learning_rate": 3.566206336311942e-06, + "loss": 1.6351, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.558645248413086, + "learning_rate": 3.564175467099919e-06, + "loss": 1.708, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.7598931789398193, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.6797, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.6048543453216553, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7117, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 1.906927466392517, + "eval_runtime": 44.5189, + "eval_samples_per_second": 22.462, + "eval_steps_per_second": 0.943, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.6124815940856934, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.7328, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.3410215377807617, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.6932, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.5247764587402344, + "learning_rate": 3.554021121039805e-06, + "loss": 1.6628, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.642958164215088, + "learning_rate": 3.551990251827783e-06, + "loss": 1.7035, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.3978517055511475, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.6796, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.3969287872314453, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.6626, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.5723509788513184, + "learning_rate": 3.545897644191714e-06, + "loss": 1.7135, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.545154571533203, + "learning_rate": 3.543866774979692e-06, + "loss": 1.6626, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.501058340072632, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.7112, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.695483684539795, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.6773, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 1.9063464403152466, + "eval_runtime": 44.2034, + "eval_samples_per_second": 22.623, + "eval_steps_per_second": 0.95, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 3.0584375858306885, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.7242, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.445220947265625, + "learning_rate": 3.535743298131601e-06, + "loss": 1.7349, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.427220106124878, + "learning_rate": 3.533712428919578e-06, + "loss": 1.677, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.5148212909698486, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.7007, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.644319534301758, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.7442, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.6118004322052, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.7509, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.5224156379699707, + "learning_rate": 3.525588952071487e-06, + "loss": 1.7279, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.4668405055999756, + "learning_rate": 3.523558082859464e-06, + "loss": 1.6754, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 3.8330910205841064, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.6815, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.569058895111084, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.6518, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 1.906802773475647, + "eval_runtime": 50.8521, + "eval_samples_per_second": 19.665, + "eval_steps_per_second": 0.826, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.705265522003174, + "learning_rate": 3.517465475223396e-06, + "loss": 1.7172, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 3.2667057514190674, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7641, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 3.5119211673736572, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.7178, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.7348034381866455, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.6815, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 3.019892454147339, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.6417, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 3.2486684322357178, + "learning_rate": 3.507311129163282e-06, + "loss": 1.6888, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.897071361541748, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7121, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.7342352867126465, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.6339, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.691178560256958, + "learning_rate": 3.501218521527214e-06, + "loss": 1.7069, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.548682689666748, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7218, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 1.9029825925827026, + "eval_runtime": 48.287, + "eval_samples_per_second": 20.71, + "eval_steps_per_second": 0.87, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.422056198120117, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.6486, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.4367074966430664, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.759, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.48002290725708, + "learning_rate": 3.493095044679123e-06, + "loss": 1.5902, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.5030295848846436, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.7108, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.5892179012298584, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.6154, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.4282920360565186, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.6807, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.4826700687408447, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7284, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.416597843170166, + "learning_rate": 3.482940698619009e-06, + "loss": 1.6972, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.4664933681488037, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.7119, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.541856288909912, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.6752, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 1.8999215364456177, + "eval_runtime": 46.3528, + "eval_samples_per_second": 21.574, + "eval_steps_per_second": 0.906, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4803109169006348, + "learning_rate": 3.476848090982941e-06, + "loss": 1.776, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.4363205432891846, + "learning_rate": 3.474817221770918e-06, + "loss": 1.7196, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.4377009868621826, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.6937, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.425791025161743, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.6547, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3216712474823, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.6789, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.65527081489563, + "learning_rate": 3.466693744922827e-06, + "loss": 1.7231, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.3415145874023438, + "learning_rate": 3.464662875710805e-06, + "loss": 1.6805, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.3864822387695312, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.7153, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.490896224975586, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.6866, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.339711904525757, + "learning_rate": 3.458570268074736e-06, + "loss": 1.6732, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 1.8985264301300049, + "eval_runtime": 46.7854, + "eval_samples_per_second": 21.374, + "eval_steps_per_second": 0.898, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.393709659576416, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.6549, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.564114809036255, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.7019, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3826403617858887, + "learning_rate": 3.452477660438668e-06, + "loss": 1.6427, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.3185465335845947, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.6916, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4503891468048096, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.6793, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.386446714401245, + "learning_rate": 3.4463850528026e-06, + "loss": 1.7186, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.6404714584350586, + "learning_rate": 3.444354183590577e-06, + "loss": 1.6605, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.342374324798584, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.6219, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.7503273487091064, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.693, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.5416243076324463, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7323, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 1.9016659259796143, + "eval_runtime": 48.5456, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.5787603855133057, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.7057, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.649528980255127, + "learning_rate": 3.434199837530463e-06, + "loss": 1.6348, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.504089832305908, + "learning_rate": 3.432168968318441e-06, + "loss": 1.7202, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.5624678134918213, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7351, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.641331911087036, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.7453, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.599500894546509, + "learning_rate": 3.426076360682372e-06, + "loss": 1.6942, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.668599843978882, + "learning_rate": 3.42404549147035e-06, + "loss": 1.6337, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.75667667388916, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.6988, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.641526937484741, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.7641, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.4972548484802246, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.7269, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 1.8979382514953613, + "eval_runtime": 45.6555, + "eval_samples_per_second": 21.903, + "eval_steps_per_second": 0.92, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.5454585552215576, + "learning_rate": 3.415922014622259e-06, + "loss": 1.7294, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.5921928882598877, + "learning_rate": 3.413891145410236e-06, + "loss": 1.7145, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.4028453826904297, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.6985, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.604522705078125, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.6586, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.528970241546631, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.734, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.565030574798584, + "learning_rate": 3.405767668562145e-06, + "loss": 1.6997, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.7735602855682373, + "learning_rate": 3.403736799350122e-06, + "loss": 1.6543, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.5116474628448486, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.7711, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.4827239513397217, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.6831, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.544309377670288, + "learning_rate": 3.397644191714054e-06, + "loss": 1.6905, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 1.8945367336273193, + "eval_runtime": 46.6027, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.456636905670166, + "learning_rate": 3.395613322502031e-06, + "loss": 1.646, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.5900352001190186, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.6967, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 3.3344573974609375, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.6947, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.5621678829193115, + "learning_rate": 3.389520714865963e-06, + "loss": 1.7084, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.5676684379577637, + "learning_rate": 3.38748984565394e-06, + "loss": 1.6788, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.600691318511963, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.6811, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.5324196815490723, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.661, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.7559397220611572, + "learning_rate": 3.381397238017872e-06, + "loss": 1.7473, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.526397466659546, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.3752379417419434, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.689, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 1.8995447158813477, + "eval_runtime": 48.0083, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.5902793407440186, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.6899, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 3.157573938369751, + "learning_rate": 3.373273761169781e-06, + "loss": 1.6792, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.628035545349121, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.7003, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.4810822010040283, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.6489, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.4859938621520996, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.6768, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.6773345470428467, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7046, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.9695260524749756, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.7341, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.8146679401397705, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.7472, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.534985065460205, + "learning_rate": 3.359057676685622e-06, + "loss": 1.6199, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.9351718425750732, + "learning_rate": 3.357026807473599e-06, + "loss": 1.6656, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 1.8962808847427368, + "eval_runtime": 47.0443, + "eval_samples_per_second": 21.257, + "eval_steps_per_second": 0.893, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.4404256343841553, + "learning_rate": 3.354995938261576e-06, + "loss": 1.6207, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.6691436767578125, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.6627, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.5436172485351562, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.6919, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.7167909145355225, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.6434, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.575446367263794, + "learning_rate": 3.346872461413485e-06, + "loss": 1.6554, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.3874504566192627, + "learning_rate": 3.344841592201463e-06, + "loss": 1.6691, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.66317081451416, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.6864, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.4396398067474365, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.7217, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.5002803802490234, + "learning_rate": 3.338748984565394e-06, + "loss": 1.7556, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.679577350616455, + "learning_rate": 3.336718115353372e-06, + "loss": 1.6822, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 1.8942477703094482, + "eval_runtime": 44.3804, + "eval_samples_per_second": 22.532, + "eval_steps_per_second": 0.946, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.6119370460510254, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.662, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.6024818420410156, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.687, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 4.128854751586914, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.6898, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3910412788391113, + "learning_rate": 3.32859463850528e-06, + "loss": 1.6274, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.8630096912384033, + "learning_rate": 3.326563769293258e-06, + "loss": 1.6554, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 4.201050281524658, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.6676, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.492168664932251, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.6319, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 3.6807360649108887, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.6776, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.7271926403045654, + "learning_rate": 3.318440292445167e-06, + "loss": 1.6383, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.5567307472229004, + "learning_rate": 3.316409423233144e-06, + "loss": 1.7325, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 1.892768383026123, + "eval_runtime": 50.909, + "eval_samples_per_second": 19.643, + "eval_steps_per_second": 0.825, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.5373101234436035, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.689, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.602923631668091, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.6861, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.507110118865967, + "learning_rate": 3.310316815597076e-06, + "loss": 1.7575, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.361786127090454, + "learning_rate": 3.308285946385053e-06, + "loss": 1.6267, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5360615253448486, + "learning_rate": 3.30625507717303e-06, + "loss": 1.6911, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.6238744258880615, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.7193, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.5358500480651855, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.6691, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.7285404205322266, + "learning_rate": 3.300162469536962e-06, + "loss": 1.7249, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.346160650253296, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.6958, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.5415399074554443, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.7544, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 1.8925012350082397, + "eval_runtime": 46.4957, + "eval_samples_per_second": 21.507, + "eval_steps_per_second": 0.903, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.5924015045166016, + "learning_rate": 3.294069861900894e-06, + "loss": 1.6901, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.4649693965911865, + "learning_rate": 3.292038992688871e-06, + "loss": 1.7234, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.3648035526275635, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.7324, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.5146145820617676, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.631, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.5919511318206787, + "learning_rate": 3.285946385052803e-06, + "loss": 1.706, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.299527168273926, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.608, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.480637788772583, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.7415, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.458183526992798, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.6806, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.4864983558654785, + "learning_rate": 3.277822908204712e-06, + "loss": 1.6373, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.3432068824768066, + "learning_rate": 3.275792038992689e-06, + "loss": 1.6832, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 1.8899296522140503, + "eval_runtime": 44.3309, + "eval_samples_per_second": 22.558, + "eval_steps_per_second": 0.947, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.365891218185425, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.6492, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.463261842727661, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.6847, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.3977205753326416, + "learning_rate": 3.269699431356621e-06, + "loss": 1.6789, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.4172558784484863, + "learning_rate": 3.267668562144598e-06, + "loss": 1.6495, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.663381814956665, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.7139, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.3702306747436523, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.6876, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 3.1759073734283447, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.7465, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.6344592571258545, + "learning_rate": 3.259545085296507e-06, + "loss": 1.6339, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.364490032196045, + "learning_rate": 3.257514216084484e-06, + "loss": 1.7394, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.5639748573303223, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.6928, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 1.8937300443649292, + "eval_runtime": 43.9356, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 3.152968645095825, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.6693, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.45776104927063, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.6999, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.5289807319641113, + "learning_rate": 3.249390739236393e-06, + "loss": 1.6523, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.4168646335601807, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.6823, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.5115208625793457, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.6816, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.685535192489624, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.6307, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.253084897994995, + "learning_rate": 3.241267262388302e-06, + "loss": 1.662, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.4825408458709717, + "learning_rate": 3.23923639317628e-06, + "loss": 1.6444, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.691168785095215, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.6912, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.978756904602051, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7536, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 1.8903727531433105, + "eval_runtime": 43.4099, + "eval_samples_per_second": 23.036, + "eval_steps_per_second": 0.968, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.317105531692505, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.6441, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 3.1030166149139404, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.7205, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.595792770385742, + "learning_rate": 3.229082047116166e-06, + "loss": 1.677, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.3315582275390625, + "learning_rate": 3.227051177904143e-06, + "loss": 1.6779, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.4262566566467285, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.7003, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.4438798427581787, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.6081, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.573531150817871, + "learning_rate": 3.220958570268075e-06, + "loss": 1.6489, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.477809429168701, + "learning_rate": 3.218927701056052e-06, + "loss": 1.6025, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.5499629974365234, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.7337, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.4699153900146484, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.6443, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 1.8867121934890747, + "eval_runtime": 42.9343, + "eval_samples_per_second": 23.291, + "eval_steps_per_second": 0.978, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.4200708866119385, + "learning_rate": 3.212835093419984e-06, + "loss": 1.6743, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.418069839477539, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.723, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.452284336090088, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.6358, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.4241347312927246, + "learning_rate": 3.206742485783916e-06, + "loss": 1.638, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.461860418319702, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.7017, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.591986894607544, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7065, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.478470802307129, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.6262, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.49831223487854, + "learning_rate": 3.198619008935825e-06, + "loss": 1.6871, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.485332727432251, + "learning_rate": 3.196588139723802e-06, + "loss": 1.6864, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.533710479736328, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.6794, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 1.88768470287323, + "eval_runtime": 43.5948, + "eval_samples_per_second": 22.938, + "eval_steps_per_second": 0.963, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.368609666824341, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.6877, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.4875223636627197, + "learning_rate": 3.190495532087734e-06, + "loss": 1.7406, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.5075156688690186, + "learning_rate": 3.188464662875711e-06, + "loss": 1.6545, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.428731918334961, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.6213, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.467618465423584, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.6263, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.561612129211426, + "learning_rate": 3.182372055239643e-06, + "loss": 1.6987, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.5035271644592285, + "learning_rate": 3.18034118602762e-06, + "loss": 1.6902, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.538311243057251, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.6523, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.486321210861206, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.6665, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.433173894882202, + "learning_rate": 3.174248578391552e-06, + "loss": 1.6605, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 1.8853652477264404, + "eval_runtime": 43.3417, + "eval_samples_per_second": 23.072, + "eval_steps_per_second": 0.969, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.4658470153808594, + "learning_rate": 3.172217709179529e-06, + "loss": 1.6615, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.452450752258301, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.6332, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.5310161113739014, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.6846, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.4504501819610596, + "learning_rate": 3.166125101543461e-06, + "loss": 1.6996, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.6004488468170166, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.6302, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.5000431537628174, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.6742, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.4869067668914795, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.6797, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.5018959045410156, + "learning_rate": 3.15800162469537e-06, + "loss": 1.673, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.6645195484161377, + "learning_rate": 3.155970755483347e-06, + "loss": 1.6908, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.3708560466766357, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.683, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 1.8829350471496582, + "eval_runtime": 43.5383, + "eval_samples_per_second": 22.968, + "eval_steps_per_second": 0.965, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.574108362197876, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.7155, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.4669814109802246, + "learning_rate": 3.149878147847279e-06, + "loss": 1.7529, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4713149070739746, + "learning_rate": 3.147847278635256e-06, + "loss": 1.6491, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.5128800868988037, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.6276, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.740370035171509, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.6778, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.6038050651550293, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.6527, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.782407283782959, + "learning_rate": 3.139723801787165e-06, + "loss": 1.6989, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.7673134803771973, + "learning_rate": 3.137692932575143e-06, + "loss": 1.6291, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.7472870349884033, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.6776, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.4887819290161133, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7009, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.8792160749435425, + "eval_runtime": 43.7334, + "eval_samples_per_second": 22.866, + "eval_steps_per_second": 0.96, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.4528491497039795, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.7245, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.5177690982818604, + "learning_rate": 3.129569455727051e-06, + "loss": 1.6852, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.9238510131835938, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.6994, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.446660041809082, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.6636, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.644211530685425, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.6983, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.697342872619629, + "learning_rate": 3.12144597887896e-06, + "loss": 1.7172, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.714172124862671, + "learning_rate": 3.119415109666938e-06, + "loss": 1.6702, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.471735715866089, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.6945, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.5089669227600098, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.7181, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.349229335784912, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.7297, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.880064845085144, + "eval_runtime": 47.0636, + "eval_samples_per_second": 21.248, + "eval_steps_per_second": 0.892, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.6843974590301514, + "learning_rate": 3.111291632818847e-06, + "loss": 1.6698, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.4325790405273438, + "learning_rate": 3.109260763606824e-06, + "loss": 1.6973, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.488459825515747, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.6552, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.4992082118988037, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.7114, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.4742603302001953, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.7073, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.556330442428589, + "learning_rate": 3.101137286758733e-06, + "loss": 1.7108, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.5711283683776855, + "learning_rate": 3.09910641754671e-06, + "loss": 1.6811, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.6467158794403076, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.7395, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.465963363647461, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.7006, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.622716188430786, + "learning_rate": 3.093013809910642e-06, + "loss": 1.7385, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.8808013200759888, + "eval_runtime": 45.5126, + "eval_samples_per_second": 21.972, + "eval_steps_per_second": 0.923, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.489701509475708, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.6949, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.4398739337921143, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.6802, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.3654873371124268, + "learning_rate": 3.086921202274574e-06, + "loss": 1.6452, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.518768548965454, + "learning_rate": 3.084890333062551e-06, + "loss": 1.6913, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.63173246383667, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.6611, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.531484603881836, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.611, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.4796738624572754, + "learning_rate": 3.078797725426483e-06, + "loss": 1.6764, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.39241099357605, + "learning_rate": 3.07676685621446e-06, + "loss": 1.6417, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3984735012054443, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.6546, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.5202760696411133, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.6879, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.879974365234375, + "eval_runtime": 44.395, + "eval_samples_per_second": 22.525, + "eval_steps_per_second": 0.946, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.5556983947753906, + "learning_rate": 3.070674248578392e-06, + "loss": 1.6733, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.3853955268859863, + "learning_rate": 3.068643379366369e-06, + "loss": 1.6274, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.4503982067108154, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.6896, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.5618021488189697, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.6592, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.6265292167663574, + "learning_rate": 3.062550771730301e-06, + "loss": 1.6634, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.4617390632629395, + "learning_rate": 3.060519902518278e-06, + "loss": 1.6032, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.4191603660583496, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.6871, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.5509119033813477, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.653, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.5399646759033203, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.7046, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.446967124938965, + "learning_rate": 3.052396425670187e-06, + "loss": 1.6547, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.8803874254226685, + "eval_runtime": 45.361, + "eval_samples_per_second": 22.045, + "eval_steps_per_second": 0.926, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.419206142425537, + "learning_rate": 3.050365556458164e-06, + "loss": 1.6517, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.4592065811157227, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.693, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.5429770946502686, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7124, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.5141642093658447, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.7103, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.362539291381836, + "learning_rate": 3.042242079610073e-06, + "loss": 1.6205, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.401175022125244, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.6479, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.5883190631866455, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.6732, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.5273513793945312, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.6673, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.6569600105285645, + "learning_rate": 3.034118602761982e-06, + "loss": 1.6396, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.8843557834625244, + "learning_rate": 3.03208773354996e-06, + "loss": 1.7136, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.8763896226882935, + "eval_runtime": 46.1886, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.6459810733795166, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.6939, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.4619321823120117, + "learning_rate": 3.028025995125914e-06, + "loss": 1.6197, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.390430450439453, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.6805, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.5494728088378906, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.6116, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.5397655963897705, + "learning_rate": 3.021933387489846e-06, + "loss": 1.6753, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.634716510772705, + "learning_rate": 3.019902518277823e-06, + "loss": 1.6892, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.5844202041625977, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.6401, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.577362060546875, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.6405, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.3554489612579346, + "learning_rate": 3.013809910641755e-06, + "loss": 1.6427, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.893918037414551, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.6739, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.8772739171981812, + "eval_runtime": 44.9972, + "eval_samples_per_second": 22.224, + "eval_steps_per_second": 0.933, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.4818382263183594, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.644, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.4990177154541016, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.6962, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.627502918243408, + "learning_rate": 3.005686433793664e-06, + "loss": 1.639, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.5949175357818604, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.7095, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.422765016555786, + "learning_rate": 3.001624695369618e-06, + "loss": 1.6572, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.708702564239502, + "learning_rate": 2.999593826157596e-06, + "loss": 1.6747, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.624866247177124, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.6643, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.5831358432769775, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.7036, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.5799436569213867, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.6628, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.7047722339630127, + "learning_rate": 2.991470349309505e-06, + "loss": 1.6591, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.8736577033996582, + "eval_runtime": 44.7095, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.411628246307373, + "learning_rate": 2.989439480097482e-06, + "loss": 1.6644, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.8774638175964355, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.7192, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.857028007507324, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7079, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.4619641304016113, + "learning_rate": 2.983346872461414e-06, + "loss": 1.6846, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.6566193103790283, + "learning_rate": 2.981316003249391e-06, + "loss": 1.6768, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.681217670440674, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.7002, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.39505672454834, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.65, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.6062734127044678, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.6659, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.6402504444122314, + "learning_rate": 2.9731925264013e-06, + "loss": 1.6491, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.497014284133911, + "learning_rate": 2.971161657189277e-06, + "loss": 1.6383, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.874670147895813, + "eval_runtime": 42.9696, + "eval_samples_per_second": 23.272, + "eval_steps_per_second": 0.977, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.9070584774017334, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.6388, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.580744981765747, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7078, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.4637644290924072, + "learning_rate": 2.965069049553209e-06, + "loss": 1.6266, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.4489941596984863, + "learning_rate": 2.963038180341186e-06, + "loss": 1.705, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.664654016494751, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.6964, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.7017202377319336, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.6356, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.3691024780273438, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.6566, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.540384292602539, + "learning_rate": 2.954914703493095e-06, + "loss": 1.6334, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.5240535736083984, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.6316, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 3.105684518814087, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.6769, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.8741132020950317, + "eval_runtime": 42.6795, + "eval_samples_per_second": 23.43, + "eval_steps_per_second": 0.984, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.494252920150757, + "learning_rate": 2.948822095857027e-06, + "loss": 1.6359, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 4.069019317626953, + "learning_rate": 2.946791226645004e-06, + "loss": 1.621, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.641447067260742, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.6502, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.6700453758239746, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.6593, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.686068296432495, + "learning_rate": 2.940698619008936e-06, + "loss": 1.67, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.409055471420288, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.6809, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.4707298278808594, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.69, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.4735982418060303, + "learning_rate": 2.934606011372868e-06, + "loss": 1.666, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.633183479309082, + "learning_rate": 2.932575142160845e-06, + "loss": 1.6597, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.4808642864227295, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.6667, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.8731157779693604, + "eval_runtime": 41.3423, + "eval_samples_per_second": 24.188, + "eval_steps_per_second": 1.016, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.4044833183288574, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.7028, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.942699909210205, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.6678, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.505645990371704, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.5972, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.700967788696289, + "learning_rate": 2.922420796100731e-06, + "loss": 1.7064, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.4805755615234375, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.6201, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.753181219100952, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.6704, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.488865613937378, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7057, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.5464885234832764, + "learning_rate": 2.91429731925264e-06, + "loss": 1.6678, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.4997060298919678, + "learning_rate": 2.912266450040618e-06, + "loss": 1.662, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.5162436962127686, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.6501, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.869848370552063, + "eval_runtime": 43.3583, + "eval_samples_per_second": 23.064, + "eval_steps_per_second": 0.969, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.616129159927368, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.6206, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.761080503463745, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7131, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.5994064807891846, + "learning_rate": 2.904142973192527e-06, + "loss": 1.6693, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 3.4927968978881836, + "learning_rate": 2.902112103980504e-06, + "loss": 1.6694, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.6050939559936523, + "learning_rate": 2.900081234768481e-06, + "loss": 1.6419, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.796086311340332, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.6354, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.5658411979675293, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.652, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.3802385330200195, + "learning_rate": 2.893988627132413e-06, + "loss": 1.5902, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.479189872741699, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.6617, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.4367568492889404, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.6691, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.868497371673584, + "eval_runtime": 45.172, + "eval_samples_per_second": 22.138, + "eval_steps_per_second": 0.93, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.51888370513916, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.6872, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.5192768573760986, + "learning_rate": 2.885865150284322e-06, + "loss": 1.6584, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.5898501873016357, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.6179, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.698101043701172, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.663, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.889942169189453, + "learning_rate": 2.879772542648254e-06, + "loss": 1.5954, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.7909929752349854, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.7264, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 3.1931703090667725, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7091, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.7934632301330566, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.7204, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.470545530319214, + "learning_rate": 2.871649065800163e-06, + "loss": 1.644, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 3.011662006378174, + "learning_rate": 2.86961819658814e-06, + "loss": 1.6743, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.8678737878799438, + "eval_runtime": 44.7875, + "eval_samples_per_second": 22.328, + "eval_steps_per_second": 0.938, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.60823655128479, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.6467, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.5347230434417725, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.7044, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.481715440750122, + "learning_rate": 2.863525588952072e-06, + "loss": 1.5969, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.4966487884521484, + "learning_rate": 2.861494719740049e-06, + "loss": 1.6231, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.8809802532196045, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.6562, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.570204734802246, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.6511, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.6073813438415527, + "learning_rate": 2.855402112103981e-06, + "loss": 1.6456, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.8125052452087402, + "learning_rate": 2.853371242891958e-06, + "loss": 1.6722, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 5.341272354125977, + "learning_rate": 2.851340373679935e-06, + "loss": 1.7025, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 3.138002395629883, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.6676, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.8733265399932861, + "eval_runtime": 45.4915, + "eval_samples_per_second": 21.982, + "eval_steps_per_second": 0.923, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 7.264789581298828, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.6335, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 3.8533623218536377, + "learning_rate": 2.845247766043867e-06, + "loss": 1.6029, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 4.1789164543151855, + "learning_rate": 2.843216896831844e-06, + "loss": 1.6368, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 3.117408037185669, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.6924, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.721306085586548, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.6914, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.7099668979644775, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.6508, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 3.7823967933654785, + "learning_rate": 2.835093419983753e-06, + "loss": 1.636, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.696370840072632, + "learning_rate": 2.833062550771731e-06, + "loss": 1.6175, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.7303690910339355, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.6654, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.867727041244507, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.6717, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.8720332384109497, + "eval_runtime": 44.164, + "eval_samples_per_second": 22.643, + "eval_steps_per_second": 0.951, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 3.1209630966186523, + "learning_rate": 2.826969943135662e-06, + "loss": 1.6635, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.5057783126831055, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.6309, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.6650168895721436, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.6365, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.371305465698242, + "learning_rate": 2.820877335499594e-06, + "loss": 1.603, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.387404203414917, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.6189, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.6265947818756104, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.6511, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.6108627319335938, + "learning_rate": 2.814784727863526e-06, + "loss": 1.6727, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.7021450996398926, + "learning_rate": 2.812753858651503e-06, + "loss": 1.6334, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.470118999481201, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.6155, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5722644329071045, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.6788, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.8727209568023682, + "eval_runtime": 44.6253, + "eval_samples_per_second": 22.409, + "eval_steps_per_second": 0.941, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.7143876552581787, + "learning_rate": 2.806661251015435e-06, + "loss": 1.6956, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.8522002696990967, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.7558, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.5352110862731934, + "learning_rate": 2.802599512591389e-06, + "loss": 1.6592, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.7065908908843994, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.6121, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.616450786590576, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.6643, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.528763771057129, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7104, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 3.725569009780884, + "learning_rate": 2.794476035743298e-06, + "loss": 1.6713, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.487497091293335, + "learning_rate": 2.792445166531276e-06, + "loss": 1.692, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.534766912460327, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.6257, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.8553240299224854, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.6147, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.8675988912582397, + "eval_runtime": 46.4445, + "eval_samples_per_second": 21.531, + "eval_steps_per_second": 0.904, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.5054166316986084, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.6082, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.4133870601654053, + "learning_rate": 2.784321689683185e-06, + "loss": 1.6148, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.5863375663757324, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7032, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.58491849899292, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.679, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.5225989818573, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.6688, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.366861343383789, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.7076, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.4159810543060303, + "learning_rate": 2.774167343623071e-06, + "loss": 1.6301, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.5394275188446045, + "learning_rate": 2.772136474411048e-06, + "loss": 1.5758, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.409821033477783, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.664, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.4548892974853516, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.643, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.8672850131988525, + "eval_runtime": 46.3149, + "eval_samples_per_second": 21.591, + "eval_steps_per_second": 0.907, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.7483904361724854, + "learning_rate": 2.76604386677498e-06, + "loss": 1.6898, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.538804769515991, + "learning_rate": 2.764012997562957e-06, + "loss": 1.6019, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.7153313159942627, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.691, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.6892242431640625, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.6386, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.550957679748535, + "learning_rate": 2.757920389926889e-06, + "loss": 1.6237, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.3658559322357178, + "learning_rate": 2.755889520714866e-06, + "loss": 1.6015, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.6524102687835693, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.6644, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.7067437171936035, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.687, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.4044861793518066, + "learning_rate": 2.749796913078798e-06, + "loss": 1.6535, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.4933829307556152, + "learning_rate": 2.747766043866775e-06, + "loss": 1.7017, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.8652688264846802, + "eval_runtime": 46.8502, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.7062129974365234, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7398, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.5704996585845947, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.6561, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.5875790119171143, + "learning_rate": 2.741673436230707e-06, + "loss": 1.643, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.7498223781585693, + "learning_rate": 2.739642567018684e-06, + "loss": 1.6652, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.7275257110595703, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.6559, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.6349947452545166, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.6489, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.7058634757995605, + "learning_rate": 2.733549959382616e-06, + "loss": 1.7176, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.481344223022461, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.685, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.468277931213379, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.6305, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.6183228492736816, + "learning_rate": 2.727457351746548e-06, + "loss": 1.6151, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.864329218864441, + "eval_runtime": 43.0443, + "eval_samples_per_second": 23.232, + "eval_steps_per_second": 0.976, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.50679349899292, + "learning_rate": 2.725426482534525e-06, + "loss": 1.6468, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.6565401554107666, + "learning_rate": 2.723395613322502e-06, + "loss": 1.6496, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.5332367420196533, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.6424, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.6645619869232178, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.8012, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.7359142303466797, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.6025, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.660728693008423, + "learning_rate": 2.715272136474411e-06, + "loss": 1.6601, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.5184221267700195, + "learning_rate": 2.713241267262389e-06, + "loss": 1.6383, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.643712043762207, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.6834, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.504753351211548, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.6292, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.3439948558807373, + "learning_rate": 2.70714865962632e-06, + "loss": 1.6558, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.861581563949585, + "eval_runtime": 43.9224, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4820125102996826, + "learning_rate": 2.705117790414298e-06, + "loss": 1.69, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.6156861782073975, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.6304, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.490123987197876, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.6573, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.4439024925231934, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.6368, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.891603708267212, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7111, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.481175661087036, + "learning_rate": 2.694963444354184e-06, + "loss": 1.6408, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.5378847122192383, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.6782, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.592477321624756, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.6085, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.9336471557617188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.6719, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.3115012645721436, + "learning_rate": 2.686839967506093e-06, + "loss": 1.5797, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.8649041652679443, + "eval_runtime": 43.711, + "eval_samples_per_second": 22.878, + "eval_steps_per_second": 0.961, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.627028465270996, + "learning_rate": 2.68480909829407e-06, + "loss": 1.5784, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.917670488357544, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7007, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.418285608291626, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.6613, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.607773780822754, + "learning_rate": 2.678716490658002e-06, + "loss": 1.6561, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.569714307785034, + "learning_rate": 2.676685621445979e-06, + "loss": 1.6541, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.359494209289551, + "learning_rate": 2.674654752233956e-06, + "loss": 1.6467, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.5155694484710693, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.7003, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.439147710800171, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.5679, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.5044116973876953, + "learning_rate": 2.668562144597888e-06, + "loss": 1.6626, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.4475765228271484, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.656, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.8636877536773682, + "eval_runtime": 43.5138, + "eval_samples_per_second": 22.981, + "eval_steps_per_second": 0.965, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4616832733154297, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.6101, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.4125123023986816, + "learning_rate": 2.66246953696182e-06, + "loss": 1.6021, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.571326494216919, + "learning_rate": 2.660438667749797e-06, + "loss": 1.6482, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.424332618713379, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.653, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.3034703731536865, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.6096, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.4758505821228027, + "learning_rate": 2.654346060113729e-06, + "loss": 1.6274, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 3.800152540206909, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.6217, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.479827880859375, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7202, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.604523181915283, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.6778, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 3.595973491668701, + "learning_rate": 2.646222583265638e-06, + "loss": 1.6678, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.8647090196609497, + "eval_runtime": 43.0415, + "eval_samples_per_second": 23.233, + "eval_steps_per_second": 0.976, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 3.297319173812866, + "learning_rate": 2.644191714053615e-06, + "loss": 1.637, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.423567056655884, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.6018, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 4.4547953605651855, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.6594, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 3.0775582790374756, + "learning_rate": 2.638099106417547e-06, + "loss": 1.6259, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.747938871383667, + "learning_rate": 2.636068237205524e-06, + "loss": 1.7208, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.5350255966186523, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.5705, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 3.0630743503570557, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.6507, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 3.5475759506225586, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7045, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.534303903579712, + "learning_rate": 2.627944760357433e-06, + "loss": 1.7173, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.6964452266693115, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.6823, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.865207552909851, + "eval_runtime": 42.9435, + "eval_samples_per_second": 23.286, + "eval_steps_per_second": 0.978, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.924957752227783, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.6036, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.7058141231536865, + "learning_rate": 2.621852152721365e-06, + "loss": 1.6701, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.735339641571045, + "learning_rate": 2.619821283509342e-06, + "loss": 1.607, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.573958396911621, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.667, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5776426792144775, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.6938, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 4.454444408416748, + "learning_rate": 2.613728675873274e-06, + "loss": 1.7139, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.4883337020874023, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.6524, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.672116994857788, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.6197, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 3.511932611465454, + "learning_rate": 2.607636068237206e-06, + "loss": 1.6739, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.5960047245025635, + "learning_rate": 2.605605199025183e-06, + "loss": 1.6516, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.8604435920715332, + "eval_runtime": 42.9247, + "eval_samples_per_second": 23.297, + "eval_steps_per_second": 0.978, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.565376043319702, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.6884, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.676114797592163, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.6693, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.582751750946045, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.6499, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.748908281326294, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.6837, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.5578067302703857, + "learning_rate": 2.595450852965069e-06, + "loss": 1.6959, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.5702805519104004, + "learning_rate": 2.593419983753047e-06, + "loss": 1.6429, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.4748787879943848, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.6878, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.6157331466674805, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.6246, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.8377959728240967, + "learning_rate": 2.587327376116978e-06, + "loss": 1.6451, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.485398292541504, + "learning_rate": 2.585296506904956e-06, + "loss": 1.6288, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.8593864440917969, + "eval_runtime": 44.6341, + "eval_samples_per_second": 22.404, + "eval_steps_per_second": 0.941, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.46850323677063, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.5766, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.673283576965332, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7383, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.935293197631836, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.5742, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.531733751296997, + "learning_rate": 2.577173030056865e-06, + "loss": 1.6714, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.528425455093384, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6802, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.5644803047180176, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.6201, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.382451295852661, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.5824, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.63320255279541, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.6685, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.488844394683838, + "learning_rate": 2.567018683996751e-06, + "loss": 1.6078, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.7845425605773926, + "learning_rate": 2.564987814784728e-06, + "loss": 1.638, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.8572624921798706, + "eval_runtime": 43.2181, + "eval_samples_per_second": 23.138, + "eval_steps_per_second": 0.972, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.709656000137329, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.6425, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.651277542114258, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7401, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.5424256324768066, + "learning_rate": 2.55889520714866e-06, + "loss": 1.6272, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.6508235931396484, + "learning_rate": 2.556864337936637e-06, + "loss": 1.6384, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.882636308670044, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.6965, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.518367290496826, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.6551, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.7977044582366943, + "learning_rate": 2.550771730300569e-06, + "loss": 1.679, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.353041172027588, + "learning_rate": 2.548740861088546e-06, + "loss": 1.5745, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.5718655586242676, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.5655, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.6955313682556152, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.6764, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.8569272756576538, + "eval_runtime": 45.3844, + "eval_samples_per_second": 22.034, + "eval_steps_per_second": 0.925, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4756696224212646, + "learning_rate": 2.542648253452478e-06, + "loss": 1.6522, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.573193311691284, + "learning_rate": 2.540617384240455e-06, + "loss": 1.6079, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.5967671871185303, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.6341, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.7896342277526855, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.6449, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.519728183746338, + "learning_rate": 2.534524776604387e-06, + "loss": 1.549, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.543139934539795, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.6335, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.5580177307128906, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6402, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.443547010421753, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.6421, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.7075212001800537, + "learning_rate": 2.526401299756296e-06, + "loss": 1.6283, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.6505324840545654, + "learning_rate": 2.524370430544273e-06, + "loss": 1.6028, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.8556184768676758, + "eval_runtime": 43.731, + "eval_samples_per_second": 22.867, + "eval_steps_per_second": 0.96, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.597172260284424, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.6631, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 3.1529037952423096, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.6219, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.5374975204467773, + "learning_rate": 2.518277822908205e-06, + "loss": 1.6476, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 3.0557401180267334, + "learning_rate": 2.516246953696182e-06, + "loss": 1.6627, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 6.759350776672363, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.6532, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.670427083969116, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.658, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 4.269686698913574, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.5873, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 13.37558650970459, + "learning_rate": 2.508123476848091e-06, + "loss": 1.6872, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 7.060877323150635, + "learning_rate": 2.506092607636069e-06, + "loss": 1.6406, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 11.259824752807617, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.6421, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.861360788345337, + "eval_runtime": 45.4438, + "eval_samples_per_second": 22.005, + "eval_steps_per_second": 0.924, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 5.861843585968018, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.701, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 3.9136719703674316, + "learning_rate": 2.5e-06, + "loss": 1.6289, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 12.720739364624023, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.6532, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 5.970798492431641, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7035, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 3.490950584411621, + "learning_rate": 2.493907392363932e-06, + "loss": 1.6565, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 5.826146602630615, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.655, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 9.610876083374023, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.7295, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 3.754833221435547, + "learning_rate": 2.487814784727864e-06, + "loss": 1.6406, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.7952232360839844, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.6463, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 6.3178558349609375, + "learning_rate": 2.483753046303818e-06, + "loss": 1.6474, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.8672106266021729, + "eval_runtime": 50.0163, + "eval_samples_per_second": 19.993, + "eval_steps_per_second": 0.84, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 7.4854960441589355, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.6798, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.8903026580810547, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.667, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.481947898864746, + "learning_rate": 2.47766043866775e-06, + "loss": 1.6677, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.8459818363189697, + "learning_rate": 2.475629569455727e-06, + "loss": 1.6129, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 5.00568962097168, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.6233, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 3.7508153915405273, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.6414, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.7895889282226562, + "learning_rate": 2.469536961819659e-06, + "loss": 1.65, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.532818555831909, + "learning_rate": 2.467506092607636e-06, + "loss": 1.6174, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.699639320373535, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7128, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.5636072158813477, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.6416, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.8608473539352417, + "eval_runtime": 43.8315, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.7466976642608643, + "learning_rate": 2.461413484971568e-06, + "loss": 1.7246, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.690990686416626, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.596, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.868340492248535, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.622, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.4456427097320557, + "learning_rate": 2.4553208773355e-06, + "loss": 1.6446, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.5499041080474854, + "learning_rate": 2.453290008123477e-06, + "loss": 1.6587, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.68770170211792, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.6612, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.589557647705078, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.6428, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.7014052867889404, + "learning_rate": 2.447197400487409e-06, + "loss": 1.6605, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.6242196559906006, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.6897, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.5285110473632812, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.6636, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.8544306755065918, + "eval_runtime": 45.6109, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.484952449798584, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.6636, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.5793070793151855, + "learning_rate": 2.439073923639318e-06, + "loss": 1.6044, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.5558605194091797, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.5885, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.6365678310394287, + "learning_rate": 2.435012185215272e-06, + "loss": 1.6236, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.428422451019287, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.6446, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.535776138305664, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.6252, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.743934154510498, + "learning_rate": 2.428919577579204e-06, + "loss": 1.729, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.4823739528656006, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.6251, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.4650003910064697, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.6191, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.4338109493255615, + "learning_rate": 2.422826969943136e-06, + "loss": 1.5919, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.8501240015029907, + "eval_runtime": 43.8558, + "eval_samples_per_second": 22.802, + "eval_steps_per_second": 0.958, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.6054813861846924, + "learning_rate": 2.420796100731113e-06, + "loss": 1.6485, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.390143394470215, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.661, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.467470645904541, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.6308, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.360908269882202, + "learning_rate": 2.414703493095045e-06, + "loss": 1.6394, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.5966176986694336, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.6138, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.528402328491211, + "learning_rate": 2.410641754670999e-06, + "loss": 1.6817, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.5319254398345947, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.6189, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.559260129928589, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.6476, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.5595016479492188, + "learning_rate": 2.404549147034931e-06, + "loss": 1.6878, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.488196849822998, + "learning_rate": 2.402518277822908e-06, + "loss": 1.6095, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.8484610319137573, + "eval_runtime": 43.4829, + "eval_samples_per_second": 22.998, + "eval_steps_per_second": 0.966, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6928117275238037, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.6533, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.485538959503174, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.6174, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.4725208282470703, + "learning_rate": 2.39642567018684e-06, + "loss": 1.6306, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.456028461456299, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.6586, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3731250762939453, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.6009, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.4567086696624756, + "learning_rate": 2.390333062550772e-06, + "loss": 1.5257, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.577141284942627, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7127, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.4612233638763428, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.638, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.463212728500366, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.5933, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.5121495723724365, + "learning_rate": 2.382209585702681e-06, + "loss": 1.68, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.8499811887741089, + "eval_runtime": 45.066, + "eval_samples_per_second": 22.19, + "eval_steps_per_second": 0.932, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.658245801925659, + "learning_rate": 2.380178716490658e-06, + "loss": 1.6354, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.6265509128570557, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.6755, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 3.016807794570923, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.6764, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.6552720069885254, + "learning_rate": 2.37408610885459e-06, + "loss": 1.6281, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.520078659057617, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.7077, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.528329849243164, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.6494, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.623776912689209, + "learning_rate": 2.367993501218522e-06, + "loss": 1.6436, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.56669020652771, + "learning_rate": 2.365962632006499e-06, + "loss": 1.6762, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.5415968894958496, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.6711, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.5164198875427246, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.6725, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.8490949869155884, + "eval_runtime": 43.612, + "eval_samples_per_second": 22.929, + "eval_steps_per_second": 0.963, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.626594066619873, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.6637, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3480069637298584, + "learning_rate": 2.357839155158408e-06, + "loss": 1.6561, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.644103527069092, + "learning_rate": 2.355808285946385e-06, + "loss": 1.6763, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.5539023876190186, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.653, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.5055317878723145, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.6643, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.4204092025756836, + "learning_rate": 2.349715678310317e-06, + "loss": 1.6133, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.5867843627929688, + "learning_rate": 2.347684809098294e-06, + "loss": 1.6804, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.5993685722351074, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.6101, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.565439462661743, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.6221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.5296084880828857, + "learning_rate": 2.341592201462226e-06, + "loss": 1.6234, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.8482695817947388, + "eval_runtime": 45.1972, + "eval_samples_per_second": 22.125, + "eval_steps_per_second": 0.929, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.6142499446868896, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.5886, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.486616849899292, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.6097, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.5778701305389404, + "learning_rate": 2.335499593826158e-06, + "loss": 1.6585, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.6165313720703125, + "learning_rate": 2.333468724614135e-06, + "loss": 1.663, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.4976565837860107, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.6083, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.485017776489258, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.6006, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.620819330215454, + "learning_rate": 2.327376116978067e-06, + "loss": 1.6641, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.528599262237549, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.6128, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.541442632675171, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.6545, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.4722139835357666, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.6146, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.848878264427185, + "eval_runtime": 43.7961, + "eval_samples_per_second": 22.833, + "eval_steps_per_second": 0.959, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.6106903553009033, + "learning_rate": 2.319252640129976e-06, + "loss": 1.7086, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.5255045890808105, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.6738, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.43829083442688, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.5861, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.527411460876465, + "learning_rate": 2.313160032493908e-06, + "loss": 1.6634, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.4998528957366943, + "learning_rate": 2.311129163281885e-06, + "loss": 1.6187, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.4838368892669678, + "learning_rate": 2.309098294069862e-06, + "loss": 1.5999, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.466900110244751, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.6446, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.583103895187378, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.6658, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.478386163711548, + "learning_rate": 2.303005686433794e-06, + "loss": 1.6261, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.5724942684173584, + "learning_rate": 2.300974817221771e-06, + "loss": 1.5955, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.8444501161575317, + "eval_runtime": 50.0263, + "eval_samples_per_second": 19.989, + "eval_steps_per_second": 0.84, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.601644277572632, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.6493, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.5638768672943115, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.6702, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.4802567958831787, + "learning_rate": 2.294882209585703e-06, + "loss": 1.6036, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.5114285945892334, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.642, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.487563133239746, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.673, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.6014463901519775, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.6261, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.53680157661438, + "learning_rate": 2.286758732737612e-06, + "loss": 1.6364, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.764408588409424, + "learning_rate": 2.284727863525589e-06, + "loss": 1.6791, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.8998825550079346, + "learning_rate": 2.282696994313566e-06, + "loss": 1.644, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.5797383785247803, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.6267, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.8467868566513062, + "eval_runtime": 49.4778, + "eval_samples_per_second": 20.211, + "eval_steps_per_second": 0.849, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.447178602218628, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.5955, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.6275408267974854, + "learning_rate": 2.276604386677498e-06, + "loss": 1.6117, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.6952037811279297, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.6894, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.5299718379974365, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.6361, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.837455987930298, + "learning_rate": 2.27051177904143e-06, + "loss": 1.6269, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.5406227111816406, + "learning_rate": 2.268480909829407e-06, + "loss": 1.6538, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.716038227081299, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.6691, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.883439064025879, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.6298, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.7631118297576904, + "learning_rate": 2.262388302193339e-06, + "loss": 1.6245, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.6777117252349854, + "learning_rate": 2.260357432981316e-06, + "loss": 1.6446, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.845224142074585, + "eval_runtime": 46.8476, + "eval_samples_per_second": 21.346, + "eval_steps_per_second": 0.897, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.6280243396759033, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.6398, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.471925973892212, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.6403, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.6293203830718994, + "learning_rate": 2.254264825345248e-06, + "loss": 1.642, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.651144027709961, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.6445, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.6489877700805664, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.6761, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.5869088172912598, + "learning_rate": 2.24817221770918e-06, + "loss": 1.6593, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.411802053451538, + "learning_rate": 2.246141348497157e-06, + "loss": 1.6232, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.522610902786255, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.6358, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.5430991649627686, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.6605, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.4440951347351074, + "learning_rate": 2.240048740861089e-06, + "loss": 1.5354, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.8440407514572144, + "eval_runtime": 48.0239, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.5492775440216064, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.668, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.7511909008026123, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6424, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.5176775455474854, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.6336, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.5408577919006348, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.5843, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.4791393280029297, + "learning_rate": 2.229894394800975e-06, + "loss": 1.555, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.383148670196533, + "learning_rate": 2.227863525588952e-06, + "loss": 1.598, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.429348945617676, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.5943, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.708796501159668, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.6382, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 3.0029661655426025, + "learning_rate": 2.221770917952884e-06, + "loss": 1.6699, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.5410728454589844, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.5965, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.8443207740783691, + "eval_runtime": 47.8671, + "eval_samples_per_second": 20.891, + "eval_steps_per_second": 0.877, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.5718178749084473, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.6214, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.474177837371826, + "learning_rate": 2.215678310316816e-06, + "loss": 1.6498, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.5960052013397217, + "learning_rate": 2.213647441104793e-06, + "loss": 1.6156, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.6217992305755615, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.6311, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.5601158142089844, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.6008, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.7786355018615723, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.62, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.548356771469116, + "learning_rate": 2.205523964256702e-06, + "loss": 1.6473, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.6265289783477783, + "learning_rate": 2.203493095044679e-06, + "loss": 1.6565, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.522379159927368, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.6764, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.5479297637939453, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.6194, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.8434110879898071, + "eval_runtime": 46.6043, + "eval_samples_per_second": 21.457, + "eval_steps_per_second": 0.901, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.5401806831359863, + "learning_rate": 2.197400487408611e-06, + "loss": 1.5911, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.466893434524536, + "learning_rate": 2.195369618196588e-06, + "loss": 1.6145, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.566653251647949, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.6558, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.664053201675415, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7089, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 3.6594645977020264, + "learning_rate": 2.18927701056052e-06, + "loss": 1.6549, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.68450927734375, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.6306, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.7094614505767822, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.6147, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.481332540512085, + "learning_rate": 2.183184402924452e-06, + "loss": 1.6612, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5651051998138428, + "learning_rate": 2.181153533712429e-06, + "loss": 1.6394, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.5570919513702393, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.6162, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.8433003425598145, + "eval_runtime": 46.8673, + "eval_samples_per_second": 21.337, + "eval_steps_per_second": 0.896, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.6206037998199463, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.6869, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.8962907791137695, + "learning_rate": 2.175060926076361e-06, + "loss": 1.7022, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.6211588382720947, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.6249, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.7770566940307617, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.5821, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.543658494949341, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.6587, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.508188486099243, + "learning_rate": 2.16693744922827e-06, + "loss": 1.6114, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.703335762023926, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.6152, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.469688892364502, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.6121, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.4836199283599854, + "learning_rate": 2.160844841592202e-06, + "loss": 1.5918, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.521851062774658, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.6228, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.845065712928772, + "eval_runtime": 45.3003, + "eval_samples_per_second": 22.075, + "eval_steps_per_second": 0.927, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.7808425426483154, + "learning_rate": 2.156783103168156e-06, + "loss": 1.5887, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.696708917617798, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.6517, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.7222893238067627, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.6817, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.504417896270752, + "learning_rate": 2.150690495532088e-06, + "loss": 1.647, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.574836015701294, + "learning_rate": 2.148659626320065e-06, + "loss": 1.6554, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.5004355907440186, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.6089, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.6979293823242188, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.6226, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.5239524841308594, + "learning_rate": 2.142567018683997e-06, + "loss": 1.6875, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.5218467712402344, + "learning_rate": 2.140536149471974e-06, + "loss": 1.6095, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.5778310298919678, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.6155, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.8443745374679565, + "eval_runtime": 46.6273, + "eval_samples_per_second": 21.447, + "eval_steps_per_second": 0.901, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.671973705291748, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.6696, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.5520615577697754, + "learning_rate": 2.134443541835906e-06, + "loss": 1.6291, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.39286208152771, + "learning_rate": 2.132412672623883e-06, + "loss": 1.6012, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.8277461528778076, + "learning_rate": 2.13038180341186e-06, + "loss": 1.6398, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.7147881984710693, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.6685, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.499044418334961, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.6128, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.514784574508667, + "learning_rate": 2.124289195775792e-06, + "loss": 1.6263, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.505342483520508, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.5532, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.608139753341675, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.5946, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.5578057765960693, + "learning_rate": 2.118196588139724e-06, + "loss": 1.6103, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.8425958156585693, + "eval_runtime": 44.206, + "eval_samples_per_second": 22.621, + "eval_steps_per_second": 0.95, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.4881014823913574, + "learning_rate": 2.116165718927701e-06, + "loss": 1.6364, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.4141550064086914, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.5943, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.649827718734741, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.5994, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.547238826751709, + "learning_rate": 2.110073111291633e-06, + "loss": 1.6247, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.6515395641326904, + "learning_rate": 2.10804224207961e-06, + "loss": 1.6543, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.5575809478759766, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.6618, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.466344118118286, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.5437, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4897878170013428, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6199, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.6578829288482666, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.5989, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.6173501014709473, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.6392, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.838721752166748, + "eval_runtime": 49.4357, + "eval_samples_per_second": 20.228, + "eval_steps_per_second": 0.85, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.5655064582824707, + "learning_rate": 2.095857026807474e-06, + "loss": 1.6276, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.685790777206421, + "learning_rate": 2.093826157595451e-06, + "loss": 1.6142, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.7278990745544434, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.6292, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.5769667625427246, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.6774, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.5120608806610107, + "learning_rate": 2.087733549959383e-06, + "loss": 1.5972, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.4658212661743164, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.5683, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.526524782180786, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.6687, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.3989250659942627, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.6184, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.589115858078003, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.6153, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.4985275268554688, + "learning_rate": 2.077579203899269e-06, + "loss": 1.5682, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.8374594449996948, + "eval_runtime": 47.0329, + "eval_samples_per_second": 21.262, + "eval_steps_per_second": 0.893, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5770835876464844, + "learning_rate": 2.075548334687246e-06, + "loss": 1.6437, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.5111560821533203, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.5711, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.6110589504241943, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.6125, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.6229705810546875, + "learning_rate": 2.069455727051178e-06, + "loss": 1.595, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.648716926574707, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.5908, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.673701763153076, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.6652, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.667478561401367, + "learning_rate": 2.06336311941511e-06, + "loss": 1.6088, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 3.5688586235046387, + "learning_rate": 2.061332250203087e-06, + "loss": 1.5852, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.512603282928467, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.6218, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.5296075344085693, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.6382, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.8406264781951904, + "eval_runtime": 43.8351, + "eval_samples_per_second": 22.813, + "eval_steps_per_second": 0.958, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.598994255065918, + "learning_rate": 2.055239642567019e-06, + "loss": 1.6263, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.93025279045105, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.6502, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 3.3655412197113037, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.5872, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.955120325088501, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.6404, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.5226638317108154, + "learning_rate": 2.047116165718928e-06, + "loss": 1.6102, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.5462093353271484, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.6757, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 4.066190242767334, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.6072, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.8548827171325684, + "learning_rate": 2.04102355808286e-06, + "loss": 1.5717, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.6206634044647217, + "learning_rate": 2.038992688870837e-06, + "loss": 1.5943, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.8793396949768066, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.6121, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.83925199508667, + "eval_runtime": 44.747, + "eval_samples_per_second": 22.348, + "eval_steps_per_second": 0.939, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.6215157508850098, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.6492, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.6103456020355225, + "learning_rate": 2.032900081234769e-06, + "loss": 1.6531, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.485775947570801, + "learning_rate": 2.030869212022746e-06, + "loss": 1.6086, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.399557113647461, + "learning_rate": 2.028838342810723e-06, + "loss": 1.6126, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.4711368083953857, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.5985, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.654580593109131, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.6557, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.5091657638549805, + "learning_rate": 2.022745735174655e-06, + "loss": 1.6181, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.5723490715026855, + "learning_rate": 2.020714865962632e-06, + "loss": 1.6421, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4211535453796387, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.6226, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.5550055503845215, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.5729, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.8397995233535767, + "eval_runtime": 44.4762, + "eval_samples_per_second": 22.484, + "eval_steps_per_second": 0.944, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.618758201599121, + "learning_rate": 2.014622258326564e-06, + "loss": 1.6356, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.6717913150787354, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.6246, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.5464835166931152, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.6027, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.557669162750244, + "learning_rate": 2.008529650690496e-06, + "loss": 1.6531, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.581970453262329, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.6076, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.609236478805542, + "learning_rate": 2.00446791226645e-06, + "loss": 1.5932, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.4249160289764404, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.6372, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.5093257427215576, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.6627, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.3989412784576416, + "learning_rate": 1.998375304630382e-06, + "loss": 1.6292, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4299519062042236, + "learning_rate": 1.996344435418359e-06, + "loss": 1.6385, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.8367400169372559, + "eval_runtime": 45.921, + "eval_samples_per_second": 21.777, + "eval_steps_per_second": 0.915, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.509530782699585, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.6441, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.591426134109497, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.6382, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.6569297313690186, + "learning_rate": 1.990251827782291e-06, + "loss": 1.6469, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.6453206539154053, + "learning_rate": 1.988220958570268e-06, + "loss": 1.6348, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.5522994995117188, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.6311, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.5981009006500244, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.585, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.617323637008667, + "learning_rate": 1.9821283509342e-06, + "loss": 1.584, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.6743686199188232, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.6334, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.418383836746216, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.5934, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.7952582836151123, + "learning_rate": 1.976035743298132e-06, + "loss": 1.6779, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.8379877805709839, + "eval_runtime": 46.545, + "eval_samples_per_second": 21.485, + "eval_steps_per_second": 0.902, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.5364768505096436, + "learning_rate": 1.974004874086109e-06, + "loss": 1.5802, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.446056365966797, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6141, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.527529239654541, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.6042, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.5953001976013184, + "learning_rate": 1.967912266450041e-06, + "loss": 1.5959, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.545994758605957, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.7091, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.3900482654571533, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.5946, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.51353120803833, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.6166, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.550389528274536, + "learning_rate": 1.95978878960195e-06, + "loss": 1.6817, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.6815762519836426, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.6594, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.591909885406494, + "learning_rate": 1.955727051177904e-06, + "loss": 1.5991, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.8390737771987915, + "eval_runtime": 46.8495, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.5111982822418213, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.5993, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.655891180038452, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.5915, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.471090316772461, + "learning_rate": 1.949634443541836e-06, + "loss": 1.5867, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.8599209785461426, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.6485, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4986093044281006, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.6084, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.6345789432525635, + "learning_rate": 1.943541835905768e-06, + "loss": 1.6719, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.5636799335479736, + "learning_rate": 1.941510966693745e-06, + "loss": 1.5817, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.5867199897766113, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.6498, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.6369481086730957, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.6791, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.4554147720336914, + "learning_rate": 1.935418359057677e-06, + "loss": 1.6319, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.8408559560775757, + "eval_runtime": 49.3515, + "eval_samples_per_second": 20.263, + "eval_steps_per_second": 0.851, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.5412509441375732, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.6673, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.5755672454833984, + "learning_rate": 1.931356620633631e-06, + "loss": 1.5636, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.602530002593994, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.6261, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.755247116088867, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.5817, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.528991460800171, + "learning_rate": 1.925264012997563e-06, + "loss": 1.6173, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.473057508468628, + "learning_rate": 1.92323314378554e-06, + "loss": 1.648, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.62630033493042, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.5864, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.7876687049865723, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.5804, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.491473436355591, + "learning_rate": 1.917140536149472e-06, + "loss": 1.5716, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.656989336013794, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.6595, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.839701771736145, + "eval_runtime": 44.5976, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.6800942420959473, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.6415, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.4725751876831055, + "learning_rate": 1.911047928513404e-06, + "loss": 1.5901, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.552624464035034, + "learning_rate": 1.909017059301381e-06, + "loss": 1.5956, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.6042184829711914, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.6328, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.5159223079681396, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.604, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.488041639328003, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.5539, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.659151077270508, + "learning_rate": 1.90089358245329e-06, + "loss": 1.5999, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.700378179550171, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.6485, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.3943607807159424, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.5736, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.8171095848083496, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6259, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.8379307985305786, + "eval_runtime": 48.6918, + "eval_samples_per_second": 20.537, + "eval_steps_per_second": 0.863, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.562655210494995, + "learning_rate": 1.892770105605199e-06, + "loss": 1.6439, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.5640337467193604, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.6113, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.5324809551239014, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.6331, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.393224000930786, + "learning_rate": 1.886677497969131e-06, + "loss": 1.5228, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 3.073945999145508, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.632, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.7066094875335693, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.5508, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.981933832168579, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6973, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.903444528579712, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.6036, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.8616178035736084, + "learning_rate": 1.876523151909017e-06, + "loss": 1.5723, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.8243541717529297, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.555, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.8437565565109253, + "eval_runtime": 48.5449, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.803142547607422, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.5861, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 3.146677017211914, + "learning_rate": 1.870430544272949e-06, + "loss": 1.6791, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 3.2554781436920166, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.5626, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.955395460128784, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.5893, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.941377878189087, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.5962, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.765620708465576, + "learning_rate": 1.862307067424858e-06, + "loss": 1.595, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.8529839515686035, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.6296, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.7770590782165527, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.5711, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.827406883239746, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.5966, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.927119016647339, + "learning_rate": 1.854183590576767e-06, + "loss": 1.6463, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.841707468032837, + "eval_runtime": 49.0965, + "eval_samples_per_second": 20.368, + "eval_steps_per_second": 0.855, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.8223392963409424, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.559, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.862473726272583, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.5189, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.9060091972351074, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.5631, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.8727498054504395, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6013, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.862298011779785, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.549, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.8221330642700195, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.6341, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.938336133956909, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.602, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.719489812850952, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.5626, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.6615874767303467, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.5535, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.835413932800293, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.5742, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.8375873565673828, + "eval_runtime": 45.731, + "eval_samples_per_second": 21.867, + "eval_steps_per_second": 0.918, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.773707389831543, + "learning_rate": 1.831844029244517e-06, + "loss": 1.5499, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 3.0959930419921875, + "learning_rate": 1.829813160032494e-06, + "loss": 1.5656, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.8521792888641357, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.5982, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.970618486404419, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.5731, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 3.0716512203216553, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.5824, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.8776299953460693, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.6283, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.72183895111084, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.5422, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.853234052658081, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.5533, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.7193424701690674, + "learning_rate": 1.815597075548335e-06, + "loss": 1.5923, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.9768450260162354, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.6277, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.8401803970336914, + "eval_runtime": 45.8062, + "eval_samples_per_second": 21.831, + "eval_steps_per_second": 0.917, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.752772569656372, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.5241, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.784980297088623, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.5657, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.7485857009887695, + "learning_rate": 1.807473598700244e-06, + "loss": 1.5501, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.7191033363342285, + "learning_rate": 1.805442729488221e-06, + "loss": 1.5718, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.697540044784546, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.5804, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.7175710201263428, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.5809, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.764824151992798, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.545, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.946669816970825, + "learning_rate": 1.79731925264013e-06, + "loss": 1.595, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.689326524734497, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.5676, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.759997606277466, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.6032, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.8379274606704712, + "eval_runtime": 45.5941, + "eval_samples_per_second": 21.933, + "eval_steps_per_second": 0.921, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.8656060695648193, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.6086, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.7155873775482178, + "learning_rate": 1.789195775792039e-06, + "loss": 1.5218, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.8487389087677, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6285, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.8070144653320312, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.5479, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.856282949447632, + "learning_rate": 1.783103168155971e-06, + "loss": 1.5689, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.7465922832489014, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.5733, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.824997901916504, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.5583, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.8030786514282227, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.5383, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.911377429962158, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.586, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.7452991008758545, + "learning_rate": 1.772948822095857e-06, + "loss": 1.6409, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.8399248123168945, + "eval_runtime": 46.1241, + "eval_samples_per_second": 21.681, + "eval_steps_per_second": 0.911, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.8255481719970703, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.6134, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.865100622177124, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.5791, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.740133047103882, + "learning_rate": 1.766856214459789e-06, + "loss": 1.5651, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.7219033241271973, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.5453, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.7266135215759277, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.6375, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.8476688861846924, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.5877, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.810105085372925, + "learning_rate": 1.758732737611698e-06, + "loss": 1.5796, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.78798246383667, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.5708, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.750525951385498, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.5567, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.818673849105835, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.5661, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.8374614715576172, + "eval_runtime": 43.5618, + "eval_samples_per_second": 22.956, + "eval_steps_per_second": 0.964, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.7967586517333984, + "learning_rate": 1.750609260763607e-06, + "loss": 1.5846, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.766272783279419, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.5721, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.8006162643432617, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.5261, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 3.714587688446045, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.6599, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 3.143052339553833, + "learning_rate": 1.742485783915516e-06, + "loss": 1.659, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.8256232738494873, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.5076, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.8843302726745605, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.5862, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 6.692182540893555, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.6144, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.928053617477417, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.571, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.8071024417877197, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.5861, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.837381362915039, + "eval_runtime": 44.7089, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 3.070399522781372, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.5636, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 3.298100709915161, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.5717, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 5.891775131225586, + "learning_rate": 1.726238830219334e-06, + "loss": 1.5757, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.838676691055298, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.6369, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.8816466331481934, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.5885, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 3.1703031063079834, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.5966, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 3.504777669906616, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.4932, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 3.1451492309570312, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.5071, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.7158446311950684, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.5725, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.8391170501708984, + "learning_rate": 1.712022745735175e-06, + "loss": 1.5742, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.8371202945709229, + "eval_runtime": 43.2464, + "eval_samples_per_second": 23.123, + "eval_steps_per_second": 0.971, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.8346052169799805, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.5808, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.767813205718994, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.5209, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.7955846786499023, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.581, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.8616271018981934, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.5758, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.7975544929504395, + "learning_rate": 1.701868399675061e-06, + "loss": 1.5912, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 3.095912456512451, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.6522, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 3.102691888809204, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.6004, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.8405165672302246, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.5649, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.8038063049316406, + "learning_rate": 1.69374492282697e-06, + "loss": 1.545, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.8661468029022217, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6297, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.8386919498443604, + "eval_runtime": 46.0334, + "eval_samples_per_second": 21.723, + "eval_steps_per_second": 0.912, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.8217391967773438, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.5694, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 3.099724769592285, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.5986, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.9951272010803223, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.5429, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.9366989135742188, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.5666, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 3.0171356201171875, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.6212, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.952259063720703, + "learning_rate": 1.679528838342811e-06, + "loss": 1.5611, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.667905330657959, + "learning_rate": 1.677497969130788e-06, + "loss": 1.5556, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.8001160621643066, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.5854, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.6954760551452637, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.5278, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.8026583194732666, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.5103, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.8384881019592285, + "eval_runtime": 46.1899, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.898869752883911, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6188, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 3.0432069301605225, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.5945, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.7744672298431396, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.5642, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.8844289779663086, + "learning_rate": 1.663281884646629e-06, + "loss": 1.5657, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 3.0022218227386475, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.6297, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.918320417404175, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.5928, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.913848876953125, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.6219, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.954702615737915, + "learning_rate": 1.655158407798538e-06, + "loss": 1.5812, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.7966420650482178, + "learning_rate": 1.653127538586515e-06, + "loss": 1.6034, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 3.1328368186950684, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.6168, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.836020827293396, + "eval_runtime": 46.1979, + "eval_samples_per_second": 21.646, + "eval_steps_per_second": 0.909, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.8109259605407715, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.6091, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.823570728302002, + "learning_rate": 1.647034930950447e-06, + "loss": 1.5504, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.885016918182373, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.5784, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 3.612942934036255, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.6014, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.8981637954711914, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.5552, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.6703410148620605, + "learning_rate": 1.638911454102356e-06, + "loss": 1.5567, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.7087886333465576, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.613, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.7711687088012695, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.5586, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.8473854064941406, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.565, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 3.783824920654297, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.5909, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.8384798765182495, + "eval_runtime": 46.2339, + "eval_samples_per_second": 21.629, + "eval_steps_per_second": 0.908, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.9971532821655273, + "learning_rate": 1.628757108042242e-06, + "loss": 1.6251, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.8581976890563965, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.5632, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.9119839668273926, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.5806, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.840101718902588, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.6706, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.8574070930480957, + "learning_rate": 1.620633631194151e-06, + "loss": 1.6138, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.820692539215088, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.5534, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7690017223358154, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.5922, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 3.062229871749878, + "learning_rate": 1.614541023558083e-06, + "loss": 1.5893, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.727062225341797, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.5851, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.7693302631378174, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.6031, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.8332058191299438, + "eval_runtime": 45.0301, + "eval_samples_per_second": 22.207, + "eval_steps_per_second": 0.933, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.803758382797241, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.5504, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.775770902633667, + "learning_rate": 1.606417546709992e-06, + "loss": 1.5707, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.7698636054992676, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.5763, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.739940881729126, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.5386, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.7594099044799805, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.5805, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.853273391723633, + "learning_rate": 1.598294069861901e-06, + "loss": 1.5548, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.9859204292297363, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6278, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.742898941040039, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.555, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.769071340560913, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.5504, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.7191832065582275, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6014, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.8336049318313599, + "eval_runtime": 47.0078, + "eval_samples_per_second": 21.273, + "eval_steps_per_second": 0.893, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.842529773712158, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.5723, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.768066883087158, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.5272, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7481656074523926, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.5843, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.806687593460083, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.5598, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.8214614391326904, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.5972, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.8541247844696045, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.5988, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 3.000246286392212, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.583, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.9164586067199707, + "learning_rate": 1.573923639317628e-06, + "loss": 1.5949, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.8099935054779053, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.5959, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.88322377204895, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.6062, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.832599401473999, + "eval_runtime": 45.9576, + "eval_samples_per_second": 21.759, + "eval_steps_per_second": 0.914, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.8019371032714844, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.5445, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.8585665225982666, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6079, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.7423956394195557, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.5537, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.7420766353607178, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.5846, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.8864388465881348, + "learning_rate": 1.559707554833469e-06, + "loss": 1.6213, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.7483463287353516, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.5012, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.7340080738067627, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.555, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.8000667095184326, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.5259, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.9176089763641357, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.5726, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.9147658348083496, + "learning_rate": 1.549553208773355e-06, + "loss": 1.5685, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.834657907485962, + "eval_runtime": 45.8709, + "eval_samples_per_second": 21.8, + "eval_steps_per_second": 0.916, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.9109392166137695, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.5533, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 3.0263078212738037, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.6277, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.9347317218780518, + "learning_rate": 1.543460601137287e-06, + "loss": 1.5738, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.8229548931121826, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.5969, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.906954288482666, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.5772, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.9214975833892822, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.5147, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.844081163406372, + "learning_rate": 1.535337124289196e-06, + "loss": 1.5635, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 3.0214555263519287, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.6466, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.9254050254821777, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.5837, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.8986024856567383, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.5997, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.8318188190460205, + "eval_runtime": 46.7695, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 2000 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 6.23461881872384e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2000/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/latest b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/latest new file mode 100644 index 0000000000000000000000000000000000000000..adbdf3f70e1e0db88a47bff3edcb0592ed0a2e61 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/latest @@ -0,0 +1 @@ +global_step2200 \ No newline at end of file diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2200/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/generation_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/latest b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/latest new file mode 100644 index 0000000000000000000000000000000000000000..78dc01ce4ce1498b753887dfb7af92409705a69b --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/latest @@ -0,0 +1 @@ +global_step2300 \ No newline at end of file diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/rng_state_0.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..cf9f7cc0f22e38cdfea7c36a09f25ba146158728 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b93ade60213134037a6e2a03060d84a67c5041245aff2ad374a5c97f337ee4c2 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/rng_state_1.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..22044561af392bd37b8a514ffc971741f1a025c2 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e17eca23ec632d21ab0115d8db1a9360104fa7b08d928954eee2eed9adf46b8a +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..24228040a3cbfaaf0d0e62b9e12dbef736514695 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6614221a631cfb519fe3c961fa724d0b1c62628b10e8afa55fbb30d5380c88e3 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/scheduler.pt b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..7355043f7da4b08f5c1e48c2ed2c78ec18824d26 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7153ac187458a95dd3da74cffb0bbaa448f03e890a0ceddddf306a5ac055b012 +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/special_tokens_map.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..5cd92fe00e49b65af3365e1e6a245ad985ec3cb0 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/trainer_state.json @@ -0,0 +1,17973 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.5205479452054793, + "eval_steps": 10, + "global_step": 2300, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.5818793773651123, + "learning_rate": 4.569455727051178e-06, + "loss": 1.85, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 3.0734691619873047, + "learning_rate": 4.567424857839156e-06, + "loss": 1.8682, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.496253252029419, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9007, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.360891342163086, + "learning_rate": 4.5633631194151105e-06, + "loss": 1.8262, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.5731117725372314, + "learning_rate": 4.561332250203087e-06, + "loss": 1.8369, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.700558662414551, + "learning_rate": 4.559301380991065e-06, + "loss": 1.7825, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.561306953430176, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.8174, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.618889570236206, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.8812, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.4671308994293213, + "learning_rate": 4.553208773354996e-06, + "loss": 1.7959, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.4571878910064697, + "learning_rate": 4.551177904142974e-06, + "loss": 1.8943, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.010777711868286, + "eval_runtime": 43.7424, + "eval_samples_per_second": 22.861, + "eval_steps_per_second": 0.96, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.4786667823791504, + "learning_rate": 4.5491470349309505e-06, + "loss": 1.8147, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.2955000400543213, + "learning_rate": 4.547116165718929e-06, + "loss": 1.8547, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.3873913288116455, + "learning_rate": 4.545085296506905e-06, + "loss": 1.8156, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.2449734210968018, + "learning_rate": 4.543054427294882e-06, + "loss": 1.8053, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.3969857692718506, + "learning_rate": 4.54102355808286e-06, + "loss": 1.8085, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.3446357250213623, + "learning_rate": 4.538992688870837e-06, + "loss": 1.8355, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.408762216567993, + "learning_rate": 4.536961819658814e-06, + "loss": 1.8319, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.362962245941162, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.8422, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.402984857559204, + "learning_rate": 4.532900081234769e-06, + "loss": 1.8125, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.3964452743530273, + "learning_rate": 4.530869212022746e-06, + "loss": 1.8417, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.006908416748047, + "eval_runtime": 44.2701, + "eval_samples_per_second": 22.589, + "eval_steps_per_second": 0.949, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.595233201980591, + "learning_rate": 4.528838342810723e-06, + "loss": 1.8275, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.326117753982544, + "learning_rate": 4.526807473598701e-06, + "loss": 1.8405, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.4904465675354004, + "learning_rate": 4.524776604386678e-06, + "loss": 1.8172, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.4591610431671143, + "learning_rate": 4.522745735174655e-06, + "loss": 1.876, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.319800853729248, + "learning_rate": 4.520714865962632e-06, + "loss": 1.8294, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.28790020942688, + "learning_rate": 4.51868399675061e-06, + "loss": 1.8537, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.284822702407837, + "learning_rate": 4.516653127538587e-06, + "loss": 1.8087, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.273820638656616, + "learning_rate": 4.514622258326564e-06, + "loss": 1.7992, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.2025654315948486, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.8248, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 2.3218843936920166, + "learning_rate": 4.510560519902519e-06, + "loss": 1.8647, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.0048956871032715, + "eval_runtime": 44.5972, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.515766143798828, + "learning_rate": 4.508529650690496e-06, + "loss": 1.7673, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.4376487731933594, + "learning_rate": 4.506498781478473e-06, + "loss": 1.7755, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.2801272869110107, + "learning_rate": 4.504467912266451e-06, + "loss": 1.8273, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.4707891941070557, + "learning_rate": 4.502437043054428e-06, + "loss": 1.8971, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.301301956176758, + "learning_rate": 4.500406173842405e-06, + "loss": 1.8345, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.2317373752593994, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.8248, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.447373867034912, + "learning_rate": 4.49634443541836e-06, + "loss": 1.8148, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.337245225906372, + "learning_rate": 4.494313566206336e-06, + "loss": 1.8811, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3717031478881836, + "learning_rate": 4.492282696994314e-06, + "loss": 1.8311, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.287677526473999, + "learning_rate": 4.490251827782291e-06, + "loss": 1.7728, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.004889965057373, + "eval_runtime": 43.8132, + "eval_samples_per_second": 22.824, + "eval_steps_per_second": 0.959, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.441445827484131, + "learning_rate": 4.488220958570269e-06, + "loss": 1.7927, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.3191535472869873, + "learning_rate": 4.486190089358245e-06, + "loss": 1.759, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.3118820190429688, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8509, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.4711859226226807, + "learning_rate": 4.4821283509342e-06, + "loss": 1.8402, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.3142173290252686, + "learning_rate": 4.480097481722178e-06, + "loss": 1.8238, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.2665774822235107, + "learning_rate": 4.478066612510154e-06, + "loss": 1.7826, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.1692490577697754, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.8138, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.412783145904541, + "learning_rate": 4.474004874086109e-06, + "loss": 1.9421, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3692686557769775, + "learning_rate": 4.471974004874087e-06, + "loss": 1.7722, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.3224151134490967, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.8753, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.0016305446624756, + "eval_runtime": 45.564, + "eval_samples_per_second": 21.947, + "eval_steps_per_second": 0.922, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.51043438911438, + "learning_rate": 4.467912266450041e-06, + "loss": 1.8325, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.4568657875061035, + "learning_rate": 4.465881397238018e-06, + "loss": 1.8081, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.3279001712799072, + "learning_rate": 4.463850528025995e-06, + "loss": 1.8439, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.6041171550750732, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.8623, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.3780531883239746, + "learning_rate": 4.45978878960195e-06, + "loss": 1.8269, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.511500597000122, + "learning_rate": 4.457757920389927e-06, + "loss": 1.7749, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.707789659500122, + "learning_rate": 4.455727051177904e-06, + "loss": 1.8157, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.5495495796203613, + "learning_rate": 4.453696181965882e-06, + "loss": 1.8534, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.255876302719116, + "learning_rate": 4.451665312753859e-06, + "loss": 1.8377, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.415097951889038, + "learning_rate": 4.449634443541836e-06, + "loss": 1.8764, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 1.9943474531173706, + "eval_runtime": 42.2985, + "eval_samples_per_second": 23.642, + "eval_steps_per_second": 0.993, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.2604713439941406, + "learning_rate": 4.4476035743298135e-06, + "loss": 1.7898, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.5643842220306396, + "learning_rate": 4.445572705117791e-06, + "loss": 1.7953, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.412548780441284, + "learning_rate": 4.443541835905768e-06, + "loss": 1.8611, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.5423877239227295, + "learning_rate": 4.441510966693745e-06, + "loss": 1.8049, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.4986631870269775, + "learning_rate": 4.439480097481723e-06, + "loss": 1.8759, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.8349273204803467, + "learning_rate": 4.4374492282697e-06, + "loss": 1.7416, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.452017068862915, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8268, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.6425273418426514, + "learning_rate": 4.433387489845654e-06, + "loss": 1.8334, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.328740358352661, + "learning_rate": 4.431356620633632e-06, + "loss": 1.7931, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.262209415435791, + "learning_rate": 4.429325751421609e-06, + "loss": 1.8151, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 1.9917123317718506, + "eval_runtime": 42.4309, + "eval_samples_per_second": 23.568, + "eval_steps_per_second": 0.99, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.2984049320220947, + "learning_rate": 4.427294882209586e-06, + "loss": 1.8187, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.3910019397735596, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.7512, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.4494471549987793, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8202, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.3857171535491943, + "learning_rate": 4.421202274573518e-06, + "loss": 1.8279, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.4526500701904297, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.7136, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.2216501235961914, + "learning_rate": 4.417140536149473e-06, + "loss": 1.7574, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.7639758586883545, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8403, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.3177521228790283, + "learning_rate": 4.413078797725427e-06, + "loss": 1.7898, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.527902364730835, + "learning_rate": 4.411047928513404e-06, + "loss": 1.8184, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.34061336517334, + "learning_rate": 4.409017059301382e-06, + "loss": 1.811, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 1.9882216453552246, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.6462934017181396, + "learning_rate": 4.406986190089358e-06, + "loss": 1.8501, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.2725913524627686, + "learning_rate": 4.404955320877336e-06, + "loss": 1.8262, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.390084981918335, + "learning_rate": 4.402924451665313e-06, + "loss": 1.7842, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.314305305480957, + "learning_rate": 4.400893582453291e-06, + "loss": 1.7692, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.2828257083892822, + "learning_rate": 4.398862713241267e-06, + "loss": 1.8031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.3493599891662598, + "learning_rate": 4.396831844029245e-06, + "loss": 1.8183, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.423365831375122, + "learning_rate": 4.394800974817222e-06, + "loss": 1.8012, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.438015937805176, + "learning_rate": 4.392770105605199e-06, + "loss": 1.8217, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.357172727584839, + "learning_rate": 4.390739236393176e-06, + "loss": 1.8382, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.3383474349975586, + "learning_rate": 4.388708367181154e-06, + "loss": 1.841, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 1.9878787994384766, + "eval_runtime": 44.0694, + "eval_samples_per_second": 22.691, + "eval_steps_per_second": 0.953, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.459385633468628, + "learning_rate": 4.386677497969131e-06, + "loss": 1.8377, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.356788158416748, + "learning_rate": 4.384646628757108e-06, + "loss": 1.8718, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.3217222690582275, + "learning_rate": 4.3826157595450855e-06, + "loss": 1.8073, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.3134331703186035, + "learning_rate": 4.380584890333063e-06, + "loss": 1.7675, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.3273653984069824, + "learning_rate": 4.37855402112104e-06, + "loss": 1.7169, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.611034393310547, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8702, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.2077300548553467, + "learning_rate": 4.3744922826969946e-06, + "loss": 1.7951, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.3393311500549316, + "learning_rate": 4.372461413484972e-06, + "loss": 1.8136, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.3306660652160645, + "learning_rate": 4.370430544272949e-06, + "loss": 1.8372, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.4440250396728516, + "learning_rate": 4.368399675060926e-06, + "loss": 1.844, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 1.98439621925354, + "eval_runtime": 44.4595, + "eval_samples_per_second": 22.492, + "eval_steps_per_second": 0.945, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.205542802810669, + "learning_rate": 4.366368805848904e-06, + "loss": 1.8182, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.6030094623565674, + "learning_rate": 4.364337936636881e-06, + "loss": 1.7524, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.4000444412231445, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8269, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.319526433944702, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.7722, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.1624808311462402, + "learning_rate": 4.358245329000813e-06, + "loss": 1.8081, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.26615309715271, + "learning_rate": 4.35621445978879e-06, + "loss": 1.8049, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.4895036220550537, + "learning_rate": 4.354183590576767e-06, + "loss": 1.7932, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.437161445617676, + "learning_rate": 4.352152721364745e-06, + "loss": 1.772, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.344069480895996, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8213, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.360734224319458, + "learning_rate": 4.348090982940699e-06, + "loss": 1.8268, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 1.9846479892730713, + "eval_runtime": 44.947, + "eval_samples_per_second": 22.248, + "eval_steps_per_second": 0.934, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.59487247467041, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.8303, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.6512107849121094, + "learning_rate": 4.344029244516654e-06, + "loss": 1.7979, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.2900476455688477, + "learning_rate": 4.341998375304631e-06, + "loss": 1.8707, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.409494638442993, + "learning_rate": 4.339967506092607e-06, + "loss": 1.8075, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.527205467224121, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.7884, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.33461332321167, + "learning_rate": 4.335905767668562e-06, + "loss": 1.7839, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.3555047512054443, + "learning_rate": 4.33387489845654e-06, + "loss": 1.8245, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.397148609161377, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.8115, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.232139825820923, + "learning_rate": 4.329813160032495e-06, + "loss": 1.7747, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.1684389114379883, + "learning_rate": 4.327782290820471e-06, + "loss": 1.748, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 1.980297565460205, + "eval_runtime": 42.9183, + "eval_samples_per_second": 23.3, + "eval_steps_per_second": 0.979, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.2454638481140137, + "learning_rate": 4.325751421608449e-06, + "loss": 1.83, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.3264827728271484, + "learning_rate": 4.323720552396426e-06, + "loss": 1.7856, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.298804998397827, + "learning_rate": 4.321689683184404e-06, + "loss": 1.78, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2202727794647217, + "learning_rate": 4.31965881397238e-06, + "loss": 1.7875, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.494479179382324, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.8363, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3520524501800537, + "learning_rate": 4.315597075548335e-06, + "loss": 1.8266, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.373878002166748, + "learning_rate": 4.313566206336312e-06, + "loss": 1.8348, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.4232168197631836, + "learning_rate": 4.311535337124289e-06, + "loss": 1.77, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.2572121620178223, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.7914, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.5817952156066895, + "learning_rate": 4.307473598700244e-06, + "loss": 1.8099, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 1.9784427881240845, + "eval_runtime": 43.2981, + "eval_samples_per_second": 23.096, + "eval_steps_per_second": 0.97, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.646040916442871, + "learning_rate": 4.305442729488221e-06, + "loss": 1.7452, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.370220184326172, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8399, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.4599130153656006, + "learning_rate": 4.301380991064176e-06, + "loss": 1.8273, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 2.4135093688964844, + "learning_rate": 4.299350121852153e-06, + "loss": 1.852, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.371021270751953, + "learning_rate": 4.29731925264013e-06, + "loss": 1.8339, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.6858229637145996, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.792, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.4101266860961914, + "learning_rate": 4.293257514216085e-06, + "loss": 1.7556, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.769761800765991, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8223, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.5777804851531982, + "learning_rate": 4.289195775792039e-06, + "loss": 1.8406, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.376864433288574, + "learning_rate": 4.287164906580017e-06, + "loss": 1.7846, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 1.9758412837982178, + "eval_runtime": 43.9353, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.7461960315704346, + "learning_rate": 4.285134037367994e-06, + "loss": 1.776, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.549576759338379, + "learning_rate": 4.283103168155971e-06, + "loss": 1.7644, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.568074941635132, + "learning_rate": 4.281072298943948e-06, + "loss": 1.7571, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.8195722103118896, + "learning_rate": 4.279041429731926e-06, + "loss": 1.7959, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.471623182296753, + "learning_rate": 4.277010560519903e-06, + "loss": 1.7584, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.664689302444458, + "learning_rate": 4.27497969130788e-06, + "loss": 1.7909, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.684504985809326, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.8205, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.5247528553009033, + "learning_rate": 4.270917952883835e-06, + "loss": 1.8365, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.757972478866577, + "learning_rate": 4.268887083671812e-06, + "loss": 1.7665, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.6032614707946777, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8702, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 1.9716291427612305, + "eval_runtime": 43.5482, + "eval_samples_per_second": 22.963, + "eval_steps_per_second": 0.964, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.5324370861053467, + "learning_rate": 4.264825345247766e-06, + "loss": 1.738, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.730712890625, + "learning_rate": 4.262794476035744e-06, + "loss": 1.7918, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.5554966926574707, + "learning_rate": 4.26076360682372e-06, + "loss": 1.7218, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.4774436950683594, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.7823, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.77431058883667, + "learning_rate": 4.256701868399675e-06, + "loss": 1.794, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.258828639984131, + "learning_rate": 4.254670999187653e-06, + "loss": 1.805, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.384399175643921, + "learning_rate": 4.252640129975629e-06, + "loss": 1.818, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3278932571411133, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.7719, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.207348108291626, + "learning_rate": 4.248578391551584e-06, + "loss": 1.7362, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.4525978565216064, + "learning_rate": 4.246547522339562e-06, + "loss": 1.778, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 1.9694374799728394, + "eval_runtime": 42.5977, + "eval_samples_per_second": 23.475, + "eval_steps_per_second": 0.986, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.377835273742676, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.7926, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.4623866081237793, + "learning_rate": 4.242485783915517e-06, + "loss": 1.7366, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.542219400405884, + "learning_rate": 4.240454914703493e-06, + "loss": 1.7409, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.1955819129943848, + "learning_rate": 4.23842404549147e-06, + "loss": 1.7596, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.5834269523620605, + "learning_rate": 4.236393176279448e-06, + "loss": 1.7343, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.219660520553589, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8089, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.283200740814209, + "learning_rate": 4.232331437855402e-06, + "loss": 1.8248, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.274127244949341, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.825, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.3550686836242676, + "learning_rate": 4.228269699431357e-06, + "loss": 1.7957, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.588027238845825, + "learning_rate": 4.226238830219334e-06, + "loss": 1.7865, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 1.9662212133407593, + "eval_runtime": 43.2661, + "eval_samples_per_second": 23.113, + "eval_steps_per_second": 0.971, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.2445170879364014, + "learning_rate": 4.224207961007311e-06, + "loss": 1.791, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.399174690246582, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.7944, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.586184501647949, + "learning_rate": 4.220146222583266e-06, + "loss": 1.7856, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.4174695014953613, + "learning_rate": 4.218115353371243e-06, + "loss": 1.7842, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2684640884399414, + "learning_rate": 4.21608448415922e-06, + "loss": 1.7482, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 3.1782288551330566, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8286, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.4046335220336914, + "learning_rate": 4.212022745735175e-06, + "loss": 1.7228, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 3.3722083568573, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8044, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.293466329574585, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.7415, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.28663969039917, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8282, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 1.9627619981765747, + "eval_runtime": 43.6092, + "eval_samples_per_second": 22.931, + "eval_steps_per_second": 0.963, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.255870819091797, + "learning_rate": 4.203899268887084e-06, + "loss": 1.7983, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.4964773654937744, + "learning_rate": 4.201868399675061e-06, + "loss": 1.8141, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.3372740745544434, + "learning_rate": 4.199837530463039e-06, + "loss": 1.7142, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.3506968021392822, + "learning_rate": 4.197806661251016e-06, + "loss": 1.7898, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.242410898208618, + "learning_rate": 4.195775792038993e-06, + "loss": 1.7677, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.3864502906799316, + "learning_rate": 4.19374492282697e-06, + "loss": 1.7097, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.482349157333374, + "learning_rate": 4.191714053614948e-06, + "loss": 1.8052, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.317652940750122, + "learning_rate": 4.189683184402924e-06, + "loss": 1.7692, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.2777392864227295, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8246, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.3506035804748535, + "learning_rate": 4.185621445978879e-06, + "loss": 1.7576, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 1.9610412120819092, + "eval_runtime": 43.3119, + "eval_samples_per_second": 23.088, + "eval_steps_per_second": 0.97, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.3011536598205566, + "learning_rate": 4.183590576766857e-06, + "loss": 1.7891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.2614853382110596, + "learning_rate": 4.181559707554833e-06, + "loss": 1.8258, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.219175100326538, + "learning_rate": 4.179528838342811e-06, + "loss": 1.7494, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.4239423274993896, + "learning_rate": 4.177497969130788e-06, + "loss": 1.7611, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.470712423324585, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8281, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.504060745239258, + "learning_rate": 4.173436230706742e-06, + "loss": 1.7641, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.326601982116699, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.7671, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.4280030727386475, + "learning_rate": 4.169374492282697e-06, + "loss": 1.7573, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.455716371536255, + "learning_rate": 4.167343623070675e-06, + "loss": 1.809, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.3029844760894775, + "learning_rate": 4.165312753858652e-06, + "loss": 1.77, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 1.9604709148406982, + "eval_runtime": 44.4627, + "eval_samples_per_second": 22.491, + "eval_steps_per_second": 0.945, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.3833982944488525, + "learning_rate": 4.163281884646629e-06, + "loss": 1.7652, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.2392966747283936, + "learning_rate": 4.161251015434607e-06, + "loss": 1.7904, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.4546353816986084, + "learning_rate": 4.159220146222583e-06, + "loss": 1.7324, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.13592529296875, + "learning_rate": 4.157189277010561e-06, + "loss": 1.759, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.381657361984253, + "learning_rate": 4.155158407798538e-06, + "loss": 1.7871, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.4142940044403076, + "learning_rate": 4.153127538586516e-06, + "loss": 1.7832, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.305697441101074, + "learning_rate": 4.151096669374492e-06, + "loss": 1.7619, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.193429708480835, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.7738, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.3007619380950928, + "learning_rate": 4.147034930950447e-06, + "loss": 1.7839, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.6337807178497314, + "learning_rate": 4.145004061738425e-06, + "loss": 1.7748, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 1.9597856998443604, + "eval_runtime": 43.8644, + "eval_samples_per_second": 22.798, + "eval_steps_per_second": 0.957, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.4294793605804443, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.7001, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.512155532836914, + "learning_rate": 4.140942323314379e-06, + "loss": 1.7334, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.290808916091919, + "learning_rate": 4.138911454102356e-06, + "loss": 1.7522, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.483161211013794, + "learning_rate": 4.136880584890333e-06, + "loss": 1.798, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2973921298980713, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.7847, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.253084659576416, + "learning_rate": 4.132818846466288e-06, + "loss": 1.7812, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.306342840194702, + "learning_rate": 4.130787977254265e-06, + "loss": 1.7886, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.5153164863586426, + "learning_rate": 4.128757108042242e-06, + "loss": 1.7954, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.4122278690338135, + "learning_rate": 4.12672623883022e-06, + "loss": 1.8525, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.4520716667175293, + "learning_rate": 4.124695369618197e-06, + "loss": 1.7648, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 1.9547661542892456, + "eval_runtime": 45.3774, + "eval_samples_per_second": 22.037, + "eval_steps_per_second": 0.926, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2959611415863037, + "learning_rate": 4.122664500406174e-06, + "loss": 1.7626, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.2280471324920654, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.7954, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.3870065212249756, + "learning_rate": 4.118602761982129e-06, + "loss": 1.7377, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.292612075805664, + "learning_rate": 4.116571892770106e-06, + "loss": 1.7, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.4607021808624268, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8338, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.532824993133545, + "learning_rate": 4.112510154346061e-06, + "loss": 1.7549, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2139623165130615, + "learning_rate": 4.110479285134038e-06, + "loss": 1.78, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.4749462604522705, + "learning_rate": 4.108448415922015e-06, + "loss": 1.8034, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.609265089035034, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.7941, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.367753028869629, + "learning_rate": 4.10438667749797e-06, + "loss": 1.6653, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 1.9539440870285034, + "eval_runtime": 48.8438, + "eval_samples_per_second": 20.473, + "eval_steps_per_second": 0.86, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.9723567962646484, + "learning_rate": 4.102355808285947e-06, + "loss": 1.8317, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.388806104660034, + "learning_rate": 4.100324939073924e-06, + "loss": 1.7597, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4882986545562744, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.7708, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.7586395740509033, + "learning_rate": 4.096263200649879e-06, + "loss": 1.7804, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.237025260925293, + "learning_rate": 4.094232331437856e-06, + "loss": 1.7415, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.421330213546753, + "learning_rate": 4.092201462225833e-06, + "loss": 1.734, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.7281360626220703, + "learning_rate": 4.090170593013811e-06, + "loss": 1.7716, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.2200379371643066, + "learning_rate": 4.088139723801787e-06, + "loss": 1.7266, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.465440034866333, + "learning_rate": 4.086108854589765e-06, + "loss": 1.741, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.39835262298584, + "learning_rate": 4.084077985377742e-06, + "loss": 1.757, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 1.9514222145080566, + "eval_runtime": 52.8068, + "eval_samples_per_second": 18.937, + "eval_steps_per_second": 0.795, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.526684522628784, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8195, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.341954469680786, + "learning_rate": 4.080016246953696e-06, + "loss": 1.7196, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.3755977153778076, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8115, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.3952431678771973, + "learning_rate": 4.075954508529651e-06, + "loss": 1.812, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.554089069366455, + "learning_rate": 4.073923639317629e-06, + "loss": 1.8184, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.2319302558898926, + "learning_rate": 4.071892770105605e-06, + "loss": 1.7448, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.284856081008911, + "learning_rate": 4.069861900893583e-06, + "loss": 1.7154, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.4081835746765137, + "learning_rate": 4.06783103168156e-06, + "loss": 1.7311, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.6115968227386475, + "learning_rate": 4.065800162469538e-06, + "loss": 1.7436, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.368004322052002, + "learning_rate": 4.063769293257514e-06, + "loss": 1.7811, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 1.9508955478668213, + "eval_runtime": 46.8711, + "eval_samples_per_second": 21.335, + "eval_steps_per_second": 0.896, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.4607419967651367, + "learning_rate": 4.061738424045492e-06, + "loss": 1.7563, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.2978763580322266, + "learning_rate": 4.059707554833469e-06, + "loss": 1.7755, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.3316237926483154, + "learning_rate": 4.057676685621446e-06, + "loss": 1.8226, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.3699169158935547, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.7755, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.195596218109131, + "learning_rate": 4.053614947197401e-06, + "loss": 1.6916, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.4524662494659424, + "learning_rate": 4.051584077985378e-06, + "loss": 1.774, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.573888063430786, + "learning_rate": 4.049553208773355e-06, + "loss": 1.7953, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.5157253742218018, + "learning_rate": 4.047522339561333e-06, + "loss": 1.7597, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.5588185787200928, + "learning_rate": 4.04549147034931e-06, + "loss": 1.7553, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.8950047492980957, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8647, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 1.945882797241211, + "eval_runtime": 45.481, + "eval_samples_per_second": 21.987, + "eval_steps_per_second": 0.923, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.217116117477417, + "learning_rate": 4.041429731925264e-06, + "loss": 1.7465, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.5601565837860107, + "learning_rate": 4.039398862713242e-06, + "loss": 1.7575, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.4540457725524902, + "learning_rate": 4.037367993501219e-06, + "loss": 1.8291, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.327181100845337, + "learning_rate": 4.035337124289196e-06, + "loss": 1.7409, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.485457181930542, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.7192, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.4613442420959473, + "learning_rate": 4.031275385865151e-06, + "loss": 1.7613, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.303217649459839, + "learning_rate": 4.029244516653128e-06, + "loss": 1.7799, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.3830621242523193, + "learning_rate": 4.027213647441105e-06, + "loss": 1.7045, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.342332601547241, + "learning_rate": 4.025182778229083e-06, + "loss": 1.7246, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.4014711380004883, + "learning_rate": 4.02315190901706e-06, + "loss": 1.7621, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 1.95199453830719, + "eval_runtime": 45.4754, + "eval_samples_per_second": 21.99, + "eval_steps_per_second": 0.924, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.6015400886535645, + "learning_rate": 4.021121039805037e-06, + "loss": 1.7536, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.866978406906128, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8405, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.389810085296631, + "learning_rate": 4.017059301380992e-06, + "loss": 1.7365, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.280442953109741, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8273, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.509330987930298, + "learning_rate": 4.012997562956945e-06, + "loss": 1.7127, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.452078104019165, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.7457, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.368337631225586, + "learning_rate": 4.0089358245329e-06, + "loss": 1.7412, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.4529192447662354, + "learning_rate": 4.006904955320878e-06, + "loss": 1.7434, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.647761821746826, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8013, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.387424945831299, + "learning_rate": 4.002843216896833e-06, + "loss": 1.7948, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 1.946029543876648, + "eval_runtime": 43.9666, + "eval_samples_per_second": 22.745, + "eval_steps_per_second": 0.955, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.9217846393585205, + "learning_rate": 4.000812347684809e-06, + "loss": 1.7384, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.400040864944458, + "learning_rate": 3.998781478472787e-06, + "loss": 1.7496, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.4050440788269043, + "learning_rate": 3.996750609260764e-06, + "loss": 1.7165, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 3.6001529693603516, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8679, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 2.4040098190307617, + "learning_rate": 3.992688870836718e-06, + "loss": 1.7671, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.5645527839660645, + "learning_rate": 3.990658001624696e-06, + "loss": 1.7374, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.799394130706787, + "learning_rate": 3.988627132412673e-06, + "loss": 1.796, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.5481607913970947, + "learning_rate": 3.98659626320065e-06, + "loss": 1.753, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.8851330280303955, + "learning_rate": 3.984565393988627e-06, + "loss": 1.7505, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.6668694019317627, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.7604, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 1.939759612083435, + "eval_runtime": 44.5019, + "eval_samples_per_second": 22.471, + "eval_steps_per_second": 0.944, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.320935010910034, + "learning_rate": 3.980503655564582e-06, + "loss": 1.7749, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.829897880554199, + "learning_rate": 3.978472786352559e-06, + "loss": 1.7763, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.8420908451080322, + "learning_rate": 3.976441917140536e-06, + "loss": 1.7598, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.8183, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.781, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 3.9271891117095947, + "learning_rate": 3.972380178716491e-06, + "loss": 1.7949, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.277897834777832, + "learning_rate": 3.970349309504468e-06, + "loss": 1.707, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 3.241063117980957, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.7175, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.514458417892456, + "learning_rate": 3.966287571080423e-06, + "loss": 1.7764, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.420212745666504, + "learning_rate": 3.9642567018684e-06, + "loss": 1.7618, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 1.9401386976242065, + "eval_runtime": 44.1672, + "eval_samples_per_second": 22.641, + "eval_steps_per_second": 0.951, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.3980538845062256, + "learning_rate": 3.962225832656377e-06, + "loss": 1.6922, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.892646312713623, + "learning_rate": 3.960194963444355e-06, + "loss": 1.7571, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.499260902404785, + "learning_rate": 3.958164094232332e-06, + "loss": 1.7501, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.409695625305176, + "learning_rate": 3.956133225020309e-06, + "loss": 1.7553, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.640069007873535, + "learning_rate": 3.954102355808286e-06, + "loss": 1.7526, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.9101760387420654, + "learning_rate": 3.952071486596264e-06, + "loss": 1.7483, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.2579474449157715, + "learning_rate": 3.950040617384241e-06, + "loss": 1.7379, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 4.039298057556152, + "learning_rate": 3.948009748172218e-06, + "loss": 1.7988, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.3232030868530273, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.7592, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.697164297103882, + "learning_rate": 3.943948009748173e-06, + "loss": 1.7622, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 1.9367746114730835, + "eval_runtime": 43.7013, + "eval_samples_per_second": 22.883, + "eval_steps_per_second": 0.961, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.480297803878784, + "learning_rate": 3.94191714053615e-06, + "loss": 1.8006, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.3624603748321533, + "learning_rate": 3.939886271324127e-06, + "loss": 1.7152, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.417187213897705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.7077, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.6575164794921875, + "learning_rate": 3.935824532900082e-06, + "loss": 1.7744, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.4896557331085205, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8052, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.475693941116333, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.7008, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.3210372924804688, + "learning_rate": 3.929731925264013e-06, + "loss": 1.7339, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.433785915374756, + "learning_rate": 3.927701056051991e-06, + "loss": 1.7544, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 3.181574583053589, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.7229, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.704399347305298, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.7086, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 1.9335075616836548, + "eval_runtime": 47.0248, + "eval_samples_per_second": 21.265, + "eval_steps_per_second": 0.893, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 3.5027623176574707, + "learning_rate": 3.921608448415922e-06, + "loss": 1.7622, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.450943946838379, + "learning_rate": 3.9195775792039e-06, + "loss": 1.7286, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.4603374004364014, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.7725, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.8217649459838867, + "learning_rate": 3.915515840779855e-06, + "loss": 1.7859, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.3308372497558594, + "learning_rate": 3.913484971567831e-06, + "loss": 1.6874, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.624776840209961, + "learning_rate": 3.911454102355808e-06, + "loss": 1.7632, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.4097838401794434, + "learning_rate": 3.909423233143786e-06, + "loss": 1.7576, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.487410545349121, + "learning_rate": 3.907392363931763e-06, + "loss": 1.831, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.647245168685913, + "learning_rate": 3.90536149471974e-06, + "loss": 1.8448, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.397664785385132, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.868, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 1.930379033088684, + "eval_runtime": 51.1662, + "eval_samples_per_second": 19.544, + "eval_steps_per_second": 0.821, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.277433395385742, + "learning_rate": 3.901299756295695e-06, + "loss": 1.7394, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 3.1312496662139893, + "learning_rate": 3.899268887083672e-06, + "loss": 1.7422, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.5011534690856934, + "learning_rate": 3.897238017871649e-06, + "loss": 1.7484, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.8716166019439697, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.7253, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.4346914291381836, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8197, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.348475933074951, + "learning_rate": 3.891145410235581e-06, + "loss": 1.7737, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.3996007442474365, + "learning_rate": 3.889114541023558e-06, + "loss": 1.7333, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.355262279510498, + "learning_rate": 3.887083671811536e-06, + "loss": 1.6902, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.335932731628418, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8206, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.2893190383911133, + "learning_rate": 3.88302193338749e-06, + "loss": 1.7533, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 1.9290764331817627, + "eval_runtime": 46.9373, + "eval_samples_per_second": 21.305, + "eval_steps_per_second": 0.895, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.739842176437378, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.7318, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.587916374206543, + "learning_rate": 3.878960194963445e-06, + "loss": 1.7979, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.406548500061035, + "learning_rate": 3.876929325751422e-06, + "loss": 1.6899, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.440321922302246, + "learning_rate": 3.874898456539399e-06, + "loss": 1.7546, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.320030450820923, + "learning_rate": 3.872867587327377e-06, + "loss": 1.7254, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.3795316219329834, + "learning_rate": 3.870836718115354e-06, + "loss": 1.728, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.299802541732788, + "learning_rate": 3.868805848903331e-06, + "loss": 1.7464, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.469719409942627, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.765, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.612109899520874, + "learning_rate": 3.864744110479286e-06, + "loss": 1.7827, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.623173952102661, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8019, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 1.9268505573272705, + "eval_runtime": 46.77, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.4146616458892822, + "learning_rate": 3.86068237205524e-06, + "loss": 1.7498, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.361994743347168, + "learning_rate": 3.858651502843217e-06, + "loss": 1.6926, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.493215560913086, + "learning_rate": 3.856620633631195e-06, + "loss": 1.736, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.4341812133789062, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8152, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.391331195831299, + "learning_rate": 3.852558895207149e-06, + "loss": 1.7717, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.419492244720459, + "learning_rate": 3.850528025995126e-06, + "loss": 1.7283, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.349090576171875, + "learning_rate": 3.848497156783104e-06, + "loss": 1.7317, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.298647880554199, + "learning_rate": 3.84646628757108e-06, + "loss": 1.7389, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.4612667560577393, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.7333, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.3139944076538086, + "learning_rate": 3.842404549147035e-06, + "loss": 1.6944, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 1.924438714981079, + "eval_runtime": 45.8276, + "eval_samples_per_second": 21.821, + "eval_steps_per_second": 0.916, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 2.17958664894104, + "learning_rate": 3.840373679935013e-06, + "loss": 1.7474, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.337859869003296, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.7345, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4777672290802, + "learning_rate": 3.836311941510967e-06, + "loss": 1.77, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.280158519744873, + "learning_rate": 3.834281072298944e-06, + "loss": 1.7069, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2387664318084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.7889, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.373771905899048, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8268, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.4950132369995117, + "learning_rate": 3.828188464662876e-06, + "loss": 1.7586, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.3296444416046143, + "learning_rate": 3.826157595450853e-06, + "loss": 1.6991, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.6653530597686768, + "learning_rate": 3.82412672623883e-06, + "loss": 1.7505, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.485128402709961, + "learning_rate": 3.822095857026808e-06, + "loss": 1.7196, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 1.9221889972686768, + "eval_runtime": 46.6581, + "eval_samples_per_second": 21.433, + "eval_steps_per_second": 0.9, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.3726181983947754, + "learning_rate": 3.820064987814785e-06, + "loss": 1.7296, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.233992099761963, + "learning_rate": 3.818034118602762e-06, + "loss": 1.7313, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.5388147830963135, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.7435, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.4787333011627197, + "learning_rate": 3.813972380178717e-06, + "loss": 1.7733, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.232564926147461, + "learning_rate": 3.811941510966694e-06, + "loss": 1.693, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.3882386684417725, + "learning_rate": 3.809910641754671e-06, + "loss": 1.6529, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.4785983562469482, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.7456, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.685098886489868, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.7162, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.353523015975952, + "learning_rate": 3.803818034118603e-06, + "loss": 1.7852, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.342163562774658, + "learning_rate": 3.80178716490658e-06, + "loss": 1.7767, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 1.9235820770263672, + "eval_runtime": 47.1327, + "eval_samples_per_second": 21.217, + "eval_steps_per_second": 0.891, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.550818920135498, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.7328, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.3204798698425293, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.7058, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.4752719402313232, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8051, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.5202794075012207, + "learning_rate": 3.793663688058489e-06, + "loss": 1.7966, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.26521635055542, + "learning_rate": 3.791632818846467e-06, + "loss": 1.7879, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.331515312194824, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.7814, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.37178635597229, + "learning_rate": 3.787571080422421e-06, + "loss": 1.7635, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.321417808532715, + "learning_rate": 3.785540211210398e-06, + "loss": 1.7513, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.485989809036255, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.7306, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.7497379779815674, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8409, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 1.9275448322296143, + "eval_runtime": 46.2088, + "eval_samples_per_second": 21.641, + "eval_steps_per_second": 0.909, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 4.9214019775390625, + "learning_rate": 3.77944760357433e-06, + "loss": 1.7834, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 4.505277156829834, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.7179, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.5772030353546143, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.7766, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.254185438156128, + "learning_rate": 3.773354995938262e-06, + "loss": 1.7697, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 3.632233142852783, + "learning_rate": 3.771324126726239e-06, + "loss": 1.7968, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.174461841583252, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.7346, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.3983266353607178, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.6534, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 3.393791913986206, + "learning_rate": 3.765231519090171e-06, + "loss": 1.7561, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.2853987216949463, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.6963, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.649789810180664, + "learning_rate": 3.761169780666125e-06, + "loss": 1.7395, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 1.9242008924484253, + "eval_runtime": 46.6017, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 3.417872190475464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.784, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.316666841506958, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.7154, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.486161231994629, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.7903, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 4.538443565368652, + "learning_rate": 3.753046303818034e-06, + "loss": 1.7711, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.9772889614105225, + "learning_rate": 3.751015434606012e-06, + "loss": 1.8046, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 7.2353715896606445, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.7189, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.736, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.8026, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 12.551762580871582, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.7788, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 3.0898046493530273, + "learning_rate": 3.742891957757921e-06, + "loss": 1.7878, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 1.9646916389465332, + "eval_runtime": 44.1583, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 8.120022773742676, + "learning_rate": 3.740861088545898e-06, + "loss": 1.768, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 6.485774040222168, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.746, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.85135555267334, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.6968, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 5.8551435470581055, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.6796, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 4.45880126953125, + "learning_rate": 3.732737611697807e-06, + "loss": 1.7219, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 3.0435831546783447, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.6959, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.8194847106933594, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.8152, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 3.1266117095947266, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.6937, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 3.642854690551758, + "learning_rate": 3.724614134849716e-06, + "loss": 1.7992, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 3.344496965408325, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.7583, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 1.9293720722198486, + "eval_runtime": 50.8602, + "eval_samples_per_second": 19.662, + "eval_steps_per_second": 0.826, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.7666447162628174, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7441, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.8819239139556885, + "learning_rate": 3.718521527213648e-06, + "loss": 1.6314, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.6030640602111816, + "learning_rate": 3.716490658001625e-06, + "loss": 1.7389, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.705246925354004, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.7408, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 3.181328535079956, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7092, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 3.0449445247650146, + "learning_rate": 3.710398050365557e-06, + "loss": 1.7399, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.650517225265503, + "learning_rate": 3.708367181153534e-06, + "loss": 1.6861, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.7593672275543213, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.6701, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 3.0486466884613037, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.7191, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.502850294113159, + "learning_rate": 3.702274573517466e-06, + "loss": 1.6982, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 1.9228183031082153, + "eval_runtime": 49.565, + "eval_samples_per_second": 20.176, + "eval_steps_per_second": 0.847, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.400770664215088, + "learning_rate": 3.700243704305443e-06, + "loss": 1.6666, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.5583295822143555, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.7971, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.519505739212036, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.7076, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 3.106309413909912, + "learning_rate": 3.694151096669375e-06, + "loss": 1.7279, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.516906499862671, + "learning_rate": 3.692120227457352e-06, + "loss": 1.6265, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.603166103363037, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.7122, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.842773199081421, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.7312, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.8860342502593994, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7229, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.517254114151001, + "learning_rate": 3.683996750609261e-06, + "loss": 1.7342, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.430025100708008, + "learning_rate": 3.681965881397238e-06, + "loss": 1.6817, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 1.9156773090362549, + "eval_runtime": 44.6903, + "eval_samples_per_second": 22.376, + "eval_steps_per_second": 0.94, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.340075969696045, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.7044, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.4758968353271484, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.6935, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.878903865814209, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.7021, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.388366222381592, + "learning_rate": 3.673842404549147e-06, + "loss": 1.7229, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.5789167881011963, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8112, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.743072509765625, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.7239, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.5485000610351562, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.7641, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.53190541267395, + "learning_rate": 3.665718927701056e-06, + "loss": 1.6996, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.3354969024658203, + "learning_rate": 3.663688058489034e-06, + "loss": 1.6623, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.37371826171875, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.7109, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 1.9122159481048584, + "eval_runtime": 43.8844, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.406137228012085, + "learning_rate": 3.659626320064988e-06, + "loss": 1.6958, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.4220657348632812, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7196, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.6203134059906006, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.6014, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.5388472080230713, + "learning_rate": 3.65353371242892e-06, + "loss": 1.7272, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.672658681869507, + "learning_rate": 3.651502843216897e-06, + "loss": 1.6786, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4366531372070312, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7198, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.4872887134552, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.7082, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.371206283569336, + "learning_rate": 3.645410235580829e-06, + "loss": 1.6964, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.496103286743164, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.716, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.658316135406494, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.7263, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 1.9115047454833984, + "eval_runtime": 43.8788, + "eval_samples_per_second": 22.79, + "eval_steps_per_second": 0.957, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.5931882858276367, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.7586, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.4697139263153076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.7232, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.7274153232574463, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.6917, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.683138132095337, + "learning_rate": 3.633225020308692e-06, + "loss": 1.7518, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.4510817527770996, + "learning_rate": 3.63119415109667e-06, + "loss": 1.6533, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.6939656734466553, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.6545, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.572864532470703, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.6965, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.4816951751708984, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.6851, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.640394926071167, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7336, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.6707329750061035, + "learning_rate": 3.621039805036556e-06, + "loss": 1.7182, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 1.91274094581604, + "eval_runtime": 44.35, + "eval_samples_per_second": 22.548, + "eval_steps_per_second": 0.947, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.6947290897369385, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.7255, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.5496068000793457, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.6209, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.6196374893188477, + "learning_rate": 3.614947197400488e-06, + "loss": 1.7277, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.4534807205200195, + "learning_rate": 3.612916328188465e-06, + "loss": 1.7404, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.5428082942962646, + "learning_rate": 3.610885458976442e-06, + "loss": 1.6618, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.5431482791900635, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.6743, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.707956552505493, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.7483, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.5643982887268066, + "learning_rate": 3.604792851340374e-06, + "loss": 1.7304, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.4612927436828613, + "learning_rate": 3.602761982128351e-06, + "loss": 1.6561, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.3140358924865723, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.6834, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 1.9124692678451538, + "eval_runtime": 44.0522, + "eval_samples_per_second": 22.7, + "eval_steps_per_second": 0.953, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4636526107788086, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.7395, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.541180372238159, + "learning_rate": 3.596669374492283e-06, + "loss": 1.6579, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.621335983276367, + "learning_rate": 3.59463850528026e-06, + "loss": 1.7332, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.363450050354004, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.7047, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.8609447479248047, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7416, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.457338571548462, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.7516, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.3943030834198, + "learning_rate": 3.586515028432169e-06, + "loss": 1.6436, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.4719340801239014, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.6988, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3442330360412598, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.7165, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.3391900062561035, + "learning_rate": 3.580422420796101e-06, + "loss": 1.6263, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 1.9100929498672485, + "eval_runtime": 44.2504, + "eval_samples_per_second": 22.599, + "eval_steps_per_second": 0.949, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.533566474914551, + "learning_rate": 3.578391551584078e-06, + "loss": 1.7141, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.464540481567383, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.7335, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.4789979457855225, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7074, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.8252034187316895, + "learning_rate": 3.57229894394801e-06, + "loss": 1.6808, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.5870537757873535, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.7171, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.349393129348755, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.6687, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.3343207836151123, + "learning_rate": 3.566206336311942e-06, + "loss": 1.6351, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.558645248413086, + "learning_rate": 3.564175467099919e-06, + "loss": 1.708, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.7598931789398193, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.6797, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.6048543453216553, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7117, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 1.906927466392517, + "eval_runtime": 44.5189, + "eval_samples_per_second": 22.462, + "eval_steps_per_second": 0.943, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.6124815940856934, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.7328, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.3410215377807617, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.6932, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.5247764587402344, + "learning_rate": 3.554021121039805e-06, + "loss": 1.6628, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.642958164215088, + "learning_rate": 3.551990251827783e-06, + "loss": 1.7035, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.3978517055511475, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.6796, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.3969287872314453, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.6626, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.5723509788513184, + "learning_rate": 3.545897644191714e-06, + "loss": 1.7135, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.545154571533203, + "learning_rate": 3.543866774979692e-06, + "loss": 1.6626, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.501058340072632, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.7112, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.695483684539795, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.6773, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 1.9063464403152466, + "eval_runtime": 44.2034, + "eval_samples_per_second": 22.623, + "eval_steps_per_second": 0.95, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 3.0584375858306885, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.7242, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.445220947265625, + "learning_rate": 3.535743298131601e-06, + "loss": 1.7349, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.427220106124878, + "learning_rate": 3.533712428919578e-06, + "loss": 1.677, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.5148212909698486, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.7007, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.644319534301758, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.7442, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.6118004322052, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.7509, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.5224156379699707, + "learning_rate": 3.525588952071487e-06, + "loss": 1.7279, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.4668405055999756, + "learning_rate": 3.523558082859464e-06, + "loss": 1.6754, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 3.8330910205841064, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.6815, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.569058895111084, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.6518, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 1.906802773475647, + "eval_runtime": 50.8521, + "eval_samples_per_second": 19.665, + "eval_steps_per_second": 0.826, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.705265522003174, + "learning_rate": 3.517465475223396e-06, + "loss": 1.7172, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 3.2667057514190674, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7641, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 3.5119211673736572, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.7178, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.7348034381866455, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.6815, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 3.019892454147339, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.6417, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 3.2486684322357178, + "learning_rate": 3.507311129163282e-06, + "loss": 1.6888, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.897071361541748, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7121, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.7342352867126465, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.6339, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.691178560256958, + "learning_rate": 3.501218521527214e-06, + "loss": 1.7069, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.548682689666748, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7218, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 1.9029825925827026, + "eval_runtime": 48.287, + "eval_samples_per_second": 20.71, + "eval_steps_per_second": 0.87, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.422056198120117, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.6486, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.4367074966430664, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.759, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.48002290725708, + "learning_rate": 3.493095044679123e-06, + "loss": 1.5902, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.5030295848846436, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.7108, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.5892179012298584, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.6154, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.4282920360565186, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.6807, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.4826700687408447, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7284, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.416597843170166, + "learning_rate": 3.482940698619009e-06, + "loss": 1.6972, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.4664933681488037, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.7119, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.541856288909912, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.6752, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 1.8999215364456177, + "eval_runtime": 46.3528, + "eval_samples_per_second": 21.574, + "eval_steps_per_second": 0.906, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4803109169006348, + "learning_rate": 3.476848090982941e-06, + "loss": 1.776, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.4363205432891846, + "learning_rate": 3.474817221770918e-06, + "loss": 1.7196, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.4377009868621826, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.6937, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.425791025161743, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.6547, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3216712474823, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.6789, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.65527081489563, + "learning_rate": 3.466693744922827e-06, + "loss": 1.7231, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.3415145874023438, + "learning_rate": 3.464662875710805e-06, + "loss": 1.6805, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.3864822387695312, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.7153, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.490896224975586, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.6866, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.339711904525757, + "learning_rate": 3.458570268074736e-06, + "loss": 1.6732, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 1.8985264301300049, + "eval_runtime": 46.7854, + "eval_samples_per_second": 21.374, + "eval_steps_per_second": 0.898, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.393709659576416, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.6549, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.564114809036255, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.7019, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3826403617858887, + "learning_rate": 3.452477660438668e-06, + "loss": 1.6427, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.3185465335845947, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.6916, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4503891468048096, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.6793, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.386446714401245, + "learning_rate": 3.4463850528026e-06, + "loss": 1.7186, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.6404714584350586, + "learning_rate": 3.444354183590577e-06, + "loss": 1.6605, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.342374324798584, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.6219, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.7503273487091064, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.693, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.5416243076324463, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7323, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 1.9016659259796143, + "eval_runtime": 48.5456, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.5787603855133057, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.7057, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.649528980255127, + "learning_rate": 3.434199837530463e-06, + "loss": 1.6348, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.504089832305908, + "learning_rate": 3.432168968318441e-06, + "loss": 1.7202, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.5624678134918213, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7351, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.641331911087036, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.7453, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.599500894546509, + "learning_rate": 3.426076360682372e-06, + "loss": 1.6942, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.668599843978882, + "learning_rate": 3.42404549147035e-06, + "loss": 1.6337, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.75667667388916, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.6988, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.641526937484741, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.7641, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.4972548484802246, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.7269, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 1.8979382514953613, + "eval_runtime": 45.6555, + "eval_samples_per_second": 21.903, + "eval_steps_per_second": 0.92, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.5454585552215576, + "learning_rate": 3.415922014622259e-06, + "loss": 1.7294, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.5921928882598877, + "learning_rate": 3.413891145410236e-06, + "loss": 1.7145, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.4028453826904297, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.6985, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.604522705078125, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.6586, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.528970241546631, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.734, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.565030574798584, + "learning_rate": 3.405767668562145e-06, + "loss": 1.6997, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.7735602855682373, + "learning_rate": 3.403736799350122e-06, + "loss": 1.6543, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.5116474628448486, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.7711, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.4827239513397217, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.6831, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.544309377670288, + "learning_rate": 3.397644191714054e-06, + "loss": 1.6905, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 1.8945367336273193, + "eval_runtime": 46.6027, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.456636905670166, + "learning_rate": 3.395613322502031e-06, + "loss": 1.646, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.5900352001190186, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.6967, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 3.3344573974609375, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.6947, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.5621678829193115, + "learning_rate": 3.389520714865963e-06, + "loss": 1.7084, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.5676684379577637, + "learning_rate": 3.38748984565394e-06, + "loss": 1.6788, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.600691318511963, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.6811, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.5324196815490723, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.661, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.7559397220611572, + "learning_rate": 3.381397238017872e-06, + "loss": 1.7473, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.526397466659546, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.3752379417419434, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.689, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 1.8995447158813477, + "eval_runtime": 48.0083, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.5902793407440186, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.6899, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 3.157573938369751, + "learning_rate": 3.373273761169781e-06, + "loss": 1.6792, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.628035545349121, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.7003, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.4810822010040283, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.6489, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.4859938621520996, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.6768, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.6773345470428467, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7046, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.9695260524749756, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.7341, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.8146679401397705, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.7472, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.534985065460205, + "learning_rate": 3.359057676685622e-06, + "loss": 1.6199, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.9351718425750732, + "learning_rate": 3.357026807473599e-06, + "loss": 1.6656, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 1.8962808847427368, + "eval_runtime": 47.0443, + "eval_samples_per_second": 21.257, + "eval_steps_per_second": 0.893, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.4404256343841553, + "learning_rate": 3.354995938261576e-06, + "loss": 1.6207, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.6691436767578125, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.6627, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.5436172485351562, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.6919, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.7167909145355225, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.6434, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.575446367263794, + "learning_rate": 3.346872461413485e-06, + "loss": 1.6554, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.3874504566192627, + "learning_rate": 3.344841592201463e-06, + "loss": 1.6691, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.66317081451416, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.6864, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.4396398067474365, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.7217, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.5002803802490234, + "learning_rate": 3.338748984565394e-06, + "loss": 1.7556, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.679577350616455, + "learning_rate": 3.336718115353372e-06, + "loss": 1.6822, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 1.8942477703094482, + "eval_runtime": 44.3804, + "eval_samples_per_second": 22.532, + "eval_steps_per_second": 0.946, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.6119370460510254, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.662, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.6024818420410156, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.687, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 4.128854751586914, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.6898, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3910412788391113, + "learning_rate": 3.32859463850528e-06, + "loss": 1.6274, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.8630096912384033, + "learning_rate": 3.326563769293258e-06, + "loss": 1.6554, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 4.201050281524658, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.6676, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.492168664932251, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.6319, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 3.6807360649108887, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.6776, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.7271926403045654, + "learning_rate": 3.318440292445167e-06, + "loss": 1.6383, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.5567307472229004, + "learning_rate": 3.316409423233144e-06, + "loss": 1.7325, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 1.892768383026123, + "eval_runtime": 50.909, + "eval_samples_per_second": 19.643, + "eval_steps_per_second": 0.825, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.5373101234436035, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.689, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.602923631668091, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.6861, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.507110118865967, + "learning_rate": 3.310316815597076e-06, + "loss": 1.7575, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.361786127090454, + "learning_rate": 3.308285946385053e-06, + "loss": 1.6267, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5360615253448486, + "learning_rate": 3.30625507717303e-06, + "loss": 1.6911, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.6238744258880615, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.7193, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.5358500480651855, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.6691, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.7285404205322266, + "learning_rate": 3.300162469536962e-06, + "loss": 1.7249, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.346160650253296, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.6958, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.5415399074554443, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.7544, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 1.8925012350082397, + "eval_runtime": 46.4957, + "eval_samples_per_second": 21.507, + "eval_steps_per_second": 0.903, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.5924015045166016, + "learning_rate": 3.294069861900894e-06, + "loss": 1.6901, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.4649693965911865, + "learning_rate": 3.292038992688871e-06, + "loss": 1.7234, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.3648035526275635, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.7324, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.5146145820617676, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.631, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.5919511318206787, + "learning_rate": 3.285946385052803e-06, + "loss": 1.706, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.299527168273926, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.608, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.480637788772583, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.7415, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.458183526992798, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.6806, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.4864983558654785, + "learning_rate": 3.277822908204712e-06, + "loss": 1.6373, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.3432068824768066, + "learning_rate": 3.275792038992689e-06, + "loss": 1.6832, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 1.8899296522140503, + "eval_runtime": 44.3309, + "eval_samples_per_second": 22.558, + "eval_steps_per_second": 0.947, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.365891218185425, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.6492, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.463261842727661, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.6847, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.3977205753326416, + "learning_rate": 3.269699431356621e-06, + "loss": 1.6789, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.4172558784484863, + "learning_rate": 3.267668562144598e-06, + "loss": 1.6495, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.663381814956665, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.7139, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.3702306747436523, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.6876, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 3.1759073734283447, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.7465, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.6344592571258545, + "learning_rate": 3.259545085296507e-06, + "loss": 1.6339, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.364490032196045, + "learning_rate": 3.257514216084484e-06, + "loss": 1.7394, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.5639748573303223, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.6928, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 1.8937300443649292, + "eval_runtime": 43.9356, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 3.152968645095825, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.6693, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.45776104927063, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.6999, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.5289807319641113, + "learning_rate": 3.249390739236393e-06, + "loss": 1.6523, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.4168646335601807, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.6823, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.5115208625793457, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.6816, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.685535192489624, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.6307, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.253084897994995, + "learning_rate": 3.241267262388302e-06, + "loss": 1.662, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.4825408458709717, + "learning_rate": 3.23923639317628e-06, + "loss": 1.6444, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.691168785095215, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.6912, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.978756904602051, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7536, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 1.8903727531433105, + "eval_runtime": 43.4099, + "eval_samples_per_second": 23.036, + "eval_steps_per_second": 0.968, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.317105531692505, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.6441, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 3.1030166149139404, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.7205, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.595792770385742, + "learning_rate": 3.229082047116166e-06, + "loss": 1.677, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.3315582275390625, + "learning_rate": 3.227051177904143e-06, + "loss": 1.6779, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.4262566566467285, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.7003, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.4438798427581787, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.6081, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.573531150817871, + "learning_rate": 3.220958570268075e-06, + "loss": 1.6489, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.477809429168701, + "learning_rate": 3.218927701056052e-06, + "loss": 1.6025, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.5499629974365234, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.7337, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.4699153900146484, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.6443, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 1.8867121934890747, + "eval_runtime": 42.9343, + "eval_samples_per_second": 23.291, + "eval_steps_per_second": 0.978, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.4200708866119385, + "learning_rate": 3.212835093419984e-06, + "loss": 1.6743, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.418069839477539, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.723, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.452284336090088, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.6358, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.4241347312927246, + "learning_rate": 3.206742485783916e-06, + "loss": 1.638, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.461860418319702, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.7017, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.591986894607544, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7065, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.478470802307129, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.6262, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.49831223487854, + "learning_rate": 3.198619008935825e-06, + "loss": 1.6871, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.485332727432251, + "learning_rate": 3.196588139723802e-06, + "loss": 1.6864, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.533710479736328, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.6794, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 1.88768470287323, + "eval_runtime": 43.5948, + "eval_samples_per_second": 22.938, + "eval_steps_per_second": 0.963, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.368609666824341, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.6877, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.4875223636627197, + "learning_rate": 3.190495532087734e-06, + "loss": 1.7406, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.5075156688690186, + "learning_rate": 3.188464662875711e-06, + "loss": 1.6545, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.428731918334961, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.6213, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.467618465423584, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.6263, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.561612129211426, + "learning_rate": 3.182372055239643e-06, + "loss": 1.6987, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.5035271644592285, + "learning_rate": 3.18034118602762e-06, + "loss": 1.6902, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.538311243057251, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.6523, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.486321210861206, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.6665, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.433173894882202, + "learning_rate": 3.174248578391552e-06, + "loss": 1.6605, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 1.8853652477264404, + "eval_runtime": 43.3417, + "eval_samples_per_second": 23.072, + "eval_steps_per_second": 0.969, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.4658470153808594, + "learning_rate": 3.172217709179529e-06, + "loss": 1.6615, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.452450752258301, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.6332, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.5310161113739014, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.6846, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.4504501819610596, + "learning_rate": 3.166125101543461e-06, + "loss": 1.6996, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.6004488468170166, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.6302, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.5000431537628174, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.6742, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.4869067668914795, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.6797, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.5018959045410156, + "learning_rate": 3.15800162469537e-06, + "loss": 1.673, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.6645195484161377, + "learning_rate": 3.155970755483347e-06, + "loss": 1.6908, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.3708560466766357, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.683, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 1.8829350471496582, + "eval_runtime": 43.5383, + "eval_samples_per_second": 22.968, + "eval_steps_per_second": 0.965, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.574108362197876, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.7155, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.4669814109802246, + "learning_rate": 3.149878147847279e-06, + "loss": 1.7529, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4713149070739746, + "learning_rate": 3.147847278635256e-06, + "loss": 1.6491, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.5128800868988037, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.6276, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.740370035171509, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.6778, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.6038050651550293, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.6527, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.782407283782959, + "learning_rate": 3.139723801787165e-06, + "loss": 1.6989, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.7673134803771973, + "learning_rate": 3.137692932575143e-06, + "loss": 1.6291, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.7472870349884033, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.6776, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.4887819290161133, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7009, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.8792160749435425, + "eval_runtime": 43.7334, + "eval_samples_per_second": 22.866, + "eval_steps_per_second": 0.96, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.4528491497039795, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.7245, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.5177690982818604, + "learning_rate": 3.129569455727051e-06, + "loss": 1.6852, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.9238510131835938, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.6994, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.446660041809082, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.6636, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.644211530685425, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.6983, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.697342872619629, + "learning_rate": 3.12144597887896e-06, + "loss": 1.7172, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.714172124862671, + "learning_rate": 3.119415109666938e-06, + "loss": 1.6702, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.471735715866089, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.6945, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.5089669227600098, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.7181, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.349229335784912, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.7297, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.880064845085144, + "eval_runtime": 47.0636, + "eval_samples_per_second": 21.248, + "eval_steps_per_second": 0.892, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.6843974590301514, + "learning_rate": 3.111291632818847e-06, + "loss": 1.6698, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.4325790405273438, + "learning_rate": 3.109260763606824e-06, + "loss": 1.6973, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.488459825515747, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.6552, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.4992082118988037, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.7114, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.4742603302001953, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.7073, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.556330442428589, + "learning_rate": 3.101137286758733e-06, + "loss": 1.7108, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.5711283683776855, + "learning_rate": 3.09910641754671e-06, + "loss": 1.6811, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.6467158794403076, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.7395, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.465963363647461, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.7006, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.622716188430786, + "learning_rate": 3.093013809910642e-06, + "loss": 1.7385, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.8808013200759888, + "eval_runtime": 45.5126, + "eval_samples_per_second": 21.972, + "eval_steps_per_second": 0.923, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.489701509475708, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.6949, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.4398739337921143, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.6802, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.3654873371124268, + "learning_rate": 3.086921202274574e-06, + "loss": 1.6452, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.518768548965454, + "learning_rate": 3.084890333062551e-06, + "loss": 1.6913, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.63173246383667, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.6611, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.531484603881836, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.611, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.4796738624572754, + "learning_rate": 3.078797725426483e-06, + "loss": 1.6764, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.39241099357605, + "learning_rate": 3.07676685621446e-06, + "loss": 1.6417, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3984735012054443, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.6546, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.5202760696411133, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.6879, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.879974365234375, + "eval_runtime": 44.395, + "eval_samples_per_second": 22.525, + "eval_steps_per_second": 0.946, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.5556983947753906, + "learning_rate": 3.070674248578392e-06, + "loss": 1.6733, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.3853955268859863, + "learning_rate": 3.068643379366369e-06, + "loss": 1.6274, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.4503982067108154, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.6896, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.5618021488189697, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.6592, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.6265292167663574, + "learning_rate": 3.062550771730301e-06, + "loss": 1.6634, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.4617390632629395, + "learning_rate": 3.060519902518278e-06, + "loss": 1.6032, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.4191603660583496, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.6871, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.5509119033813477, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.653, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.5399646759033203, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.7046, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.446967124938965, + "learning_rate": 3.052396425670187e-06, + "loss": 1.6547, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.8803874254226685, + "eval_runtime": 45.361, + "eval_samples_per_second": 22.045, + "eval_steps_per_second": 0.926, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.419206142425537, + "learning_rate": 3.050365556458164e-06, + "loss": 1.6517, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.4592065811157227, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.693, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.5429770946502686, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7124, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.5141642093658447, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.7103, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.362539291381836, + "learning_rate": 3.042242079610073e-06, + "loss": 1.6205, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.401175022125244, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.6479, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.5883190631866455, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.6732, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.5273513793945312, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.6673, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.6569600105285645, + "learning_rate": 3.034118602761982e-06, + "loss": 1.6396, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.8843557834625244, + "learning_rate": 3.03208773354996e-06, + "loss": 1.7136, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.8763896226882935, + "eval_runtime": 46.1886, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.6459810733795166, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.6939, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.4619321823120117, + "learning_rate": 3.028025995125914e-06, + "loss": 1.6197, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.390430450439453, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.6805, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.5494728088378906, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.6116, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.5397655963897705, + "learning_rate": 3.021933387489846e-06, + "loss": 1.6753, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.634716510772705, + "learning_rate": 3.019902518277823e-06, + "loss": 1.6892, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.5844202041625977, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.6401, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.577362060546875, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.6405, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.3554489612579346, + "learning_rate": 3.013809910641755e-06, + "loss": 1.6427, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.893918037414551, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.6739, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.8772739171981812, + "eval_runtime": 44.9972, + "eval_samples_per_second": 22.224, + "eval_steps_per_second": 0.933, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.4818382263183594, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.644, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.4990177154541016, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.6962, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.627502918243408, + "learning_rate": 3.005686433793664e-06, + "loss": 1.639, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.5949175357818604, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.7095, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.422765016555786, + "learning_rate": 3.001624695369618e-06, + "loss": 1.6572, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.708702564239502, + "learning_rate": 2.999593826157596e-06, + "loss": 1.6747, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.624866247177124, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.6643, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.5831358432769775, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.7036, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.5799436569213867, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.6628, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.7047722339630127, + "learning_rate": 2.991470349309505e-06, + "loss": 1.6591, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.8736577033996582, + "eval_runtime": 44.7095, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.411628246307373, + "learning_rate": 2.989439480097482e-06, + "loss": 1.6644, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.8774638175964355, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.7192, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.857028007507324, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7079, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.4619641304016113, + "learning_rate": 2.983346872461414e-06, + "loss": 1.6846, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.6566193103790283, + "learning_rate": 2.981316003249391e-06, + "loss": 1.6768, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.681217670440674, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.7002, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.39505672454834, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.65, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.6062734127044678, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.6659, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.6402504444122314, + "learning_rate": 2.9731925264013e-06, + "loss": 1.6491, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.497014284133911, + "learning_rate": 2.971161657189277e-06, + "loss": 1.6383, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.874670147895813, + "eval_runtime": 42.9696, + "eval_samples_per_second": 23.272, + "eval_steps_per_second": 0.977, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.9070584774017334, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.6388, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.580744981765747, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7078, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.4637644290924072, + "learning_rate": 2.965069049553209e-06, + "loss": 1.6266, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.4489941596984863, + "learning_rate": 2.963038180341186e-06, + "loss": 1.705, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.664654016494751, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.6964, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.7017202377319336, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.6356, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.3691024780273438, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.6566, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.540384292602539, + "learning_rate": 2.954914703493095e-06, + "loss": 1.6334, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.5240535736083984, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.6316, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 3.105684518814087, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.6769, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.8741132020950317, + "eval_runtime": 42.6795, + "eval_samples_per_second": 23.43, + "eval_steps_per_second": 0.984, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.494252920150757, + "learning_rate": 2.948822095857027e-06, + "loss": 1.6359, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 4.069019317626953, + "learning_rate": 2.946791226645004e-06, + "loss": 1.621, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.641447067260742, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.6502, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.6700453758239746, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.6593, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.686068296432495, + "learning_rate": 2.940698619008936e-06, + "loss": 1.67, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.409055471420288, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.6809, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.4707298278808594, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.69, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.4735982418060303, + "learning_rate": 2.934606011372868e-06, + "loss": 1.666, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.633183479309082, + "learning_rate": 2.932575142160845e-06, + "loss": 1.6597, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.4808642864227295, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.6667, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.8731157779693604, + "eval_runtime": 41.3423, + "eval_samples_per_second": 24.188, + "eval_steps_per_second": 1.016, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.4044833183288574, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.7028, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.942699909210205, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.6678, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.505645990371704, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.5972, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.700967788696289, + "learning_rate": 2.922420796100731e-06, + "loss": 1.7064, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.4805755615234375, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.6201, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.753181219100952, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.6704, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.488865613937378, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7057, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.5464885234832764, + "learning_rate": 2.91429731925264e-06, + "loss": 1.6678, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.4997060298919678, + "learning_rate": 2.912266450040618e-06, + "loss": 1.662, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.5162436962127686, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.6501, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.869848370552063, + "eval_runtime": 43.3583, + "eval_samples_per_second": 23.064, + "eval_steps_per_second": 0.969, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.616129159927368, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.6206, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.761080503463745, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7131, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.5994064807891846, + "learning_rate": 2.904142973192527e-06, + "loss": 1.6693, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 3.4927968978881836, + "learning_rate": 2.902112103980504e-06, + "loss": 1.6694, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.6050939559936523, + "learning_rate": 2.900081234768481e-06, + "loss": 1.6419, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.796086311340332, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.6354, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.5658411979675293, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.652, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.3802385330200195, + "learning_rate": 2.893988627132413e-06, + "loss": 1.5902, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.479189872741699, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.6617, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.4367568492889404, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.6691, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.868497371673584, + "eval_runtime": 45.172, + "eval_samples_per_second": 22.138, + "eval_steps_per_second": 0.93, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.51888370513916, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.6872, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.5192768573760986, + "learning_rate": 2.885865150284322e-06, + "loss": 1.6584, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.5898501873016357, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.6179, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.698101043701172, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.663, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.889942169189453, + "learning_rate": 2.879772542648254e-06, + "loss": 1.5954, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.7909929752349854, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.7264, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 3.1931703090667725, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7091, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.7934632301330566, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.7204, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.470545530319214, + "learning_rate": 2.871649065800163e-06, + "loss": 1.644, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 3.011662006378174, + "learning_rate": 2.86961819658814e-06, + "loss": 1.6743, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.8678737878799438, + "eval_runtime": 44.7875, + "eval_samples_per_second": 22.328, + "eval_steps_per_second": 0.938, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.60823655128479, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.6467, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.5347230434417725, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.7044, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.481715440750122, + "learning_rate": 2.863525588952072e-06, + "loss": 1.5969, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.4966487884521484, + "learning_rate": 2.861494719740049e-06, + "loss": 1.6231, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.8809802532196045, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.6562, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.570204734802246, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.6511, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.6073813438415527, + "learning_rate": 2.855402112103981e-06, + "loss": 1.6456, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.8125052452087402, + "learning_rate": 2.853371242891958e-06, + "loss": 1.6722, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 5.341272354125977, + "learning_rate": 2.851340373679935e-06, + "loss": 1.7025, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 3.138002395629883, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.6676, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.8733265399932861, + "eval_runtime": 45.4915, + "eval_samples_per_second": 21.982, + "eval_steps_per_second": 0.923, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 7.264789581298828, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.6335, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 3.8533623218536377, + "learning_rate": 2.845247766043867e-06, + "loss": 1.6029, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 4.1789164543151855, + "learning_rate": 2.843216896831844e-06, + "loss": 1.6368, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 3.117408037185669, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.6924, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.721306085586548, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.6914, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.7099668979644775, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.6508, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 3.7823967933654785, + "learning_rate": 2.835093419983753e-06, + "loss": 1.636, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.696370840072632, + "learning_rate": 2.833062550771731e-06, + "loss": 1.6175, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.7303690910339355, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.6654, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.867727041244507, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.6717, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.8720332384109497, + "eval_runtime": 44.164, + "eval_samples_per_second": 22.643, + "eval_steps_per_second": 0.951, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 3.1209630966186523, + "learning_rate": 2.826969943135662e-06, + "loss": 1.6635, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.5057783126831055, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.6309, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.6650168895721436, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.6365, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.371305465698242, + "learning_rate": 2.820877335499594e-06, + "loss": 1.603, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.387404203414917, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.6189, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.6265947818756104, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.6511, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.6108627319335938, + "learning_rate": 2.814784727863526e-06, + "loss": 1.6727, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.7021450996398926, + "learning_rate": 2.812753858651503e-06, + "loss": 1.6334, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.470118999481201, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.6155, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5722644329071045, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.6788, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.8727209568023682, + "eval_runtime": 44.6253, + "eval_samples_per_second": 22.409, + "eval_steps_per_second": 0.941, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.7143876552581787, + "learning_rate": 2.806661251015435e-06, + "loss": 1.6956, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.8522002696990967, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.7558, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.5352110862731934, + "learning_rate": 2.802599512591389e-06, + "loss": 1.6592, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.7065908908843994, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.6121, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.616450786590576, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.6643, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.528763771057129, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7104, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 3.725569009780884, + "learning_rate": 2.794476035743298e-06, + "loss": 1.6713, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.487497091293335, + "learning_rate": 2.792445166531276e-06, + "loss": 1.692, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.534766912460327, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.6257, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.8553240299224854, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.6147, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.8675988912582397, + "eval_runtime": 46.4445, + "eval_samples_per_second": 21.531, + "eval_steps_per_second": 0.904, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.5054166316986084, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.6082, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.4133870601654053, + "learning_rate": 2.784321689683185e-06, + "loss": 1.6148, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.5863375663757324, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7032, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.58491849899292, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.679, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.5225989818573, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.6688, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.366861343383789, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.7076, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.4159810543060303, + "learning_rate": 2.774167343623071e-06, + "loss": 1.6301, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.5394275188446045, + "learning_rate": 2.772136474411048e-06, + "loss": 1.5758, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.409821033477783, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.664, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.4548892974853516, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.643, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.8672850131988525, + "eval_runtime": 46.3149, + "eval_samples_per_second": 21.591, + "eval_steps_per_second": 0.907, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.7483904361724854, + "learning_rate": 2.76604386677498e-06, + "loss": 1.6898, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.538804769515991, + "learning_rate": 2.764012997562957e-06, + "loss": 1.6019, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.7153313159942627, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.691, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.6892242431640625, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.6386, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.550957679748535, + "learning_rate": 2.757920389926889e-06, + "loss": 1.6237, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.3658559322357178, + "learning_rate": 2.755889520714866e-06, + "loss": 1.6015, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.6524102687835693, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.6644, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.7067437171936035, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.687, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.4044861793518066, + "learning_rate": 2.749796913078798e-06, + "loss": 1.6535, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.4933829307556152, + "learning_rate": 2.747766043866775e-06, + "loss": 1.7017, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.8652688264846802, + "eval_runtime": 46.8502, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.7062129974365234, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7398, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.5704996585845947, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.6561, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.5875790119171143, + "learning_rate": 2.741673436230707e-06, + "loss": 1.643, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.7498223781585693, + "learning_rate": 2.739642567018684e-06, + "loss": 1.6652, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.7275257110595703, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.6559, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.6349947452545166, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.6489, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.7058634757995605, + "learning_rate": 2.733549959382616e-06, + "loss": 1.7176, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.481344223022461, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.685, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.468277931213379, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.6305, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.6183228492736816, + "learning_rate": 2.727457351746548e-06, + "loss": 1.6151, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.864329218864441, + "eval_runtime": 43.0443, + "eval_samples_per_second": 23.232, + "eval_steps_per_second": 0.976, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.50679349899292, + "learning_rate": 2.725426482534525e-06, + "loss": 1.6468, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.6565401554107666, + "learning_rate": 2.723395613322502e-06, + "loss": 1.6496, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.5332367420196533, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.6424, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.6645619869232178, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.8012, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.7359142303466797, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.6025, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.660728693008423, + "learning_rate": 2.715272136474411e-06, + "loss": 1.6601, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.5184221267700195, + "learning_rate": 2.713241267262389e-06, + "loss": 1.6383, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.643712043762207, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.6834, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.504753351211548, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.6292, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.3439948558807373, + "learning_rate": 2.70714865962632e-06, + "loss": 1.6558, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.861581563949585, + "eval_runtime": 43.9224, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4820125102996826, + "learning_rate": 2.705117790414298e-06, + "loss": 1.69, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.6156861782073975, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.6304, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.490123987197876, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.6573, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.4439024925231934, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.6368, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.891603708267212, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7111, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.481175661087036, + "learning_rate": 2.694963444354184e-06, + "loss": 1.6408, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.5378847122192383, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.6782, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.592477321624756, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.6085, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.9336471557617188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.6719, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.3115012645721436, + "learning_rate": 2.686839967506093e-06, + "loss": 1.5797, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.8649041652679443, + "eval_runtime": 43.711, + "eval_samples_per_second": 22.878, + "eval_steps_per_second": 0.961, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.627028465270996, + "learning_rate": 2.68480909829407e-06, + "loss": 1.5784, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.917670488357544, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7007, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.418285608291626, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.6613, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.607773780822754, + "learning_rate": 2.678716490658002e-06, + "loss": 1.6561, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.569714307785034, + "learning_rate": 2.676685621445979e-06, + "loss": 1.6541, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.359494209289551, + "learning_rate": 2.674654752233956e-06, + "loss": 1.6467, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.5155694484710693, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.7003, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.439147710800171, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.5679, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.5044116973876953, + "learning_rate": 2.668562144597888e-06, + "loss": 1.6626, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.4475765228271484, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.656, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.8636877536773682, + "eval_runtime": 43.5138, + "eval_samples_per_second": 22.981, + "eval_steps_per_second": 0.965, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4616832733154297, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.6101, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.4125123023986816, + "learning_rate": 2.66246953696182e-06, + "loss": 1.6021, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.571326494216919, + "learning_rate": 2.660438667749797e-06, + "loss": 1.6482, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.424332618713379, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.653, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.3034703731536865, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.6096, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.4758505821228027, + "learning_rate": 2.654346060113729e-06, + "loss": 1.6274, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 3.800152540206909, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.6217, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.479827880859375, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7202, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.604523181915283, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.6778, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 3.595973491668701, + "learning_rate": 2.646222583265638e-06, + "loss": 1.6678, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.8647090196609497, + "eval_runtime": 43.0415, + "eval_samples_per_second": 23.233, + "eval_steps_per_second": 0.976, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 3.297319173812866, + "learning_rate": 2.644191714053615e-06, + "loss": 1.637, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.423567056655884, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.6018, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 4.4547953605651855, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.6594, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 3.0775582790374756, + "learning_rate": 2.638099106417547e-06, + "loss": 1.6259, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.747938871383667, + "learning_rate": 2.636068237205524e-06, + "loss": 1.7208, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.5350255966186523, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.5705, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 3.0630743503570557, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.6507, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 3.5475759506225586, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7045, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.534303903579712, + "learning_rate": 2.627944760357433e-06, + "loss": 1.7173, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.6964452266693115, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.6823, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.865207552909851, + "eval_runtime": 42.9435, + "eval_samples_per_second": 23.286, + "eval_steps_per_second": 0.978, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.924957752227783, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.6036, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.7058141231536865, + "learning_rate": 2.621852152721365e-06, + "loss": 1.6701, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.735339641571045, + "learning_rate": 2.619821283509342e-06, + "loss": 1.607, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.573958396911621, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.667, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5776426792144775, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.6938, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 4.454444408416748, + "learning_rate": 2.613728675873274e-06, + "loss": 1.7139, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.4883337020874023, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.6524, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.672116994857788, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.6197, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 3.511932611465454, + "learning_rate": 2.607636068237206e-06, + "loss": 1.6739, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.5960047245025635, + "learning_rate": 2.605605199025183e-06, + "loss": 1.6516, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.8604435920715332, + "eval_runtime": 42.9247, + "eval_samples_per_second": 23.297, + "eval_steps_per_second": 0.978, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.565376043319702, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.6884, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.676114797592163, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.6693, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.582751750946045, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.6499, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.748908281326294, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.6837, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.5578067302703857, + "learning_rate": 2.595450852965069e-06, + "loss": 1.6959, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.5702805519104004, + "learning_rate": 2.593419983753047e-06, + "loss": 1.6429, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.4748787879943848, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.6878, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.6157331466674805, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.6246, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.8377959728240967, + "learning_rate": 2.587327376116978e-06, + "loss": 1.6451, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.485398292541504, + "learning_rate": 2.585296506904956e-06, + "loss": 1.6288, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.8593864440917969, + "eval_runtime": 44.6341, + "eval_samples_per_second": 22.404, + "eval_steps_per_second": 0.941, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.46850323677063, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.5766, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.673283576965332, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7383, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.935293197631836, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.5742, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.531733751296997, + "learning_rate": 2.577173030056865e-06, + "loss": 1.6714, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.528425455093384, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6802, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.5644803047180176, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.6201, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.382451295852661, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.5824, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.63320255279541, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.6685, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.488844394683838, + "learning_rate": 2.567018683996751e-06, + "loss": 1.6078, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.7845425605773926, + "learning_rate": 2.564987814784728e-06, + "loss": 1.638, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.8572624921798706, + "eval_runtime": 43.2181, + "eval_samples_per_second": 23.138, + "eval_steps_per_second": 0.972, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.709656000137329, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.6425, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.651277542114258, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7401, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.5424256324768066, + "learning_rate": 2.55889520714866e-06, + "loss": 1.6272, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.6508235931396484, + "learning_rate": 2.556864337936637e-06, + "loss": 1.6384, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.882636308670044, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.6965, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.518367290496826, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.6551, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.7977044582366943, + "learning_rate": 2.550771730300569e-06, + "loss": 1.679, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.353041172027588, + "learning_rate": 2.548740861088546e-06, + "loss": 1.5745, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.5718655586242676, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.5655, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.6955313682556152, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.6764, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.8569272756576538, + "eval_runtime": 45.3844, + "eval_samples_per_second": 22.034, + "eval_steps_per_second": 0.925, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4756696224212646, + "learning_rate": 2.542648253452478e-06, + "loss": 1.6522, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.573193311691284, + "learning_rate": 2.540617384240455e-06, + "loss": 1.6079, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.5967671871185303, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.6341, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.7896342277526855, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.6449, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.519728183746338, + "learning_rate": 2.534524776604387e-06, + "loss": 1.549, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.543139934539795, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.6335, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.5580177307128906, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6402, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.443547010421753, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.6421, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.7075212001800537, + "learning_rate": 2.526401299756296e-06, + "loss": 1.6283, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.6505324840545654, + "learning_rate": 2.524370430544273e-06, + "loss": 1.6028, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.8556184768676758, + "eval_runtime": 43.731, + "eval_samples_per_second": 22.867, + "eval_steps_per_second": 0.96, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.597172260284424, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.6631, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 3.1529037952423096, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.6219, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.5374975204467773, + "learning_rate": 2.518277822908205e-06, + "loss": 1.6476, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 3.0557401180267334, + "learning_rate": 2.516246953696182e-06, + "loss": 1.6627, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 6.759350776672363, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.6532, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.670427083969116, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.658, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 4.269686698913574, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.5873, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 13.37558650970459, + "learning_rate": 2.508123476848091e-06, + "loss": 1.6872, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 7.060877323150635, + "learning_rate": 2.506092607636069e-06, + "loss": 1.6406, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 11.259824752807617, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.6421, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.861360788345337, + "eval_runtime": 45.4438, + "eval_samples_per_second": 22.005, + "eval_steps_per_second": 0.924, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 5.861843585968018, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.701, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 3.9136719703674316, + "learning_rate": 2.5e-06, + "loss": 1.6289, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 12.720739364624023, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.6532, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 5.970798492431641, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7035, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 3.490950584411621, + "learning_rate": 2.493907392363932e-06, + "loss": 1.6565, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 5.826146602630615, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.655, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 9.610876083374023, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.7295, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 3.754833221435547, + "learning_rate": 2.487814784727864e-06, + "loss": 1.6406, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.7952232360839844, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.6463, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 6.3178558349609375, + "learning_rate": 2.483753046303818e-06, + "loss": 1.6474, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.8672106266021729, + "eval_runtime": 50.0163, + "eval_samples_per_second": 19.993, + "eval_steps_per_second": 0.84, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 7.4854960441589355, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.6798, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.8903026580810547, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.667, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.481947898864746, + "learning_rate": 2.47766043866775e-06, + "loss": 1.6677, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.8459818363189697, + "learning_rate": 2.475629569455727e-06, + "loss": 1.6129, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 5.00568962097168, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.6233, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 3.7508153915405273, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.6414, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.7895889282226562, + "learning_rate": 2.469536961819659e-06, + "loss": 1.65, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.532818555831909, + "learning_rate": 2.467506092607636e-06, + "loss": 1.6174, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.699639320373535, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7128, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.5636072158813477, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.6416, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.8608473539352417, + "eval_runtime": 43.8315, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.7466976642608643, + "learning_rate": 2.461413484971568e-06, + "loss": 1.7246, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.690990686416626, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.596, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.868340492248535, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.622, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.4456427097320557, + "learning_rate": 2.4553208773355e-06, + "loss": 1.6446, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.5499041080474854, + "learning_rate": 2.453290008123477e-06, + "loss": 1.6587, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.68770170211792, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.6612, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.589557647705078, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.6428, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.7014052867889404, + "learning_rate": 2.447197400487409e-06, + "loss": 1.6605, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.6242196559906006, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.6897, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.5285110473632812, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.6636, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.8544306755065918, + "eval_runtime": 45.6109, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.484952449798584, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.6636, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.5793070793151855, + "learning_rate": 2.439073923639318e-06, + "loss": 1.6044, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.5558605194091797, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.5885, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.6365678310394287, + "learning_rate": 2.435012185215272e-06, + "loss": 1.6236, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.428422451019287, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.6446, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.535776138305664, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.6252, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.743934154510498, + "learning_rate": 2.428919577579204e-06, + "loss": 1.729, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.4823739528656006, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.6251, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.4650003910064697, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.6191, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.4338109493255615, + "learning_rate": 2.422826969943136e-06, + "loss": 1.5919, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.8501240015029907, + "eval_runtime": 43.8558, + "eval_samples_per_second": 22.802, + "eval_steps_per_second": 0.958, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.6054813861846924, + "learning_rate": 2.420796100731113e-06, + "loss": 1.6485, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.390143394470215, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.661, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.467470645904541, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.6308, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.360908269882202, + "learning_rate": 2.414703493095045e-06, + "loss": 1.6394, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.5966176986694336, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.6138, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.528402328491211, + "learning_rate": 2.410641754670999e-06, + "loss": 1.6817, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.5319254398345947, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.6189, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.559260129928589, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.6476, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.5595016479492188, + "learning_rate": 2.404549147034931e-06, + "loss": 1.6878, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.488196849822998, + "learning_rate": 2.402518277822908e-06, + "loss": 1.6095, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.8484610319137573, + "eval_runtime": 43.4829, + "eval_samples_per_second": 22.998, + "eval_steps_per_second": 0.966, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6928117275238037, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.6533, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.485538959503174, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.6174, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.4725208282470703, + "learning_rate": 2.39642567018684e-06, + "loss": 1.6306, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.456028461456299, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.6586, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3731250762939453, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.6009, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.4567086696624756, + "learning_rate": 2.390333062550772e-06, + "loss": 1.5257, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.577141284942627, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7127, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.4612233638763428, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.638, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.463212728500366, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.5933, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.5121495723724365, + "learning_rate": 2.382209585702681e-06, + "loss": 1.68, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.8499811887741089, + "eval_runtime": 45.066, + "eval_samples_per_second": 22.19, + "eval_steps_per_second": 0.932, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.658245801925659, + "learning_rate": 2.380178716490658e-06, + "loss": 1.6354, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.6265509128570557, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.6755, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 3.016807794570923, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.6764, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.6552720069885254, + "learning_rate": 2.37408610885459e-06, + "loss": 1.6281, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.520078659057617, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.7077, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.528329849243164, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.6494, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.623776912689209, + "learning_rate": 2.367993501218522e-06, + "loss": 1.6436, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.56669020652771, + "learning_rate": 2.365962632006499e-06, + "loss": 1.6762, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.5415968894958496, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.6711, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.5164198875427246, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.6725, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.8490949869155884, + "eval_runtime": 43.612, + "eval_samples_per_second": 22.929, + "eval_steps_per_second": 0.963, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.626594066619873, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.6637, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3480069637298584, + "learning_rate": 2.357839155158408e-06, + "loss": 1.6561, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.644103527069092, + "learning_rate": 2.355808285946385e-06, + "loss": 1.6763, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.5539023876190186, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.653, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.5055317878723145, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.6643, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.4204092025756836, + "learning_rate": 2.349715678310317e-06, + "loss": 1.6133, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.5867843627929688, + "learning_rate": 2.347684809098294e-06, + "loss": 1.6804, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.5993685722351074, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.6101, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.565439462661743, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.6221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.5296084880828857, + "learning_rate": 2.341592201462226e-06, + "loss": 1.6234, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.8482695817947388, + "eval_runtime": 45.1972, + "eval_samples_per_second": 22.125, + "eval_steps_per_second": 0.929, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.6142499446868896, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.5886, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.486616849899292, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.6097, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.5778701305389404, + "learning_rate": 2.335499593826158e-06, + "loss": 1.6585, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.6165313720703125, + "learning_rate": 2.333468724614135e-06, + "loss": 1.663, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.4976565837860107, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.6083, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.485017776489258, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.6006, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.620819330215454, + "learning_rate": 2.327376116978067e-06, + "loss": 1.6641, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.528599262237549, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.6128, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.541442632675171, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.6545, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.4722139835357666, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.6146, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.848878264427185, + "eval_runtime": 43.7961, + "eval_samples_per_second": 22.833, + "eval_steps_per_second": 0.959, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.6106903553009033, + "learning_rate": 2.319252640129976e-06, + "loss": 1.7086, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.5255045890808105, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.6738, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.43829083442688, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.5861, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.527411460876465, + "learning_rate": 2.313160032493908e-06, + "loss": 1.6634, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.4998528957366943, + "learning_rate": 2.311129163281885e-06, + "loss": 1.6187, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.4838368892669678, + "learning_rate": 2.309098294069862e-06, + "loss": 1.5999, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.466900110244751, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.6446, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.583103895187378, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.6658, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.478386163711548, + "learning_rate": 2.303005686433794e-06, + "loss": 1.6261, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.5724942684173584, + "learning_rate": 2.300974817221771e-06, + "loss": 1.5955, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.8444501161575317, + "eval_runtime": 50.0263, + "eval_samples_per_second": 19.989, + "eval_steps_per_second": 0.84, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.601644277572632, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.6493, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.5638768672943115, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.6702, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.4802567958831787, + "learning_rate": 2.294882209585703e-06, + "loss": 1.6036, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.5114285945892334, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.642, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.487563133239746, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.673, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.6014463901519775, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.6261, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.53680157661438, + "learning_rate": 2.286758732737612e-06, + "loss": 1.6364, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.764408588409424, + "learning_rate": 2.284727863525589e-06, + "loss": 1.6791, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.8998825550079346, + "learning_rate": 2.282696994313566e-06, + "loss": 1.644, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.5797383785247803, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.6267, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.8467868566513062, + "eval_runtime": 49.4778, + "eval_samples_per_second": 20.211, + "eval_steps_per_second": 0.849, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.447178602218628, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.5955, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.6275408267974854, + "learning_rate": 2.276604386677498e-06, + "loss": 1.6117, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.6952037811279297, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.6894, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.5299718379974365, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.6361, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.837455987930298, + "learning_rate": 2.27051177904143e-06, + "loss": 1.6269, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.5406227111816406, + "learning_rate": 2.268480909829407e-06, + "loss": 1.6538, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.716038227081299, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.6691, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.883439064025879, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.6298, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.7631118297576904, + "learning_rate": 2.262388302193339e-06, + "loss": 1.6245, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.6777117252349854, + "learning_rate": 2.260357432981316e-06, + "loss": 1.6446, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.845224142074585, + "eval_runtime": 46.8476, + "eval_samples_per_second": 21.346, + "eval_steps_per_second": 0.897, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.6280243396759033, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.6398, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.471925973892212, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.6403, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.6293203830718994, + "learning_rate": 2.254264825345248e-06, + "loss": 1.642, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.651144027709961, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.6445, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.6489877700805664, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.6761, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.5869088172912598, + "learning_rate": 2.24817221770918e-06, + "loss": 1.6593, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.411802053451538, + "learning_rate": 2.246141348497157e-06, + "loss": 1.6232, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.522610902786255, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.6358, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.5430991649627686, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.6605, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.4440951347351074, + "learning_rate": 2.240048740861089e-06, + "loss": 1.5354, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.8440407514572144, + "eval_runtime": 48.0239, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.5492775440216064, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.668, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.7511909008026123, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6424, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.5176775455474854, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.6336, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.5408577919006348, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.5843, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.4791393280029297, + "learning_rate": 2.229894394800975e-06, + "loss": 1.555, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.383148670196533, + "learning_rate": 2.227863525588952e-06, + "loss": 1.598, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.429348945617676, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.5943, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.708796501159668, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.6382, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 3.0029661655426025, + "learning_rate": 2.221770917952884e-06, + "loss": 1.6699, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.5410728454589844, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.5965, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.8443207740783691, + "eval_runtime": 47.8671, + "eval_samples_per_second": 20.891, + "eval_steps_per_second": 0.877, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.5718178749084473, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.6214, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.474177837371826, + "learning_rate": 2.215678310316816e-06, + "loss": 1.6498, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.5960052013397217, + "learning_rate": 2.213647441104793e-06, + "loss": 1.6156, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.6217992305755615, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.6311, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.5601158142089844, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.6008, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.7786355018615723, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.62, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.548356771469116, + "learning_rate": 2.205523964256702e-06, + "loss": 1.6473, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.6265289783477783, + "learning_rate": 2.203493095044679e-06, + "loss": 1.6565, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.522379159927368, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.6764, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.5479297637939453, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.6194, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.8434110879898071, + "eval_runtime": 46.6043, + "eval_samples_per_second": 21.457, + "eval_steps_per_second": 0.901, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.5401806831359863, + "learning_rate": 2.197400487408611e-06, + "loss": 1.5911, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.466893434524536, + "learning_rate": 2.195369618196588e-06, + "loss": 1.6145, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.566653251647949, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.6558, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.664053201675415, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7089, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 3.6594645977020264, + "learning_rate": 2.18927701056052e-06, + "loss": 1.6549, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.68450927734375, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.6306, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.7094614505767822, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.6147, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.481332540512085, + "learning_rate": 2.183184402924452e-06, + "loss": 1.6612, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5651051998138428, + "learning_rate": 2.181153533712429e-06, + "loss": 1.6394, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.5570919513702393, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.6162, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.8433003425598145, + "eval_runtime": 46.8673, + "eval_samples_per_second": 21.337, + "eval_steps_per_second": 0.896, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.6206037998199463, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.6869, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.8962907791137695, + "learning_rate": 2.175060926076361e-06, + "loss": 1.7022, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.6211588382720947, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.6249, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.7770566940307617, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.5821, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.543658494949341, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.6587, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.508188486099243, + "learning_rate": 2.16693744922827e-06, + "loss": 1.6114, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.703335762023926, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.6152, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.469688892364502, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.6121, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.4836199283599854, + "learning_rate": 2.160844841592202e-06, + "loss": 1.5918, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.521851062774658, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.6228, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.845065712928772, + "eval_runtime": 45.3003, + "eval_samples_per_second": 22.075, + "eval_steps_per_second": 0.927, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.7808425426483154, + "learning_rate": 2.156783103168156e-06, + "loss": 1.5887, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.696708917617798, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.6517, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.7222893238067627, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.6817, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.504417896270752, + "learning_rate": 2.150690495532088e-06, + "loss": 1.647, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.574836015701294, + "learning_rate": 2.148659626320065e-06, + "loss": 1.6554, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.5004355907440186, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.6089, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.6979293823242188, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.6226, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.5239524841308594, + "learning_rate": 2.142567018683997e-06, + "loss": 1.6875, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.5218467712402344, + "learning_rate": 2.140536149471974e-06, + "loss": 1.6095, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.5778310298919678, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.6155, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.8443745374679565, + "eval_runtime": 46.6273, + "eval_samples_per_second": 21.447, + "eval_steps_per_second": 0.901, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.671973705291748, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.6696, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.5520615577697754, + "learning_rate": 2.134443541835906e-06, + "loss": 1.6291, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.39286208152771, + "learning_rate": 2.132412672623883e-06, + "loss": 1.6012, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.8277461528778076, + "learning_rate": 2.13038180341186e-06, + "loss": 1.6398, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.7147881984710693, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.6685, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.499044418334961, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.6128, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.514784574508667, + "learning_rate": 2.124289195775792e-06, + "loss": 1.6263, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.505342483520508, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.5532, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.608139753341675, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.5946, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.5578057765960693, + "learning_rate": 2.118196588139724e-06, + "loss": 1.6103, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.8425958156585693, + "eval_runtime": 44.206, + "eval_samples_per_second": 22.621, + "eval_steps_per_second": 0.95, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.4881014823913574, + "learning_rate": 2.116165718927701e-06, + "loss": 1.6364, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.4141550064086914, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.5943, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.649827718734741, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.5994, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.547238826751709, + "learning_rate": 2.110073111291633e-06, + "loss": 1.6247, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.6515395641326904, + "learning_rate": 2.10804224207961e-06, + "loss": 1.6543, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.5575809478759766, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.6618, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.466344118118286, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.5437, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4897878170013428, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6199, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.6578829288482666, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.5989, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.6173501014709473, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.6392, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.838721752166748, + "eval_runtime": 49.4357, + "eval_samples_per_second": 20.228, + "eval_steps_per_second": 0.85, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.5655064582824707, + "learning_rate": 2.095857026807474e-06, + "loss": 1.6276, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.685790777206421, + "learning_rate": 2.093826157595451e-06, + "loss": 1.6142, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.7278990745544434, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.6292, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.5769667625427246, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.6774, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.5120608806610107, + "learning_rate": 2.087733549959383e-06, + "loss": 1.5972, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.4658212661743164, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.5683, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.526524782180786, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.6687, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.3989250659942627, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.6184, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.589115858078003, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.6153, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.4985275268554688, + "learning_rate": 2.077579203899269e-06, + "loss": 1.5682, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.8374594449996948, + "eval_runtime": 47.0329, + "eval_samples_per_second": 21.262, + "eval_steps_per_second": 0.893, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5770835876464844, + "learning_rate": 2.075548334687246e-06, + "loss": 1.6437, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.5111560821533203, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.5711, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.6110589504241943, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.6125, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.6229705810546875, + "learning_rate": 2.069455727051178e-06, + "loss": 1.595, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.648716926574707, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.5908, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.673701763153076, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.6652, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.667478561401367, + "learning_rate": 2.06336311941511e-06, + "loss": 1.6088, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 3.5688586235046387, + "learning_rate": 2.061332250203087e-06, + "loss": 1.5852, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.512603282928467, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.6218, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.5296075344085693, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.6382, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.8406264781951904, + "eval_runtime": 43.8351, + "eval_samples_per_second": 22.813, + "eval_steps_per_second": 0.958, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.598994255065918, + "learning_rate": 2.055239642567019e-06, + "loss": 1.6263, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.93025279045105, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.6502, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 3.3655412197113037, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.5872, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.955120325088501, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.6404, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.5226638317108154, + "learning_rate": 2.047116165718928e-06, + "loss": 1.6102, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.5462093353271484, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.6757, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 4.066190242767334, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.6072, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.8548827171325684, + "learning_rate": 2.04102355808286e-06, + "loss": 1.5717, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.6206634044647217, + "learning_rate": 2.038992688870837e-06, + "loss": 1.5943, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.8793396949768066, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.6121, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.83925199508667, + "eval_runtime": 44.747, + "eval_samples_per_second": 22.348, + "eval_steps_per_second": 0.939, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.6215157508850098, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.6492, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.6103456020355225, + "learning_rate": 2.032900081234769e-06, + "loss": 1.6531, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.485775947570801, + "learning_rate": 2.030869212022746e-06, + "loss": 1.6086, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.399557113647461, + "learning_rate": 2.028838342810723e-06, + "loss": 1.6126, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.4711368083953857, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.5985, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.654580593109131, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.6557, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.5091657638549805, + "learning_rate": 2.022745735174655e-06, + "loss": 1.6181, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.5723490715026855, + "learning_rate": 2.020714865962632e-06, + "loss": 1.6421, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4211535453796387, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.6226, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.5550055503845215, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.5729, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.8397995233535767, + "eval_runtime": 44.4762, + "eval_samples_per_second": 22.484, + "eval_steps_per_second": 0.944, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.618758201599121, + "learning_rate": 2.014622258326564e-06, + "loss": 1.6356, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.6717913150787354, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.6246, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.5464835166931152, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.6027, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.557669162750244, + "learning_rate": 2.008529650690496e-06, + "loss": 1.6531, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.581970453262329, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.6076, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.609236478805542, + "learning_rate": 2.00446791226645e-06, + "loss": 1.5932, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.4249160289764404, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.6372, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.5093257427215576, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.6627, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.3989412784576416, + "learning_rate": 1.998375304630382e-06, + "loss": 1.6292, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4299519062042236, + "learning_rate": 1.996344435418359e-06, + "loss": 1.6385, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.8367400169372559, + "eval_runtime": 45.921, + "eval_samples_per_second": 21.777, + "eval_steps_per_second": 0.915, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.509530782699585, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.6441, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.591426134109497, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.6382, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.6569297313690186, + "learning_rate": 1.990251827782291e-06, + "loss": 1.6469, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.6453206539154053, + "learning_rate": 1.988220958570268e-06, + "loss": 1.6348, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.5522994995117188, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.6311, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.5981009006500244, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.585, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.617323637008667, + "learning_rate": 1.9821283509342e-06, + "loss": 1.584, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.6743686199188232, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.6334, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.418383836746216, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.5934, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.7952582836151123, + "learning_rate": 1.976035743298132e-06, + "loss": 1.6779, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.8379877805709839, + "eval_runtime": 46.545, + "eval_samples_per_second": 21.485, + "eval_steps_per_second": 0.902, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.5364768505096436, + "learning_rate": 1.974004874086109e-06, + "loss": 1.5802, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.446056365966797, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6141, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.527529239654541, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.6042, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.5953001976013184, + "learning_rate": 1.967912266450041e-06, + "loss": 1.5959, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.545994758605957, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.7091, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.3900482654571533, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.5946, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.51353120803833, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.6166, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.550389528274536, + "learning_rate": 1.95978878960195e-06, + "loss": 1.6817, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.6815762519836426, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.6594, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.591909885406494, + "learning_rate": 1.955727051177904e-06, + "loss": 1.5991, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.8390737771987915, + "eval_runtime": 46.8495, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.5111982822418213, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.5993, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.655891180038452, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.5915, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.471090316772461, + "learning_rate": 1.949634443541836e-06, + "loss": 1.5867, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.8599209785461426, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.6485, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4986093044281006, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.6084, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.6345789432525635, + "learning_rate": 1.943541835905768e-06, + "loss": 1.6719, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.5636799335479736, + "learning_rate": 1.941510966693745e-06, + "loss": 1.5817, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.5867199897766113, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.6498, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.6369481086730957, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.6791, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.4554147720336914, + "learning_rate": 1.935418359057677e-06, + "loss": 1.6319, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.8408559560775757, + "eval_runtime": 49.3515, + "eval_samples_per_second": 20.263, + "eval_steps_per_second": 0.851, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.5412509441375732, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.6673, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.5755672454833984, + "learning_rate": 1.931356620633631e-06, + "loss": 1.5636, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.602530002593994, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.6261, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.755247116088867, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.5817, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.528991460800171, + "learning_rate": 1.925264012997563e-06, + "loss": 1.6173, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.473057508468628, + "learning_rate": 1.92323314378554e-06, + "loss": 1.648, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.62630033493042, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.5864, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.7876687049865723, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.5804, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.491473436355591, + "learning_rate": 1.917140536149472e-06, + "loss": 1.5716, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.656989336013794, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.6595, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.839701771736145, + "eval_runtime": 44.5976, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.6800942420959473, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.6415, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.4725751876831055, + "learning_rate": 1.911047928513404e-06, + "loss": 1.5901, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.552624464035034, + "learning_rate": 1.909017059301381e-06, + "loss": 1.5956, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.6042184829711914, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.6328, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.5159223079681396, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.604, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.488041639328003, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.5539, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.659151077270508, + "learning_rate": 1.90089358245329e-06, + "loss": 1.5999, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.700378179550171, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.6485, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.3943607807159424, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.5736, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.8171095848083496, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6259, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.8379307985305786, + "eval_runtime": 48.6918, + "eval_samples_per_second": 20.537, + "eval_steps_per_second": 0.863, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.562655210494995, + "learning_rate": 1.892770105605199e-06, + "loss": 1.6439, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.5640337467193604, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.6113, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.5324809551239014, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.6331, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.393224000930786, + "learning_rate": 1.886677497969131e-06, + "loss": 1.5228, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 3.073945999145508, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.632, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.7066094875335693, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.5508, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.981933832168579, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6973, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.903444528579712, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.6036, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.8616178035736084, + "learning_rate": 1.876523151909017e-06, + "loss": 1.5723, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.8243541717529297, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.555, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.8437565565109253, + "eval_runtime": 48.5449, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.803142547607422, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.5861, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 3.146677017211914, + "learning_rate": 1.870430544272949e-06, + "loss": 1.6791, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 3.2554781436920166, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.5626, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.955395460128784, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.5893, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.941377878189087, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.5962, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.765620708465576, + "learning_rate": 1.862307067424858e-06, + "loss": 1.595, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.8529839515686035, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.6296, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.7770590782165527, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.5711, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.827406883239746, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.5966, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.927119016647339, + "learning_rate": 1.854183590576767e-06, + "loss": 1.6463, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.841707468032837, + "eval_runtime": 49.0965, + "eval_samples_per_second": 20.368, + "eval_steps_per_second": 0.855, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.8223392963409424, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.559, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.862473726272583, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.5189, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.9060091972351074, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.5631, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.8727498054504395, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6013, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.862298011779785, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.549, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.8221330642700195, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.6341, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.938336133956909, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.602, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.719489812850952, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.5626, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.6615874767303467, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.5535, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.835413932800293, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.5742, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.8375873565673828, + "eval_runtime": 45.731, + "eval_samples_per_second": 21.867, + "eval_steps_per_second": 0.918, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.773707389831543, + "learning_rate": 1.831844029244517e-06, + "loss": 1.5499, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 3.0959930419921875, + "learning_rate": 1.829813160032494e-06, + "loss": 1.5656, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.8521792888641357, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.5982, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.970618486404419, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.5731, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 3.0716512203216553, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.5824, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.8776299953460693, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.6283, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.72183895111084, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.5422, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.853234052658081, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.5533, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.7193424701690674, + "learning_rate": 1.815597075548335e-06, + "loss": 1.5923, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.9768450260162354, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.6277, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.8401803970336914, + "eval_runtime": 45.8062, + "eval_samples_per_second": 21.831, + "eval_steps_per_second": 0.917, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.752772569656372, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.5241, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.784980297088623, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.5657, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.7485857009887695, + "learning_rate": 1.807473598700244e-06, + "loss": 1.5501, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.7191033363342285, + "learning_rate": 1.805442729488221e-06, + "loss": 1.5718, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.697540044784546, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.5804, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.7175710201263428, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.5809, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.764824151992798, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.545, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.946669816970825, + "learning_rate": 1.79731925264013e-06, + "loss": 1.595, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.689326524734497, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.5676, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.759997606277466, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.6032, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.8379274606704712, + "eval_runtime": 45.5941, + "eval_samples_per_second": 21.933, + "eval_steps_per_second": 0.921, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.8656060695648193, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.6086, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.7155873775482178, + "learning_rate": 1.789195775792039e-06, + "loss": 1.5218, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.8487389087677, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6285, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.8070144653320312, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.5479, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.856282949447632, + "learning_rate": 1.783103168155971e-06, + "loss": 1.5689, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.7465922832489014, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.5733, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.824997901916504, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.5583, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.8030786514282227, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.5383, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.911377429962158, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.586, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.7452991008758545, + "learning_rate": 1.772948822095857e-06, + "loss": 1.6409, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.8399248123168945, + "eval_runtime": 46.1241, + "eval_samples_per_second": 21.681, + "eval_steps_per_second": 0.911, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.8255481719970703, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.6134, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.865100622177124, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.5791, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.740133047103882, + "learning_rate": 1.766856214459789e-06, + "loss": 1.5651, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.7219033241271973, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.5453, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.7266135215759277, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.6375, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.8476688861846924, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.5877, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.810105085372925, + "learning_rate": 1.758732737611698e-06, + "loss": 1.5796, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.78798246383667, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.5708, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.750525951385498, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.5567, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.818673849105835, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.5661, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.8374614715576172, + "eval_runtime": 43.5618, + "eval_samples_per_second": 22.956, + "eval_steps_per_second": 0.964, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.7967586517333984, + "learning_rate": 1.750609260763607e-06, + "loss": 1.5846, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.766272783279419, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.5721, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.8006162643432617, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.5261, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 3.714587688446045, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.6599, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 3.143052339553833, + "learning_rate": 1.742485783915516e-06, + "loss": 1.659, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.8256232738494873, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.5076, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.8843302726745605, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.5862, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 6.692182540893555, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.6144, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.928053617477417, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.571, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.8071024417877197, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.5861, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.837381362915039, + "eval_runtime": 44.7089, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 3.070399522781372, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.5636, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 3.298100709915161, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.5717, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 5.891775131225586, + "learning_rate": 1.726238830219334e-06, + "loss": 1.5757, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.838676691055298, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.6369, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.8816466331481934, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.5885, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 3.1703031063079834, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.5966, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 3.504777669906616, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.4932, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 3.1451492309570312, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.5071, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.7158446311950684, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.5725, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.8391170501708984, + "learning_rate": 1.712022745735175e-06, + "loss": 1.5742, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.8371202945709229, + "eval_runtime": 43.2464, + "eval_samples_per_second": 23.123, + "eval_steps_per_second": 0.971, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.8346052169799805, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.5808, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.767813205718994, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.5209, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.7955846786499023, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.581, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.8616271018981934, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.5758, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.7975544929504395, + "learning_rate": 1.701868399675061e-06, + "loss": 1.5912, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 3.095912456512451, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.6522, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 3.102691888809204, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.6004, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.8405165672302246, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.5649, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.8038063049316406, + "learning_rate": 1.69374492282697e-06, + "loss": 1.545, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.8661468029022217, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6297, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.8386919498443604, + "eval_runtime": 46.0334, + "eval_samples_per_second": 21.723, + "eval_steps_per_second": 0.912, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.8217391967773438, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.5694, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 3.099724769592285, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.5986, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.9951272010803223, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.5429, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.9366989135742188, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.5666, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 3.0171356201171875, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.6212, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.952259063720703, + "learning_rate": 1.679528838342811e-06, + "loss": 1.5611, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.667905330657959, + "learning_rate": 1.677497969130788e-06, + "loss": 1.5556, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.8001160621643066, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.5854, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.6954760551452637, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.5278, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.8026583194732666, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.5103, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.8384881019592285, + "eval_runtime": 46.1899, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.898869752883911, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6188, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 3.0432069301605225, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.5945, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.7744672298431396, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.5642, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.8844289779663086, + "learning_rate": 1.663281884646629e-06, + "loss": 1.5657, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 3.0022218227386475, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.6297, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.918320417404175, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.5928, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.913848876953125, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.6219, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.954702615737915, + "learning_rate": 1.655158407798538e-06, + "loss": 1.5812, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.7966420650482178, + "learning_rate": 1.653127538586515e-06, + "loss": 1.6034, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 3.1328368186950684, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.6168, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.836020827293396, + "eval_runtime": 46.1979, + "eval_samples_per_second": 21.646, + "eval_steps_per_second": 0.909, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.8109259605407715, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.6091, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.823570728302002, + "learning_rate": 1.647034930950447e-06, + "loss": 1.5504, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.885016918182373, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.5784, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 3.612942934036255, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.6014, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.8981637954711914, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.5552, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.6703410148620605, + "learning_rate": 1.638911454102356e-06, + "loss": 1.5567, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.7087886333465576, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.613, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.7711687088012695, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.5586, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.8473854064941406, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.565, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 3.783824920654297, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.5909, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.8384798765182495, + "eval_runtime": 46.2339, + "eval_samples_per_second": 21.629, + "eval_steps_per_second": 0.908, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.9971532821655273, + "learning_rate": 1.628757108042242e-06, + "loss": 1.6251, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.8581976890563965, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.5632, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.9119839668273926, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.5806, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.840101718902588, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.6706, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.8574070930480957, + "learning_rate": 1.620633631194151e-06, + "loss": 1.6138, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.820692539215088, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.5534, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7690017223358154, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.5922, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 3.062229871749878, + "learning_rate": 1.614541023558083e-06, + "loss": 1.5893, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.727062225341797, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.5851, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.7693302631378174, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.6031, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.8332058191299438, + "eval_runtime": 45.0301, + "eval_samples_per_second": 22.207, + "eval_steps_per_second": 0.933, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.803758382797241, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.5504, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.775770902633667, + "learning_rate": 1.606417546709992e-06, + "loss": 1.5707, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.7698636054992676, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.5763, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.739940881729126, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.5386, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.7594099044799805, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.5805, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.853273391723633, + "learning_rate": 1.598294069861901e-06, + "loss": 1.5548, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.9859204292297363, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6278, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.742898941040039, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.555, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.769071340560913, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.5504, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.7191832065582275, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6014, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.8336049318313599, + "eval_runtime": 47.0078, + "eval_samples_per_second": 21.273, + "eval_steps_per_second": 0.893, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.842529773712158, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.5723, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.768066883087158, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.5272, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7481656074523926, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.5843, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.806687593460083, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.5598, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.8214614391326904, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.5972, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.8541247844696045, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.5988, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 3.000246286392212, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.583, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.9164586067199707, + "learning_rate": 1.573923639317628e-06, + "loss": 1.5949, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.8099935054779053, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.5959, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.88322377204895, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.6062, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.832599401473999, + "eval_runtime": 45.9576, + "eval_samples_per_second": 21.759, + "eval_steps_per_second": 0.914, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.8019371032714844, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.5445, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.8585665225982666, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6079, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.7423956394195557, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.5537, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.7420766353607178, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.5846, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.8864388465881348, + "learning_rate": 1.559707554833469e-06, + "loss": 1.6213, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.7483463287353516, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.5012, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.7340080738067627, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.555, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.8000667095184326, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.5259, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.9176089763641357, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.5726, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.9147658348083496, + "learning_rate": 1.549553208773355e-06, + "loss": 1.5685, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.834657907485962, + "eval_runtime": 45.8709, + "eval_samples_per_second": 21.8, + "eval_steps_per_second": 0.916, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.9109392166137695, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.5533, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 3.0263078212738037, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.6277, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.9347317218780518, + "learning_rate": 1.543460601137287e-06, + "loss": 1.5738, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.8229548931121826, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.5969, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.906954288482666, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.5772, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.9214975833892822, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.5147, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.844081163406372, + "learning_rate": 1.535337124289196e-06, + "loss": 1.5635, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 3.0214555263519287, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.6466, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.9254050254821777, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.5837, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.8986024856567383, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.5997, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.8318188190460205, + "eval_runtime": 46.7695, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.974231719970703, + "learning_rate": 1.5272136474411047e-06, + "loss": 1.6143, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.7890946865081787, + "learning_rate": 1.525182778229082e-06, + "loss": 1.5956, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.820800304412842, + "learning_rate": 1.5231519090170593e-06, + "loss": 1.5626, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.84753680229187, + "learning_rate": 1.5211210398050365e-06, + "loss": 1.5473, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.829115152359009, + "learning_rate": 1.5190901705930138e-06, + "loss": 1.591, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.786787509918213, + "learning_rate": 1.517059301380991e-06, + "loss": 1.565, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.7778282165527344, + "learning_rate": 1.5150284321689684e-06, + "loss": 1.5666, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.9727330207824707, + "learning_rate": 1.5129975629569456e-06, + "loss": 1.577, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.876600980758667, + "learning_rate": 1.510966693744923e-06, + "loss": 1.5742, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.7339680194854736, + "learning_rate": 1.5089358245329002e-06, + "loss": 1.5378, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 1.834323763847351, + "eval_runtime": 44.2017, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.9201254844665527, + "learning_rate": 1.5069049553208775e-06, + "loss": 1.5962, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.8632285594940186, + "learning_rate": 1.5048740861088547e-06, + "loss": 1.589, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.9003729820251465, + "learning_rate": 1.502843216896832e-06, + "loss": 1.5125, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.81120228767395, + "learning_rate": 1.500812347684809e-06, + "loss": 1.5809, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.747528553009033, + "learning_rate": 1.4987814784727864e-06, + "loss": 1.5611, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.6891186237335205, + "learning_rate": 1.4967506092607636e-06, + "loss": 1.5387, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.9105594158172607, + "learning_rate": 1.494719740048741e-06, + "loss": 1.5921, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.8186421394348145, + "learning_rate": 1.4926888708367182e-06, + "loss": 1.6075, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 3.0198328495025635, + "learning_rate": 1.4906580016246955e-06, + "loss": 1.611, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 3.0839414596557617, + "learning_rate": 1.4886271324126727e-06, + "loss": 1.6298, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 1.8325375318527222, + "eval_runtime": 53.1022, + "eval_samples_per_second": 18.832, + "eval_steps_per_second": 0.791, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.8375303745269775, + "learning_rate": 1.48659626320065e-06, + "loss": 1.4942, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.857344627380371, + "learning_rate": 1.4845653939886273e-06, + "loss": 1.5182, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.7296979427337646, + "learning_rate": 1.4825345247766046e-06, + "loss": 1.5081, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.9989287853240967, + "learning_rate": 1.4805036555645818e-06, + "loss": 1.5577, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.9123456478118896, + "learning_rate": 1.4784727863525591e-06, + "loss": 1.5339, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.845898151397705, + "learning_rate": 1.4764419171405362e-06, + "loss": 1.6084, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 3.089202404022217, + "learning_rate": 1.4744110479285135e-06, + "loss": 1.6133, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.9908018112182617, + "learning_rate": 1.4723801787164907e-06, + "loss": 1.5984, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.838611125946045, + "learning_rate": 1.470349309504468e-06, + "loss": 1.5971, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.902926445007324, + "learning_rate": 1.4683184402924453e-06, + "loss": 1.6029, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 1.8312546014785767, + "eval_runtime": 46.3457, + "eval_samples_per_second": 21.577, + "eval_steps_per_second": 0.906, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.8279829025268555, + "learning_rate": 1.4662875710804226e-06, + "loss": 1.65, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.7380146980285645, + "learning_rate": 1.4642567018683998e-06, + "loss": 1.5673, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.723890781402588, + "learning_rate": 1.4622258326563771e-06, + "loss": 1.5573, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.8826022148132324, + "learning_rate": 1.4601949634443544e-06, + "loss": 1.5648, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.948598861694336, + "learning_rate": 1.4581640942323317e-06, + "loss": 1.6603, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.8282432556152344, + "learning_rate": 1.456133225020309e-06, + "loss": 1.5875, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.874748468399048, + "learning_rate": 1.4541023558082862e-06, + "loss": 1.5713, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.894249677658081, + "learning_rate": 1.4520714865962635e-06, + "loss": 1.6063, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.907095193862915, + "learning_rate": 1.4500406173842406e-06, + "loss": 1.5744, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.892090082168579, + "learning_rate": 1.4480097481722178e-06, + "loss": 1.5877, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 1.8305796384811401, + "eval_runtime": 48.3583, + "eval_samples_per_second": 20.679, + "eval_steps_per_second": 0.869, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.7991275787353516, + "learning_rate": 1.4459788789601951e-06, + "loss": 1.5475, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.9188716411590576, + "learning_rate": 1.4439480097481724e-06, + "loss": 1.5385, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.8575944900512695, + "learning_rate": 1.4419171405361497e-06, + "loss": 1.5855, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.814990520477295, + "learning_rate": 1.439886271324127e-06, + "loss": 1.5923, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.7953567504882812, + "learning_rate": 1.4378554021121042e-06, + "loss": 1.5724, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.882002115249634, + "learning_rate": 1.4358245329000815e-06, + "loss": 1.6086, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.910799503326416, + "learning_rate": 1.4337936636880588e-06, + "loss": 1.5933, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.911015748977661, + "learning_rate": 1.431762794476036e-06, + "loss": 1.5356, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.9788129329681396, + "learning_rate": 1.4297319252640133e-06, + "loss": 1.5546, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.8637843132019043, + "learning_rate": 1.4277010560519906e-06, + "loss": 1.5423, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 1.8307991027832031, + "eval_runtime": 45.4486, + "eval_samples_per_second": 22.003, + "eval_steps_per_second": 0.924, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 3.007390260696411, + "learning_rate": 1.4256701868399674e-06, + "loss": 1.6244, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.908496856689453, + "learning_rate": 1.4236393176279447e-06, + "loss": 1.5767, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.9650135040283203, + "learning_rate": 1.421608448415922e-06, + "loss": 1.6639, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.8085806369781494, + "learning_rate": 1.4195775792038993e-06, + "loss": 1.5486, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.7759034633636475, + "learning_rate": 1.4175467099918765e-06, + "loss": 1.5847, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.9309468269348145, + "learning_rate": 1.4155158407798538e-06, + "loss": 1.6014, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.8542673587799072, + "learning_rate": 1.413484971567831e-06, + "loss": 1.5183, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 3.0022106170654297, + "learning_rate": 1.4114541023558084e-06, + "loss": 1.5511, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.826852321624756, + "learning_rate": 1.4094232331437856e-06, + "loss": 1.6079, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.8682327270507812, + "learning_rate": 1.407392363931763e-06, + "loss": 1.6579, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 1.8299224376678467, + "eval_runtime": 44.6401, + "eval_samples_per_second": 22.401, + "eval_steps_per_second": 0.941, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.8816471099853516, + "learning_rate": 1.4053614947197402e-06, + "loss": 1.5825, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.696047306060791, + "learning_rate": 1.4033306255077175e-06, + "loss": 1.5068, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.9243481159210205, + "learning_rate": 1.4012997562956945e-06, + "loss": 1.6172, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.810744285583496, + "learning_rate": 1.3992688870836718e-06, + "loss": 1.5976, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.7787344455718994, + "learning_rate": 1.397238017871649e-06, + "loss": 1.5829, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.817336082458496, + "learning_rate": 1.3952071486596264e-06, + "loss": 1.5482, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.7691237926483154, + "learning_rate": 1.3931762794476036e-06, + "loss": 1.6098, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.7974281311035156, + "learning_rate": 1.391145410235581e-06, + "loss": 1.5591, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.8233275413513184, + "learning_rate": 1.3891145410235582e-06, + "loss": 1.4975, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.7465834617614746, + "learning_rate": 1.3870836718115355e-06, + "loss": 1.5682, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 1.8317815065383911, + "eval_runtime": 44.8326, + "eval_samples_per_second": 22.305, + "eval_steps_per_second": 0.937, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.9384610652923584, + "learning_rate": 1.3850528025995127e-06, + "loss": 1.5627, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.6946020126342773, + "learning_rate": 1.38302193338749e-06, + "loss": 1.4851, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.875818967819214, + "learning_rate": 1.3809910641754673e-06, + "loss": 1.595, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.871393918991089, + "learning_rate": 1.3789601949634446e-06, + "loss": 1.5297, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.8451645374298096, + "learning_rate": 1.3769293257514218e-06, + "loss": 1.5912, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.808422803878784, + "learning_rate": 1.374898456539399e-06, + "loss": 1.5316, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.866929292678833, + "learning_rate": 1.3728675873273762e-06, + "loss": 1.6223, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.969078302383423, + "learning_rate": 1.3708367181153535e-06, + "loss": 1.5855, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.739272117614746, + "learning_rate": 1.3688058489033307e-06, + "loss": 1.5915, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.9254791736602783, + "learning_rate": 1.366774979691308e-06, + "loss": 1.5989, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 1.8284611701965332, + "eval_runtime": 45.0516, + "eval_samples_per_second": 22.197, + "eval_steps_per_second": 0.932, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.7639904022216797, + "learning_rate": 1.3647441104792853e-06, + "loss": 1.5735, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.782290458679199, + "learning_rate": 1.3627132412672626e-06, + "loss": 1.5365, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.8718478679656982, + "learning_rate": 1.3606823720552398e-06, + "loss": 1.5776, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.770900249481201, + "learning_rate": 1.3586515028432171e-06, + "loss": 1.5242, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.7527379989624023, + "learning_rate": 1.3566206336311944e-06, + "loss": 1.5783, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.724205732345581, + "learning_rate": 1.3545897644191717e-06, + "loss": 1.5556, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.7595510482788086, + "learning_rate": 1.352558895207149e-06, + "loss": 1.6084, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.832383394241333, + "learning_rate": 1.3505280259951258e-06, + "loss": 1.5755, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.8154635429382324, + "learning_rate": 1.348497156783103e-06, + "loss": 1.6091, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.808480739593506, + "learning_rate": 1.3464662875710804e-06, + "loss": 1.5403, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 1.8285717964172363, + "eval_runtime": 44.0301, + "eval_samples_per_second": 22.712, + "eval_steps_per_second": 0.954, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.7877376079559326, + "learning_rate": 1.3444354183590576e-06, + "loss": 1.5985, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.854957103729248, + "learning_rate": 1.342404549147035e-06, + "loss": 1.5576, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.8522772789001465, + "learning_rate": 1.3403736799350122e-06, + "loss": 1.5914, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.866199493408203, + "learning_rate": 1.3383428107229895e-06, + "loss": 1.6094, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.9277379512786865, + "learning_rate": 1.3363119415109667e-06, + "loss": 1.6065, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.8854689598083496, + "learning_rate": 1.334281072298944e-06, + "loss": 1.5892, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.7135169506073, + "learning_rate": 1.3322502030869213e-06, + "loss": 1.5733, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.8702995777130127, + "learning_rate": 1.3302193338748986e-06, + "loss": 1.6054, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.8330469131469727, + "learning_rate": 1.3281884646628758e-06, + "loss": 1.5902, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 3.016582727432251, + "learning_rate": 1.3261575954508531e-06, + "loss": 1.5931, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 1.82985520362854, + "eval_runtime": 45.8601, + "eval_samples_per_second": 21.805, + "eval_steps_per_second": 0.916, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.8396756649017334, + "learning_rate": 1.3241267262388302e-06, + "loss": 1.5882, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.7104387283325195, + "learning_rate": 1.3220958570268074e-06, + "loss": 1.5223, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.787862777709961, + "learning_rate": 1.3200649878147847e-06, + "loss": 1.5549, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.9957401752471924, + "learning_rate": 1.318034118602762e-06, + "loss": 1.6036, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.916990280151367, + "learning_rate": 1.3160032493907393e-06, + "loss": 1.5947, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.972477912902832, + "learning_rate": 1.3139723801787166e-06, + "loss": 1.624, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.84188175201416, + "learning_rate": 1.3119415109666938e-06, + "loss": 1.5708, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.908295154571533, + "learning_rate": 1.309910641754671e-06, + "loss": 1.5412, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.8915605545043945, + "learning_rate": 1.3078797725426484e-06, + "loss": 1.5463, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.7805707454681396, + "learning_rate": 1.3058489033306257e-06, + "loss": 1.5555, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 1.8281073570251465, + "eval_runtime": 44.7855, + "eval_samples_per_second": 22.329, + "eval_steps_per_second": 0.938, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.933260679244995, + "learning_rate": 1.303818034118603e-06, + "loss": 1.578, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.970050096511841, + "learning_rate": 1.3017871649065802e-06, + "loss": 1.5606, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.717027187347412, + "learning_rate": 1.2997562956945573e-06, + "loss": 1.5446, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.966905355453491, + "learning_rate": 1.2977254264825345e-06, + "loss": 1.549, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.847358465194702, + "learning_rate": 1.2956945572705118e-06, + "loss": 1.581, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.814897060394287, + "learning_rate": 1.293663688058489e-06, + "loss": 1.5416, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.788320302963257, + "learning_rate": 1.2916328188464664e-06, + "loss": 1.5018, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.944441556930542, + "learning_rate": 1.2896019496344436e-06, + "loss": 1.5318, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.8705525398254395, + "learning_rate": 1.287571080422421e-06, + "loss": 1.5953, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.7703897953033447, + "learning_rate": 1.2855402112103982e-06, + "loss": 1.5287, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 1.8304927349090576, + "eval_runtime": 47.1736, + "eval_samples_per_second": 21.198, + "eval_steps_per_second": 0.89, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.862663745880127, + "learning_rate": 1.2835093419983755e-06, + "loss": 1.541, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.8288817405700684, + "learning_rate": 1.2814784727863528e-06, + "loss": 1.5466, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 3.0720043182373047, + "learning_rate": 1.27944760357433e-06, + "loss": 1.6106, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.9464635848999023, + "learning_rate": 1.2774167343623073e-06, + "loss": 1.6051, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.9039576053619385, + "learning_rate": 1.2753858651502846e-06, + "loss": 1.5525, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.8839735984802246, + "learning_rate": 1.2733549959382616e-06, + "loss": 1.5551, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.915053129196167, + "learning_rate": 1.271324126726239e-06, + "loss": 1.5853, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.862844467163086, + "learning_rate": 1.2692932575142162e-06, + "loss": 1.5233, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.717557430267334, + "learning_rate": 1.2672623883021935e-06, + "loss": 1.5138, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 3.1381704807281494, + "learning_rate": 1.2652315190901707e-06, + "loss": 1.6073, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 1.82700514793396, + "eval_runtime": 51.3349, + "eval_samples_per_second": 19.48, + "eval_steps_per_second": 0.818, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.8603286743164062, + "learning_rate": 1.263200649878148e-06, + "loss": 1.5713, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.9590084552764893, + "learning_rate": 1.2611697806661253e-06, + "loss": 1.5952, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.8419439792633057, + "learning_rate": 1.2591389114541026e-06, + "loss": 1.5729, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.7028636932373047, + "learning_rate": 1.2571080422420798e-06, + "loss": 1.5427, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.819112539291382, + "learning_rate": 1.2550771730300571e-06, + "loss": 1.6221, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.821469783782959, + "learning_rate": 1.2530463038180344e-06, + "loss": 1.5756, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.979747772216797, + "learning_rate": 1.2510154346060117e-06, + "loss": 1.5662, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 3.176698684692383, + "learning_rate": 1.2489845653939887e-06, + "loss": 1.591, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.8962442874908447, + "learning_rate": 1.246953696181966e-06, + "loss": 1.6549, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.840876817703247, + "learning_rate": 1.2449228269699433e-06, + "loss": 1.5636, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 1.8279746770858765, + "eval_runtime": 45.3966, + "eval_samples_per_second": 22.028, + "eval_steps_per_second": 0.925, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.8487303256988525, + "learning_rate": 1.2428919577579204e-06, + "loss": 1.5345, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.8009743690490723, + "learning_rate": 1.2408610885458976e-06, + "loss": 1.5673, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.7863199710845947, + "learning_rate": 1.238830219333875e-06, + "loss": 1.5351, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 3.4834156036376953, + "learning_rate": 1.2367993501218522e-06, + "loss": 1.6171, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.9101381301879883, + "learning_rate": 1.2347684809098295e-06, + "loss": 1.6026, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.750434160232544, + "learning_rate": 1.2327376116978067e-06, + "loss": 1.5199, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.8388402462005615, + "learning_rate": 1.230706742485784e-06, + "loss": 1.5253, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 3.000380754470825, + "learning_rate": 1.2286758732737613e-06, + "loss": 1.6463, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 3.0284345149993896, + "learning_rate": 1.2266450040617386e-06, + "loss": 1.637, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.9016034603118896, + "learning_rate": 1.2246141348497158e-06, + "loss": 1.5794, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 1.8288570642471313, + "eval_runtime": 47.0365, + "eval_samples_per_second": 21.26, + "eval_steps_per_second": 0.893, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 3.27822208404541, + "learning_rate": 1.2225832656376931e-06, + "loss": 1.5734, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.8195855617523193, + "learning_rate": 1.2205523964256704e-06, + "loss": 1.5512, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.7829315662384033, + "learning_rate": 1.2185215272136477e-06, + "loss": 1.5804, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.8760147094726562, + "learning_rate": 1.2164906580016247e-06, + "loss": 1.5858, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.8194808959960938, + "learning_rate": 1.214459788789602e-06, + "loss": 1.485, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.901355266571045, + "learning_rate": 1.2124289195775793e-06, + "loss": 1.5287, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 3.0580921173095703, + "learning_rate": 1.2103980503655566e-06, + "loss": 1.5596, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 3.0313403606414795, + "learning_rate": 1.2083671811535338e-06, + "loss": 1.5536, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.7663581371307373, + "learning_rate": 1.2063363119415111e-06, + "loss": 1.5479, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.7729651927948, + "learning_rate": 1.2043054427294882e-06, + "loss": 1.56, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 1.8264360427856445, + "eval_runtime": 45.8995, + "eval_samples_per_second": 21.787, + "eval_steps_per_second": 0.915, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 3.0862157344818115, + "learning_rate": 1.2022745735174654e-06, + "loss": 1.5869, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.9891650676727295, + "learning_rate": 1.2002437043054427e-06, + "loss": 1.6249, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.8348777294158936, + "learning_rate": 1.19821283509342e-06, + "loss": 1.6116, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.915382146835327, + "learning_rate": 1.1961819658813973e-06, + "loss": 1.5783, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 3.086989402770996, + "learning_rate": 1.1941510966693746e-06, + "loss": 1.5473, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.8165829181671143, + "learning_rate": 1.1921202274573518e-06, + "loss": 1.5617, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 3.121267795562744, + "learning_rate": 1.190089358245329e-06, + "loss": 1.6125, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.889802932739258, + "learning_rate": 1.1880584890333064e-06, + "loss": 1.5637, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.8893845081329346, + "learning_rate": 1.1860276198212837e-06, + "loss": 1.5929, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.900737762451172, + "learning_rate": 1.183996750609261e-06, + "loss": 1.5798, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 1.826514720916748, + "eval_runtime": 46.0672, + "eval_samples_per_second": 21.707, + "eval_steps_per_second": 0.912, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.8097946643829346, + "learning_rate": 1.1819658813972382e-06, + "loss": 1.5653, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.82616925239563, + "learning_rate": 1.1799350121852153e-06, + "loss": 1.612, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 3.3667945861816406, + "learning_rate": 1.1779041429731925e-06, + "loss": 1.6107, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.9167160987854004, + "learning_rate": 1.1758732737611698e-06, + "loss": 1.5424, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.9599924087524414, + "learning_rate": 1.173842404549147e-06, + "loss": 1.6195, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.8982062339782715, + "learning_rate": 1.1718115353371244e-06, + "loss": 1.5579, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.915224075317383, + "learning_rate": 1.1697806661251016e-06, + "loss": 1.6204, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.9213180541992188, + "learning_rate": 1.167749796913079e-06, + "loss": 1.5988, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.9090816974639893, + "learning_rate": 1.1657189277010562e-06, + "loss": 1.5773, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.8555307388305664, + "learning_rate": 1.1636880584890335e-06, + "loss": 1.5687, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 1.8246150016784668, + "eval_runtime": 45.9157, + "eval_samples_per_second": 21.779, + "eval_steps_per_second": 0.915, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.8956985473632812, + "learning_rate": 1.1616571892770108e-06, + "loss": 1.5878, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 4.177687644958496, + "learning_rate": 1.159626320064988e-06, + "loss": 1.579, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.9070162773132324, + "learning_rate": 1.1575954508529653e-06, + "loss": 1.5926, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.7478721141815186, + "learning_rate": 1.1555645816409426e-06, + "loss": 1.5898, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.6728436946868896, + "learning_rate": 1.1535337124289196e-06, + "loss": 1.5345, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.809030532836914, + "learning_rate": 1.151502843216897e-06, + "loss": 1.5174, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 3.078012466430664, + "learning_rate": 1.1494719740048742e-06, + "loss": 1.5914, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 4.21104621887207, + "learning_rate": 1.1474411047928515e-06, + "loss": 1.5694, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 3.0975046157836914, + "learning_rate": 1.1454102355808287e-06, + "loss": 1.545, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.925477981567383, + "learning_rate": 1.143379366368806e-06, + "loss": 1.5396, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 1.8294752836227417, + "eval_runtime": 47.0872, + "eval_samples_per_second": 21.237, + "eval_steps_per_second": 0.892, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 3.391608238220215, + "learning_rate": 1.141348497156783e-06, + "loss": 1.6501, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 3.053389549255371, + "learning_rate": 1.1393176279447604e-06, + "loss": 1.6376, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.8790252208709717, + "learning_rate": 1.1372867587327376e-06, + "loss": 1.5002, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 3.0458412170410156, + "learning_rate": 1.135255889520715e-06, + "loss": 1.5955, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.8105313777923584, + "learning_rate": 1.1332250203086922e-06, + "loss": 1.5849, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.893143892288208, + "learning_rate": 1.1311941510966695e-06, + "loss": 1.5554, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.86181640625, + "learning_rate": 1.1291632818846467e-06, + "loss": 1.5409, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.749263286590576, + "learning_rate": 1.127132412672624e-06, + "loss": 1.544, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.759432554244995, + "learning_rate": 1.1251015434606013e-06, + "loss": 1.5787, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.8302371501922607, + "learning_rate": 1.1230706742485786e-06, + "loss": 1.6204, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 1.8242546319961548, + "eval_runtime": 45.7883, + "eval_samples_per_second": 21.84, + "eval_steps_per_second": 0.917, + "step": 2200 + }, + { + "epoch": 2.412054794520548, + "grad_norm": 2.8081696033477783, + "learning_rate": 1.1210398050365558e-06, + "loss": 1.5497, + "step": 2201 + }, + { + "epoch": 2.413150684931507, + "grad_norm": 2.777566432952881, + "learning_rate": 1.1190089358245331e-06, + "loss": 1.5249, + "step": 2202 + }, + { + "epoch": 2.4142465753424656, + "grad_norm": 2.851879835128784, + "learning_rate": 1.1169780666125102e-06, + "loss": 1.5885, + "step": 2203 + }, + { + "epoch": 2.4153424657534246, + "grad_norm": 2.859072685241699, + "learning_rate": 1.1149471974004875e-06, + "loss": 1.5438, + "step": 2204 + }, + { + "epoch": 2.4164383561643836, + "grad_norm": 2.908538818359375, + "learning_rate": 1.1129163281884647e-06, + "loss": 1.6276, + "step": 2205 + }, + { + "epoch": 2.417534246575342, + "grad_norm": 2.7648227214813232, + "learning_rate": 1.110885458976442e-06, + "loss": 1.5367, + "step": 2206 + }, + { + "epoch": 2.4186301369863012, + "grad_norm": 2.9282727241516113, + "learning_rate": 1.1088545897644193e-06, + "loss": 1.6055, + "step": 2207 + }, + { + "epoch": 2.4197260273972603, + "grad_norm": 2.7551443576812744, + "learning_rate": 1.1068237205523966e-06, + "loss": 1.5984, + "step": 2208 + }, + { + "epoch": 2.4208219178082193, + "grad_norm": 2.88720440864563, + "learning_rate": 1.1047928513403736e-06, + "loss": 1.5988, + "step": 2209 + }, + { + "epoch": 2.421917808219178, + "grad_norm": 2.75961971282959, + "learning_rate": 1.102761982128351e-06, + "loss": 1.5291, + "step": 2210 + }, + { + "epoch": 2.421917808219178, + "eval_loss": 1.8230210542678833, + "eval_runtime": 45.9261, + "eval_samples_per_second": 21.774, + "eval_steps_per_second": 0.915, + "step": 2210 + }, + { + "epoch": 2.423013698630137, + "grad_norm": 2.7809903621673584, + "learning_rate": 1.1007311129163282e-06, + "loss": 1.5563, + "step": 2211 + }, + { + "epoch": 2.424109589041096, + "grad_norm": 2.7741317749023438, + "learning_rate": 1.0987002437043055e-06, + "loss": 1.5594, + "step": 2212 + }, + { + "epoch": 2.425205479452055, + "grad_norm": 2.840890884399414, + "learning_rate": 1.0966693744922827e-06, + "loss": 1.52, + "step": 2213 + }, + { + "epoch": 2.4263013698630136, + "grad_norm": 2.9647538661956787, + "learning_rate": 1.09463850528026e-06, + "loss": 1.5447, + "step": 2214 + }, + { + "epoch": 2.4273972602739726, + "grad_norm": 2.9884531497955322, + "learning_rate": 1.0926076360682373e-06, + "loss": 1.5426, + "step": 2215 + }, + { + "epoch": 2.4284931506849317, + "grad_norm": 2.8884270191192627, + "learning_rate": 1.0905767668562146e-06, + "loss": 1.5546, + "step": 2216 + }, + { + "epoch": 2.4295890410958902, + "grad_norm": 2.97591495513916, + "learning_rate": 1.0885458976441918e-06, + "loss": 1.632, + "step": 2217 + }, + { + "epoch": 2.4306849315068493, + "grad_norm": 2.944126605987549, + "learning_rate": 1.0865150284321691e-06, + "loss": 1.5845, + "step": 2218 + }, + { + "epoch": 2.4317808219178083, + "grad_norm": 2.9660258293151855, + "learning_rate": 1.0844841592201464e-06, + "loss": 1.5939, + "step": 2219 + }, + { + "epoch": 2.432876712328767, + "grad_norm": 2.8848304748535156, + "learning_rate": 1.0824532900081237e-06, + "loss": 1.5848, + "step": 2220 + }, + { + "epoch": 2.432876712328767, + "eval_loss": 1.8266901969909668, + "eval_runtime": 51.9703, + "eval_samples_per_second": 19.242, + "eval_steps_per_second": 0.808, + "step": 2220 + }, + { + "epoch": 2.433972602739726, + "grad_norm": 2.9656991958618164, + "learning_rate": 1.080422420796101e-06, + "loss": 1.568, + "step": 2221 + }, + { + "epoch": 2.435068493150685, + "grad_norm": 2.9461898803710938, + "learning_rate": 1.078391551584078e-06, + "loss": 1.5922, + "step": 2222 + }, + { + "epoch": 2.436164383561644, + "grad_norm": 2.937289237976074, + "learning_rate": 1.0763606823720553e-06, + "loss": 1.6414, + "step": 2223 + }, + { + "epoch": 2.4372602739726026, + "grad_norm": 2.8320538997650146, + "learning_rate": 1.0743298131600326e-06, + "loss": 1.5995, + "step": 2224 + }, + { + "epoch": 2.4383561643835616, + "grad_norm": 2.8221585750579834, + "learning_rate": 1.0722989439480098e-06, + "loss": 1.549, + "step": 2225 + }, + { + "epoch": 2.4394520547945207, + "grad_norm": 2.854454517364502, + "learning_rate": 1.070268074735987e-06, + "loss": 1.5474, + "step": 2226 + }, + { + "epoch": 2.4405479452054797, + "grad_norm": 2.7813878059387207, + "learning_rate": 1.0682372055239644e-06, + "loss": 1.5638, + "step": 2227 + }, + { + "epoch": 2.4416438356164383, + "grad_norm": 2.985661029815674, + "learning_rate": 1.0662063363119414e-06, + "loss": 1.5457, + "step": 2228 + }, + { + "epoch": 2.4427397260273973, + "grad_norm": 2.863631010055542, + "learning_rate": 1.0641754670999187e-06, + "loss": 1.5824, + "step": 2229 + }, + { + "epoch": 2.4438356164383563, + "grad_norm": 2.8328969478607178, + "learning_rate": 1.062144597887896e-06, + "loss": 1.5708, + "step": 2230 + }, + { + "epoch": 2.4438356164383563, + "eval_loss": 1.8261159658432007, + "eval_runtime": 50.0095, + "eval_samples_per_second": 19.996, + "eval_steps_per_second": 0.84, + "step": 2230 + }, + { + "epoch": 2.444931506849315, + "grad_norm": 2.9791696071624756, + "learning_rate": 1.0601137286758733e-06, + "loss": 1.6242, + "step": 2231 + }, + { + "epoch": 2.446027397260274, + "grad_norm": 2.7748308181762695, + "learning_rate": 1.0580828594638505e-06, + "loss": 1.499, + "step": 2232 + }, + { + "epoch": 2.447123287671233, + "grad_norm": 2.8511860370635986, + "learning_rate": 1.0560519902518278e-06, + "loss": 1.5787, + "step": 2233 + }, + { + "epoch": 2.4482191780821916, + "grad_norm": 2.802671194076538, + "learning_rate": 1.054021121039805e-06, + "loss": 1.4995, + "step": 2234 + }, + { + "epoch": 2.4493150684931506, + "grad_norm": 2.8005664348602295, + "learning_rate": 1.0519902518277824e-06, + "loss": 1.5083, + "step": 2235 + }, + { + "epoch": 2.4504109589041096, + "grad_norm": 2.92136287689209, + "learning_rate": 1.0499593826157596e-06, + "loss": 1.5807, + "step": 2236 + }, + { + "epoch": 2.4515068493150687, + "grad_norm": 2.850313186645508, + "learning_rate": 1.047928513403737e-06, + "loss": 1.5592, + "step": 2237 + }, + { + "epoch": 2.4526027397260273, + "grad_norm": 2.745370864868164, + "learning_rate": 1.0458976441917142e-06, + "loss": 1.495, + "step": 2238 + }, + { + "epoch": 2.4536986301369863, + "grad_norm": 2.794156789779663, + "learning_rate": 1.0438667749796915e-06, + "loss": 1.5024, + "step": 2239 + }, + { + "epoch": 2.4547945205479453, + "grad_norm": 2.8183324337005615, + "learning_rate": 1.0418359057676688e-06, + "loss": 1.5461, + "step": 2240 + }, + { + "epoch": 2.4547945205479453, + "eval_loss": 1.8265395164489746, + "eval_runtime": 47.1962, + "eval_samples_per_second": 21.188, + "eval_steps_per_second": 0.89, + "step": 2240 + }, + { + "epoch": 2.455890410958904, + "grad_norm": 2.863630771636963, + "learning_rate": 1.0398050365556458e-06, + "loss": 1.5977, + "step": 2241 + }, + { + "epoch": 2.456986301369863, + "grad_norm": 2.7688169479370117, + "learning_rate": 1.037774167343623e-06, + "loss": 1.5399, + "step": 2242 + }, + { + "epoch": 2.458082191780822, + "grad_norm": 2.8812880516052246, + "learning_rate": 1.0357432981316004e-06, + "loss": 1.5722, + "step": 2243 + }, + { + "epoch": 2.459178082191781, + "grad_norm": 3.2114691734313965, + "learning_rate": 1.0337124289195776e-06, + "loss": 1.6012, + "step": 2244 + }, + { + "epoch": 2.4602739726027396, + "grad_norm": 2.961271047592163, + "learning_rate": 1.031681559707555e-06, + "loss": 1.606, + "step": 2245 + }, + { + "epoch": 2.4613698630136986, + "grad_norm": 2.835495710372925, + "learning_rate": 1.0296506904955322e-06, + "loss": 1.5518, + "step": 2246 + }, + { + "epoch": 2.4624657534246577, + "grad_norm": 2.8796403408050537, + "learning_rate": 1.0276198212835095e-06, + "loss": 1.573, + "step": 2247 + }, + { + "epoch": 2.4635616438356163, + "grad_norm": 2.774409770965576, + "learning_rate": 1.0255889520714867e-06, + "loss": 1.5485, + "step": 2248 + }, + { + "epoch": 2.4646575342465753, + "grad_norm": 2.8847763538360596, + "learning_rate": 1.023558082859464e-06, + "loss": 1.5168, + "step": 2249 + }, + { + "epoch": 2.4657534246575343, + "grad_norm": 2.9533803462982178, + "learning_rate": 1.0215272136474413e-06, + "loss": 1.6179, + "step": 2250 + }, + { + "epoch": 2.4657534246575343, + "eval_loss": 1.8261420726776123, + "eval_runtime": 46.1356, + "eval_samples_per_second": 21.675, + "eval_steps_per_second": 0.91, + "step": 2250 + }, + { + "epoch": 2.466849315068493, + "grad_norm": 2.8762729167938232, + "learning_rate": 1.0194963444354186e-06, + "loss": 1.5109, + "step": 2251 + }, + { + "epoch": 2.467945205479452, + "grad_norm": 2.9228789806365967, + "learning_rate": 1.0174654752233958e-06, + "loss": 1.5861, + "step": 2252 + }, + { + "epoch": 2.469041095890411, + "grad_norm": 2.9467368125915527, + "learning_rate": 1.015434606011373e-06, + "loss": 1.5533, + "step": 2253 + }, + { + "epoch": 2.47013698630137, + "grad_norm": 2.9693222045898438, + "learning_rate": 1.0134037367993502e-06, + "loss": 1.5741, + "step": 2254 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 2.933368682861328, + "learning_rate": 1.0113728675873275e-06, + "loss": 1.5555, + "step": 2255 + }, + { + "epoch": 2.4723287671232876, + "grad_norm": 2.9675047397613525, + "learning_rate": 1.0093419983753047e-06, + "loss": 1.5803, + "step": 2256 + }, + { + "epoch": 2.4734246575342467, + "grad_norm": 2.931260108947754, + "learning_rate": 1.007311129163282e-06, + "loss": 1.5263, + "step": 2257 + }, + { + "epoch": 2.4745205479452057, + "grad_norm": 2.8160653114318848, + "learning_rate": 1.0052802599512593e-06, + "loss": 1.5281, + "step": 2258 + }, + { + "epoch": 2.4756164383561643, + "grad_norm": 3.0168352127075195, + "learning_rate": 1.0032493907392364e-06, + "loss": 1.5746, + "step": 2259 + }, + { + "epoch": 2.4767123287671233, + "grad_norm": 2.7623541355133057, + "learning_rate": 1.0012185215272136e-06, + "loss": 1.5181, + "step": 2260 + }, + { + "epoch": 2.4767123287671233, + "eval_loss": 1.8249253034591675, + "eval_runtime": 47.2027, + "eval_samples_per_second": 21.185, + "eval_steps_per_second": 0.89, + "step": 2260 + }, + { + "epoch": 2.4778082191780824, + "grad_norm": 2.8098902702331543, + "learning_rate": 9.99187652315191e-07, + "loss": 1.6099, + "step": 2261 + }, + { + "epoch": 2.478904109589041, + "grad_norm": 2.8562352657318115, + "learning_rate": 9.971567831031682e-07, + "loss": 1.5795, + "step": 2262 + }, + { + "epoch": 2.48, + "grad_norm": 2.7867422103881836, + "learning_rate": 9.951259138911455e-07, + "loss": 1.5104, + "step": 2263 + }, + { + "epoch": 2.481095890410959, + "grad_norm": 2.9658195972442627, + "learning_rate": 9.930950446791227e-07, + "loss": 1.5891, + "step": 2264 + }, + { + "epoch": 2.4821917808219176, + "grad_norm": 2.824355363845825, + "learning_rate": 9.910641754671e-07, + "loss": 1.6385, + "step": 2265 + }, + { + "epoch": 2.4832876712328766, + "grad_norm": 2.833772897720337, + "learning_rate": 9.890333062550773e-07, + "loss": 1.5759, + "step": 2266 + }, + { + "epoch": 2.4843835616438357, + "grad_norm": 2.9420840740203857, + "learning_rate": 9.870024370430546e-07, + "loss": 1.573, + "step": 2267 + }, + { + "epoch": 2.4854794520547947, + "grad_norm": 2.84629225730896, + "learning_rate": 9.849715678310318e-07, + "loss": 1.5606, + "step": 2268 + }, + { + "epoch": 2.4865753424657533, + "grad_norm": 2.7983481884002686, + "learning_rate": 9.829406986190091e-07, + "loss": 1.5718, + "step": 2269 + }, + { + "epoch": 2.4876712328767123, + "grad_norm": 2.8418283462524414, + "learning_rate": 9.809098294069864e-07, + "loss": 1.5052, + "step": 2270 + }, + { + "epoch": 2.4876712328767123, + "eval_loss": 1.8237050771713257, + "eval_runtime": 53.2157, + "eval_samples_per_second": 18.791, + "eval_steps_per_second": 0.789, + "step": 2270 + }, + { + "epoch": 2.4887671232876714, + "grad_norm": 2.8386685848236084, + "learning_rate": 9.788789601949637e-07, + "loss": 1.5638, + "step": 2271 + }, + { + "epoch": 2.48986301369863, + "grad_norm": 2.8796494007110596, + "learning_rate": 9.768480909829407e-07, + "loss": 1.6258, + "step": 2272 + }, + { + "epoch": 2.490958904109589, + "grad_norm": 2.8450164794921875, + "learning_rate": 9.74817221770918e-07, + "loss": 1.5567, + "step": 2273 + }, + { + "epoch": 2.492054794520548, + "grad_norm": 2.8501436710357666, + "learning_rate": 9.727863525588953e-07, + "loss": 1.5731, + "step": 2274 + }, + { + "epoch": 2.493150684931507, + "grad_norm": 2.8499205112457275, + "learning_rate": 9.707554833468726e-07, + "loss": 1.5383, + "step": 2275 + }, + { + "epoch": 2.4942465753424656, + "grad_norm": 2.904446601867676, + "learning_rate": 9.687246141348498e-07, + "loss": 1.5423, + "step": 2276 + }, + { + "epoch": 2.4953424657534247, + "grad_norm": 2.7885360717773438, + "learning_rate": 9.666937449228271e-07, + "loss": 1.5583, + "step": 2277 + }, + { + "epoch": 2.4964383561643837, + "grad_norm": 2.8230857849121094, + "learning_rate": 9.646628757108042e-07, + "loss": 1.531, + "step": 2278 + }, + { + "epoch": 2.4975342465753423, + "grad_norm": 2.9093189239501953, + "learning_rate": 9.626320064987814e-07, + "loss": 1.5901, + "step": 2279 + }, + { + "epoch": 2.4986301369863013, + "grad_norm": 2.853430986404419, + "learning_rate": 9.606011372867587e-07, + "loss": 1.5425, + "step": 2280 + }, + { + "epoch": 2.4986301369863013, + "eval_loss": 1.824331283569336, + "eval_runtime": 51.6196, + "eval_samples_per_second": 19.372, + "eval_steps_per_second": 0.814, + "step": 2280 + }, + { + "epoch": 2.4997260273972604, + "grad_norm": 2.857558488845825, + "learning_rate": 9.58570268074736e-07, + "loss": 1.555, + "step": 2281 + }, + { + "epoch": 2.500821917808219, + "grad_norm": 2.758384943008423, + "learning_rate": 9.565393988627133e-07, + "loss": 1.5546, + "step": 2282 + }, + { + "epoch": 2.501917808219178, + "grad_norm": 2.8295326232910156, + "learning_rate": 9.545085296506906e-07, + "loss": 1.5842, + "step": 2283 + }, + { + "epoch": 2.503013698630137, + "grad_norm": 2.855534315109253, + "learning_rate": 9.524776604386677e-07, + "loss": 1.5636, + "step": 2284 + }, + { + "epoch": 2.504109589041096, + "grad_norm": 2.7737905979156494, + "learning_rate": 9.50446791226645e-07, + "loss": 1.5101, + "step": 2285 + }, + { + "epoch": 2.5052054794520546, + "grad_norm": 2.7556989192962646, + "learning_rate": 9.484159220146223e-07, + "loss": 1.6216, + "step": 2286 + }, + { + "epoch": 2.5063013698630137, + "grad_norm": 2.8598365783691406, + "learning_rate": 9.463850528025995e-07, + "loss": 1.5818, + "step": 2287 + }, + { + "epoch": 2.5073972602739727, + "grad_norm": 3.0465996265411377, + "learning_rate": 9.443541835905768e-07, + "loss": 1.6237, + "step": 2288 + }, + { + "epoch": 2.5084931506849317, + "grad_norm": 3.067974328994751, + "learning_rate": 9.423233143785541e-07, + "loss": 1.6346, + "step": 2289 + }, + { + "epoch": 2.5095890410958903, + "grad_norm": 2.9208505153656006, + "learning_rate": 9.402924451665313e-07, + "loss": 1.543, + "step": 2290 + }, + { + "epoch": 2.5095890410958903, + "eval_loss": 1.8239408731460571, + "eval_runtime": 52.1634, + "eval_samples_per_second": 19.171, + "eval_steps_per_second": 0.805, + "step": 2290 + }, + { + "epoch": 2.5106849315068493, + "grad_norm": 3.0587332248687744, + "learning_rate": 9.382615759545085e-07, + "loss": 1.5814, + "step": 2291 + }, + { + "epoch": 2.5117808219178084, + "grad_norm": 2.9887471199035645, + "learning_rate": 9.362307067424858e-07, + "loss": 1.5481, + "step": 2292 + }, + { + "epoch": 2.512876712328767, + "grad_norm": 3.127073049545288, + "learning_rate": 9.341998375304631e-07, + "loss": 1.6121, + "step": 2293 + }, + { + "epoch": 2.513972602739726, + "grad_norm": 2.7613704204559326, + "learning_rate": 9.321689683184404e-07, + "loss": 1.5337, + "step": 2294 + }, + { + "epoch": 2.515068493150685, + "grad_norm": 2.9512479305267334, + "learning_rate": 9.301380991064176e-07, + "loss": 1.5161, + "step": 2295 + }, + { + "epoch": 2.5161643835616436, + "grad_norm": 2.9375245571136475, + "learning_rate": 9.281072298943949e-07, + "loss": 1.5538, + "step": 2296 + }, + { + "epoch": 2.5172602739726027, + "grad_norm": 3.8778791427612305, + "learning_rate": 9.260763606823721e-07, + "loss": 1.5736, + "step": 2297 + }, + { + "epoch": 2.5183561643835617, + "grad_norm": 3.176246404647827, + "learning_rate": 9.240454914703494e-07, + "loss": 1.5855, + "step": 2298 + }, + { + "epoch": 2.5194520547945203, + "grad_norm": 2.826871395111084, + "learning_rate": 9.220146222583266e-07, + "loss": 1.526, + "step": 2299 + }, + { + "epoch": 2.5205479452054793, + "grad_norm": 2.7983620166778564, + "learning_rate": 9.199837530463039e-07, + "loss": 1.5656, + "step": 2300 + }, + { + "epoch": 2.5205479452054793, + "eval_loss": 1.8233634233474731, + "eval_runtime": 51.1613, + "eval_samples_per_second": 19.546, + "eval_steps_per_second": 0.821, + "step": 2300 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 7.169811641532416e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2300/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2400/scheduler.pt b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2400/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..f88aabcb0353a3d746e3ce76304eddf9a7dea116 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2400/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5e8abcc134c3178fa2dd27afe60a0d71d01592061bcabd0bec2e0eaebaea2bf4 +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/generation_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/latest b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/latest new file mode 100644 index 0000000000000000000000000000000000000000..98f8bed9a5485ee900d9931cc06950de69499848 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/latest @@ -0,0 +1 @@ +global_step2500 \ No newline at end of file diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..cf71113371921ea2e7fb737a20484b1a62f38481 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a08b70b84bf331849cc389e31e88e5bf289f53f5f48e4ac516e4a788f9bfae7d +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/scheduler.pt b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..e70d2c2d326c257007c3a3527fba50cd79f158ce --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d3934d9047e04a6364d4fe8c5d63097b7f00ebf0aba610f23dc8c7d84571f3e8 +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/special_tokens_map.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/tokenizer.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..34da28b24b0bfa0a2a69eb0e72a18f3dd85fe10a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/trainer_state.json @@ -0,0 +1,19533 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.73972602739726, + "eval_steps": 10, + "global_step": 2500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.5818793773651123, + "learning_rate": 4.569455727051178e-06, + "loss": 1.85, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 3.0734691619873047, + "learning_rate": 4.567424857839156e-06, + "loss": 1.8682, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.496253252029419, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9007, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.360891342163086, + "learning_rate": 4.5633631194151105e-06, + "loss": 1.8262, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.5731117725372314, + "learning_rate": 4.561332250203087e-06, + "loss": 1.8369, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.700558662414551, + "learning_rate": 4.559301380991065e-06, + "loss": 1.7825, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.561306953430176, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.8174, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.618889570236206, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.8812, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.4671308994293213, + "learning_rate": 4.553208773354996e-06, + "loss": 1.7959, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.4571878910064697, + "learning_rate": 4.551177904142974e-06, + "loss": 1.8943, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.010777711868286, + "eval_runtime": 43.7424, + "eval_samples_per_second": 22.861, + "eval_steps_per_second": 0.96, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.4786667823791504, + "learning_rate": 4.5491470349309505e-06, + "loss": 1.8147, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.2955000400543213, + "learning_rate": 4.547116165718929e-06, + "loss": 1.8547, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.3873913288116455, + "learning_rate": 4.545085296506905e-06, + "loss": 1.8156, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.2449734210968018, + "learning_rate": 4.543054427294882e-06, + "loss": 1.8053, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.3969857692718506, + "learning_rate": 4.54102355808286e-06, + "loss": 1.8085, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.3446357250213623, + "learning_rate": 4.538992688870837e-06, + "loss": 1.8355, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.408762216567993, + "learning_rate": 4.536961819658814e-06, + "loss": 1.8319, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.362962245941162, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.8422, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.402984857559204, + "learning_rate": 4.532900081234769e-06, + "loss": 1.8125, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.3964452743530273, + "learning_rate": 4.530869212022746e-06, + "loss": 1.8417, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.006908416748047, + "eval_runtime": 44.2701, + "eval_samples_per_second": 22.589, + "eval_steps_per_second": 0.949, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.595233201980591, + "learning_rate": 4.528838342810723e-06, + "loss": 1.8275, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.326117753982544, + "learning_rate": 4.526807473598701e-06, + "loss": 1.8405, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.4904465675354004, + "learning_rate": 4.524776604386678e-06, + "loss": 1.8172, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.4591610431671143, + "learning_rate": 4.522745735174655e-06, + "loss": 1.876, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.319800853729248, + "learning_rate": 4.520714865962632e-06, + "loss": 1.8294, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.28790020942688, + "learning_rate": 4.51868399675061e-06, + "loss": 1.8537, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.284822702407837, + "learning_rate": 4.516653127538587e-06, + "loss": 1.8087, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.273820638656616, + "learning_rate": 4.514622258326564e-06, + "loss": 1.7992, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.2025654315948486, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.8248, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 2.3218843936920166, + "learning_rate": 4.510560519902519e-06, + "loss": 1.8647, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.0048956871032715, + "eval_runtime": 44.5972, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.515766143798828, + "learning_rate": 4.508529650690496e-06, + "loss": 1.7673, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.4376487731933594, + "learning_rate": 4.506498781478473e-06, + "loss": 1.7755, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.2801272869110107, + "learning_rate": 4.504467912266451e-06, + "loss": 1.8273, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.4707891941070557, + "learning_rate": 4.502437043054428e-06, + "loss": 1.8971, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.301301956176758, + "learning_rate": 4.500406173842405e-06, + "loss": 1.8345, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.2317373752593994, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.8248, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.447373867034912, + "learning_rate": 4.49634443541836e-06, + "loss": 1.8148, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.337245225906372, + "learning_rate": 4.494313566206336e-06, + "loss": 1.8811, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3717031478881836, + "learning_rate": 4.492282696994314e-06, + "loss": 1.8311, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.287677526473999, + "learning_rate": 4.490251827782291e-06, + "loss": 1.7728, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.004889965057373, + "eval_runtime": 43.8132, + "eval_samples_per_second": 22.824, + "eval_steps_per_second": 0.959, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.441445827484131, + "learning_rate": 4.488220958570269e-06, + "loss": 1.7927, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.3191535472869873, + "learning_rate": 4.486190089358245e-06, + "loss": 1.759, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.3118820190429688, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8509, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.4711859226226807, + "learning_rate": 4.4821283509342e-06, + "loss": 1.8402, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.3142173290252686, + "learning_rate": 4.480097481722178e-06, + "loss": 1.8238, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.2665774822235107, + "learning_rate": 4.478066612510154e-06, + "loss": 1.7826, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.1692490577697754, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.8138, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.412783145904541, + "learning_rate": 4.474004874086109e-06, + "loss": 1.9421, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3692686557769775, + "learning_rate": 4.471974004874087e-06, + "loss": 1.7722, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.3224151134490967, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.8753, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.0016305446624756, + "eval_runtime": 45.564, + "eval_samples_per_second": 21.947, + "eval_steps_per_second": 0.922, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.51043438911438, + "learning_rate": 4.467912266450041e-06, + "loss": 1.8325, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.4568657875061035, + "learning_rate": 4.465881397238018e-06, + "loss": 1.8081, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.3279001712799072, + "learning_rate": 4.463850528025995e-06, + "loss": 1.8439, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.6041171550750732, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.8623, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.3780531883239746, + "learning_rate": 4.45978878960195e-06, + "loss": 1.8269, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.511500597000122, + "learning_rate": 4.457757920389927e-06, + "loss": 1.7749, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.707789659500122, + "learning_rate": 4.455727051177904e-06, + "loss": 1.8157, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.5495495796203613, + "learning_rate": 4.453696181965882e-06, + "loss": 1.8534, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.255876302719116, + "learning_rate": 4.451665312753859e-06, + "loss": 1.8377, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.415097951889038, + "learning_rate": 4.449634443541836e-06, + "loss": 1.8764, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 1.9943474531173706, + "eval_runtime": 42.2985, + "eval_samples_per_second": 23.642, + "eval_steps_per_second": 0.993, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.2604713439941406, + "learning_rate": 4.4476035743298135e-06, + "loss": 1.7898, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.5643842220306396, + "learning_rate": 4.445572705117791e-06, + "loss": 1.7953, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.412548780441284, + "learning_rate": 4.443541835905768e-06, + "loss": 1.8611, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.5423877239227295, + "learning_rate": 4.441510966693745e-06, + "loss": 1.8049, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.4986631870269775, + "learning_rate": 4.439480097481723e-06, + "loss": 1.8759, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.8349273204803467, + "learning_rate": 4.4374492282697e-06, + "loss": 1.7416, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.452017068862915, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8268, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.6425273418426514, + "learning_rate": 4.433387489845654e-06, + "loss": 1.8334, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.328740358352661, + "learning_rate": 4.431356620633632e-06, + "loss": 1.7931, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.262209415435791, + "learning_rate": 4.429325751421609e-06, + "loss": 1.8151, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 1.9917123317718506, + "eval_runtime": 42.4309, + "eval_samples_per_second": 23.568, + "eval_steps_per_second": 0.99, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.2984049320220947, + "learning_rate": 4.427294882209586e-06, + "loss": 1.8187, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.3910019397735596, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.7512, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.4494471549987793, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8202, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.3857171535491943, + "learning_rate": 4.421202274573518e-06, + "loss": 1.8279, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.4526500701904297, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.7136, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.2216501235961914, + "learning_rate": 4.417140536149473e-06, + "loss": 1.7574, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.7639758586883545, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8403, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.3177521228790283, + "learning_rate": 4.413078797725427e-06, + "loss": 1.7898, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.527902364730835, + "learning_rate": 4.411047928513404e-06, + "loss": 1.8184, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.34061336517334, + "learning_rate": 4.409017059301382e-06, + "loss": 1.811, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 1.9882216453552246, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.6462934017181396, + "learning_rate": 4.406986190089358e-06, + "loss": 1.8501, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.2725913524627686, + "learning_rate": 4.404955320877336e-06, + "loss": 1.8262, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.390084981918335, + "learning_rate": 4.402924451665313e-06, + "loss": 1.7842, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.314305305480957, + "learning_rate": 4.400893582453291e-06, + "loss": 1.7692, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.2828257083892822, + "learning_rate": 4.398862713241267e-06, + "loss": 1.8031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.3493599891662598, + "learning_rate": 4.396831844029245e-06, + "loss": 1.8183, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.423365831375122, + "learning_rate": 4.394800974817222e-06, + "loss": 1.8012, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.438015937805176, + "learning_rate": 4.392770105605199e-06, + "loss": 1.8217, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.357172727584839, + "learning_rate": 4.390739236393176e-06, + "loss": 1.8382, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.3383474349975586, + "learning_rate": 4.388708367181154e-06, + "loss": 1.841, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 1.9878787994384766, + "eval_runtime": 44.0694, + "eval_samples_per_second": 22.691, + "eval_steps_per_second": 0.953, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.459385633468628, + "learning_rate": 4.386677497969131e-06, + "loss": 1.8377, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.356788158416748, + "learning_rate": 4.384646628757108e-06, + "loss": 1.8718, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.3217222690582275, + "learning_rate": 4.3826157595450855e-06, + "loss": 1.8073, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.3134331703186035, + "learning_rate": 4.380584890333063e-06, + "loss": 1.7675, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.3273653984069824, + "learning_rate": 4.37855402112104e-06, + "loss": 1.7169, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.611034393310547, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8702, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.2077300548553467, + "learning_rate": 4.3744922826969946e-06, + "loss": 1.7951, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.3393311500549316, + "learning_rate": 4.372461413484972e-06, + "loss": 1.8136, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.3306660652160645, + "learning_rate": 4.370430544272949e-06, + "loss": 1.8372, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.4440250396728516, + "learning_rate": 4.368399675060926e-06, + "loss": 1.844, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 1.98439621925354, + "eval_runtime": 44.4595, + "eval_samples_per_second": 22.492, + "eval_steps_per_second": 0.945, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.205542802810669, + "learning_rate": 4.366368805848904e-06, + "loss": 1.8182, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.6030094623565674, + "learning_rate": 4.364337936636881e-06, + "loss": 1.7524, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.4000444412231445, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8269, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.319526433944702, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.7722, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.1624808311462402, + "learning_rate": 4.358245329000813e-06, + "loss": 1.8081, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.26615309715271, + "learning_rate": 4.35621445978879e-06, + "loss": 1.8049, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.4895036220550537, + "learning_rate": 4.354183590576767e-06, + "loss": 1.7932, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.437161445617676, + "learning_rate": 4.352152721364745e-06, + "loss": 1.772, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.344069480895996, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8213, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.360734224319458, + "learning_rate": 4.348090982940699e-06, + "loss": 1.8268, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 1.9846479892730713, + "eval_runtime": 44.947, + "eval_samples_per_second": 22.248, + "eval_steps_per_second": 0.934, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.59487247467041, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.8303, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.6512107849121094, + "learning_rate": 4.344029244516654e-06, + "loss": 1.7979, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.2900476455688477, + "learning_rate": 4.341998375304631e-06, + "loss": 1.8707, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.409494638442993, + "learning_rate": 4.339967506092607e-06, + "loss": 1.8075, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.527205467224121, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.7884, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.33461332321167, + "learning_rate": 4.335905767668562e-06, + "loss": 1.7839, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.3555047512054443, + "learning_rate": 4.33387489845654e-06, + "loss": 1.8245, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.397148609161377, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.8115, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.232139825820923, + "learning_rate": 4.329813160032495e-06, + "loss": 1.7747, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.1684389114379883, + "learning_rate": 4.327782290820471e-06, + "loss": 1.748, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 1.980297565460205, + "eval_runtime": 42.9183, + "eval_samples_per_second": 23.3, + "eval_steps_per_second": 0.979, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.2454638481140137, + "learning_rate": 4.325751421608449e-06, + "loss": 1.83, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.3264827728271484, + "learning_rate": 4.323720552396426e-06, + "loss": 1.7856, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.298804998397827, + "learning_rate": 4.321689683184404e-06, + "loss": 1.78, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2202727794647217, + "learning_rate": 4.31965881397238e-06, + "loss": 1.7875, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.494479179382324, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.8363, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3520524501800537, + "learning_rate": 4.315597075548335e-06, + "loss": 1.8266, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.373878002166748, + "learning_rate": 4.313566206336312e-06, + "loss": 1.8348, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.4232168197631836, + "learning_rate": 4.311535337124289e-06, + "loss": 1.77, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.2572121620178223, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.7914, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.5817952156066895, + "learning_rate": 4.307473598700244e-06, + "loss": 1.8099, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 1.9784427881240845, + "eval_runtime": 43.2981, + "eval_samples_per_second": 23.096, + "eval_steps_per_second": 0.97, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.646040916442871, + "learning_rate": 4.305442729488221e-06, + "loss": 1.7452, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.370220184326172, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8399, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.4599130153656006, + "learning_rate": 4.301380991064176e-06, + "loss": 1.8273, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 2.4135093688964844, + "learning_rate": 4.299350121852153e-06, + "loss": 1.852, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.371021270751953, + "learning_rate": 4.29731925264013e-06, + "loss": 1.8339, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.6858229637145996, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.792, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.4101266860961914, + "learning_rate": 4.293257514216085e-06, + "loss": 1.7556, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.769761800765991, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8223, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.5777804851531982, + "learning_rate": 4.289195775792039e-06, + "loss": 1.8406, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.376864433288574, + "learning_rate": 4.287164906580017e-06, + "loss": 1.7846, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 1.9758412837982178, + "eval_runtime": 43.9353, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.7461960315704346, + "learning_rate": 4.285134037367994e-06, + "loss": 1.776, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.549576759338379, + "learning_rate": 4.283103168155971e-06, + "loss": 1.7644, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.568074941635132, + "learning_rate": 4.281072298943948e-06, + "loss": 1.7571, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.8195722103118896, + "learning_rate": 4.279041429731926e-06, + "loss": 1.7959, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.471623182296753, + "learning_rate": 4.277010560519903e-06, + "loss": 1.7584, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.664689302444458, + "learning_rate": 4.27497969130788e-06, + "loss": 1.7909, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.684504985809326, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.8205, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.5247528553009033, + "learning_rate": 4.270917952883835e-06, + "loss": 1.8365, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.757972478866577, + "learning_rate": 4.268887083671812e-06, + "loss": 1.7665, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.6032614707946777, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8702, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 1.9716291427612305, + "eval_runtime": 43.5482, + "eval_samples_per_second": 22.963, + "eval_steps_per_second": 0.964, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.5324370861053467, + "learning_rate": 4.264825345247766e-06, + "loss": 1.738, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.730712890625, + "learning_rate": 4.262794476035744e-06, + "loss": 1.7918, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.5554966926574707, + "learning_rate": 4.26076360682372e-06, + "loss": 1.7218, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.4774436950683594, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.7823, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.77431058883667, + "learning_rate": 4.256701868399675e-06, + "loss": 1.794, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.258828639984131, + "learning_rate": 4.254670999187653e-06, + "loss": 1.805, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.384399175643921, + "learning_rate": 4.252640129975629e-06, + "loss": 1.818, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3278932571411133, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.7719, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.207348108291626, + "learning_rate": 4.248578391551584e-06, + "loss": 1.7362, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.4525978565216064, + "learning_rate": 4.246547522339562e-06, + "loss": 1.778, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 1.9694374799728394, + "eval_runtime": 42.5977, + "eval_samples_per_second": 23.475, + "eval_steps_per_second": 0.986, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.377835273742676, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.7926, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.4623866081237793, + "learning_rate": 4.242485783915517e-06, + "loss": 1.7366, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.542219400405884, + "learning_rate": 4.240454914703493e-06, + "loss": 1.7409, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.1955819129943848, + "learning_rate": 4.23842404549147e-06, + "loss": 1.7596, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.5834269523620605, + "learning_rate": 4.236393176279448e-06, + "loss": 1.7343, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.219660520553589, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8089, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.283200740814209, + "learning_rate": 4.232331437855402e-06, + "loss": 1.8248, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.274127244949341, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.825, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.3550686836242676, + "learning_rate": 4.228269699431357e-06, + "loss": 1.7957, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.588027238845825, + "learning_rate": 4.226238830219334e-06, + "loss": 1.7865, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 1.9662212133407593, + "eval_runtime": 43.2661, + "eval_samples_per_second": 23.113, + "eval_steps_per_second": 0.971, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.2445170879364014, + "learning_rate": 4.224207961007311e-06, + "loss": 1.791, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.399174690246582, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.7944, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.586184501647949, + "learning_rate": 4.220146222583266e-06, + "loss": 1.7856, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.4174695014953613, + "learning_rate": 4.218115353371243e-06, + "loss": 1.7842, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2684640884399414, + "learning_rate": 4.21608448415922e-06, + "loss": 1.7482, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 3.1782288551330566, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8286, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.4046335220336914, + "learning_rate": 4.212022745735175e-06, + "loss": 1.7228, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 3.3722083568573, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8044, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.293466329574585, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.7415, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.28663969039917, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8282, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 1.9627619981765747, + "eval_runtime": 43.6092, + "eval_samples_per_second": 22.931, + "eval_steps_per_second": 0.963, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.255870819091797, + "learning_rate": 4.203899268887084e-06, + "loss": 1.7983, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.4964773654937744, + "learning_rate": 4.201868399675061e-06, + "loss": 1.8141, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.3372740745544434, + "learning_rate": 4.199837530463039e-06, + "loss": 1.7142, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.3506968021392822, + "learning_rate": 4.197806661251016e-06, + "loss": 1.7898, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.242410898208618, + "learning_rate": 4.195775792038993e-06, + "loss": 1.7677, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.3864502906799316, + "learning_rate": 4.19374492282697e-06, + "loss": 1.7097, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.482349157333374, + "learning_rate": 4.191714053614948e-06, + "loss": 1.8052, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.317652940750122, + "learning_rate": 4.189683184402924e-06, + "loss": 1.7692, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.2777392864227295, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8246, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.3506035804748535, + "learning_rate": 4.185621445978879e-06, + "loss": 1.7576, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 1.9610412120819092, + "eval_runtime": 43.3119, + "eval_samples_per_second": 23.088, + "eval_steps_per_second": 0.97, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.3011536598205566, + "learning_rate": 4.183590576766857e-06, + "loss": 1.7891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.2614853382110596, + "learning_rate": 4.181559707554833e-06, + "loss": 1.8258, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.219175100326538, + "learning_rate": 4.179528838342811e-06, + "loss": 1.7494, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.4239423274993896, + "learning_rate": 4.177497969130788e-06, + "loss": 1.7611, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.470712423324585, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8281, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.504060745239258, + "learning_rate": 4.173436230706742e-06, + "loss": 1.7641, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.326601982116699, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.7671, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.4280030727386475, + "learning_rate": 4.169374492282697e-06, + "loss": 1.7573, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.455716371536255, + "learning_rate": 4.167343623070675e-06, + "loss": 1.809, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.3029844760894775, + "learning_rate": 4.165312753858652e-06, + "loss": 1.77, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 1.9604709148406982, + "eval_runtime": 44.4627, + "eval_samples_per_second": 22.491, + "eval_steps_per_second": 0.945, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.3833982944488525, + "learning_rate": 4.163281884646629e-06, + "loss": 1.7652, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.2392966747283936, + "learning_rate": 4.161251015434607e-06, + "loss": 1.7904, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.4546353816986084, + "learning_rate": 4.159220146222583e-06, + "loss": 1.7324, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.13592529296875, + "learning_rate": 4.157189277010561e-06, + "loss": 1.759, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.381657361984253, + "learning_rate": 4.155158407798538e-06, + "loss": 1.7871, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.4142940044403076, + "learning_rate": 4.153127538586516e-06, + "loss": 1.7832, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.305697441101074, + "learning_rate": 4.151096669374492e-06, + "loss": 1.7619, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.193429708480835, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.7738, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.3007619380950928, + "learning_rate": 4.147034930950447e-06, + "loss": 1.7839, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.6337807178497314, + "learning_rate": 4.145004061738425e-06, + "loss": 1.7748, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 1.9597856998443604, + "eval_runtime": 43.8644, + "eval_samples_per_second": 22.798, + "eval_steps_per_second": 0.957, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.4294793605804443, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.7001, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.512155532836914, + "learning_rate": 4.140942323314379e-06, + "loss": 1.7334, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.290808916091919, + "learning_rate": 4.138911454102356e-06, + "loss": 1.7522, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.483161211013794, + "learning_rate": 4.136880584890333e-06, + "loss": 1.798, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2973921298980713, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.7847, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.253084659576416, + "learning_rate": 4.132818846466288e-06, + "loss": 1.7812, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.306342840194702, + "learning_rate": 4.130787977254265e-06, + "loss": 1.7886, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.5153164863586426, + "learning_rate": 4.128757108042242e-06, + "loss": 1.7954, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.4122278690338135, + "learning_rate": 4.12672623883022e-06, + "loss": 1.8525, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.4520716667175293, + "learning_rate": 4.124695369618197e-06, + "loss": 1.7648, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 1.9547661542892456, + "eval_runtime": 45.3774, + "eval_samples_per_second": 22.037, + "eval_steps_per_second": 0.926, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2959611415863037, + "learning_rate": 4.122664500406174e-06, + "loss": 1.7626, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.2280471324920654, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.7954, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.3870065212249756, + "learning_rate": 4.118602761982129e-06, + "loss": 1.7377, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.292612075805664, + "learning_rate": 4.116571892770106e-06, + "loss": 1.7, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.4607021808624268, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8338, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.532824993133545, + "learning_rate": 4.112510154346061e-06, + "loss": 1.7549, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2139623165130615, + "learning_rate": 4.110479285134038e-06, + "loss": 1.78, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.4749462604522705, + "learning_rate": 4.108448415922015e-06, + "loss": 1.8034, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.609265089035034, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.7941, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.367753028869629, + "learning_rate": 4.10438667749797e-06, + "loss": 1.6653, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 1.9539440870285034, + "eval_runtime": 48.8438, + "eval_samples_per_second": 20.473, + "eval_steps_per_second": 0.86, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.9723567962646484, + "learning_rate": 4.102355808285947e-06, + "loss": 1.8317, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.388806104660034, + "learning_rate": 4.100324939073924e-06, + "loss": 1.7597, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4882986545562744, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.7708, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.7586395740509033, + "learning_rate": 4.096263200649879e-06, + "loss": 1.7804, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.237025260925293, + "learning_rate": 4.094232331437856e-06, + "loss": 1.7415, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.421330213546753, + "learning_rate": 4.092201462225833e-06, + "loss": 1.734, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.7281360626220703, + "learning_rate": 4.090170593013811e-06, + "loss": 1.7716, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.2200379371643066, + "learning_rate": 4.088139723801787e-06, + "loss": 1.7266, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.465440034866333, + "learning_rate": 4.086108854589765e-06, + "loss": 1.741, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.39835262298584, + "learning_rate": 4.084077985377742e-06, + "loss": 1.757, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 1.9514222145080566, + "eval_runtime": 52.8068, + "eval_samples_per_second": 18.937, + "eval_steps_per_second": 0.795, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.526684522628784, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8195, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.341954469680786, + "learning_rate": 4.080016246953696e-06, + "loss": 1.7196, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.3755977153778076, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8115, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.3952431678771973, + "learning_rate": 4.075954508529651e-06, + "loss": 1.812, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.554089069366455, + "learning_rate": 4.073923639317629e-06, + "loss": 1.8184, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.2319302558898926, + "learning_rate": 4.071892770105605e-06, + "loss": 1.7448, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.284856081008911, + "learning_rate": 4.069861900893583e-06, + "loss": 1.7154, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.4081835746765137, + "learning_rate": 4.06783103168156e-06, + "loss": 1.7311, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.6115968227386475, + "learning_rate": 4.065800162469538e-06, + "loss": 1.7436, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.368004322052002, + "learning_rate": 4.063769293257514e-06, + "loss": 1.7811, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 1.9508955478668213, + "eval_runtime": 46.8711, + "eval_samples_per_second": 21.335, + "eval_steps_per_second": 0.896, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.4607419967651367, + "learning_rate": 4.061738424045492e-06, + "loss": 1.7563, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.2978763580322266, + "learning_rate": 4.059707554833469e-06, + "loss": 1.7755, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.3316237926483154, + "learning_rate": 4.057676685621446e-06, + "loss": 1.8226, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.3699169158935547, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.7755, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.195596218109131, + "learning_rate": 4.053614947197401e-06, + "loss": 1.6916, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.4524662494659424, + "learning_rate": 4.051584077985378e-06, + "loss": 1.774, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.573888063430786, + "learning_rate": 4.049553208773355e-06, + "loss": 1.7953, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.5157253742218018, + "learning_rate": 4.047522339561333e-06, + "loss": 1.7597, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.5588185787200928, + "learning_rate": 4.04549147034931e-06, + "loss": 1.7553, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.8950047492980957, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8647, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 1.945882797241211, + "eval_runtime": 45.481, + "eval_samples_per_second": 21.987, + "eval_steps_per_second": 0.923, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.217116117477417, + "learning_rate": 4.041429731925264e-06, + "loss": 1.7465, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.5601565837860107, + "learning_rate": 4.039398862713242e-06, + "loss": 1.7575, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.4540457725524902, + "learning_rate": 4.037367993501219e-06, + "loss": 1.8291, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.327181100845337, + "learning_rate": 4.035337124289196e-06, + "loss": 1.7409, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.485457181930542, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.7192, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.4613442420959473, + "learning_rate": 4.031275385865151e-06, + "loss": 1.7613, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.303217649459839, + "learning_rate": 4.029244516653128e-06, + "loss": 1.7799, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.3830621242523193, + "learning_rate": 4.027213647441105e-06, + "loss": 1.7045, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.342332601547241, + "learning_rate": 4.025182778229083e-06, + "loss": 1.7246, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.4014711380004883, + "learning_rate": 4.02315190901706e-06, + "loss": 1.7621, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 1.95199453830719, + "eval_runtime": 45.4754, + "eval_samples_per_second": 21.99, + "eval_steps_per_second": 0.924, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.6015400886535645, + "learning_rate": 4.021121039805037e-06, + "loss": 1.7536, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.866978406906128, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8405, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.389810085296631, + "learning_rate": 4.017059301380992e-06, + "loss": 1.7365, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.280442953109741, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8273, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.509330987930298, + "learning_rate": 4.012997562956945e-06, + "loss": 1.7127, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.452078104019165, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.7457, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.368337631225586, + "learning_rate": 4.0089358245329e-06, + "loss": 1.7412, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.4529192447662354, + "learning_rate": 4.006904955320878e-06, + "loss": 1.7434, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.647761821746826, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8013, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.387424945831299, + "learning_rate": 4.002843216896833e-06, + "loss": 1.7948, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 1.946029543876648, + "eval_runtime": 43.9666, + "eval_samples_per_second": 22.745, + "eval_steps_per_second": 0.955, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.9217846393585205, + "learning_rate": 4.000812347684809e-06, + "loss": 1.7384, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.400040864944458, + "learning_rate": 3.998781478472787e-06, + "loss": 1.7496, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.4050440788269043, + "learning_rate": 3.996750609260764e-06, + "loss": 1.7165, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 3.6001529693603516, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8679, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 2.4040098190307617, + "learning_rate": 3.992688870836718e-06, + "loss": 1.7671, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.5645527839660645, + "learning_rate": 3.990658001624696e-06, + "loss": 1.7374, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.799394130706787, + "learning_rate": 3.988627132412673e-06, + "loss": 1.796, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.5481607913970947, + "learning_rate": 3.98659626320065e-06, + "loss": 1.753, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.8851330280303955, + "learning_rate": 3.984565393988627e-06, + "loss": 1.7505, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.6668694019317627, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.7604, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 1.939759612083435, + "eval_runtime": 44.5019, + "eval_samples_per_second": 22.471, + "eval_steps_per_second": 0.944, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.320935010910034, + "learning_rate": 3.980503655564582e-06, + "loss": 1.7749, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.829897880554199, + "learning_rate": 3.978472786352559e-06, + "loss": 1.7763, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.8420908451080322, + "learning_rate": 3.976441917140536e-06, + "loss": 1.7598, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.8183, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.781, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 3.9271891117095947, + "learning_rate": 3.972380178716491e-06, + "loss": 1.7949, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.277897834777832, + "learning_rate": 3.970349309504468e-06, + "loss": 1.707, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 3.241063117980957, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.7175, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.514458417892456, + "learning_rate": 3.966287571080423e-06, + "loss": 1.7764, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.420212745666504, + "learning_rate": 3.9642567018684e-06, + "loss": 1.7618, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 1.9401386976242065, + "eval_runtime": 44.1672, + "eval_samples_per_second": 22.641, + "eval_steps_per_second": 0.951, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.3980538845062256, + "learning_rate": 3.962225832656377e-06, + "loss": 1.6922, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.892646312713623, + "learning_rate": 3.960194963444355e-06, + "loss": 1.7571, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.499260902404785, + "learning_rate": 3.958164094232332e-06, + "loss": 1.7501, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.409695625305176, + "learning_rate": 3.956133225020309e-06, + "loss": 1.7553, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.640069007873535, + "learning_rate": 3.954102355808286e-06, + "loss": 1.7526, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.9101760387420654, + "learning_rate": 3.952071486596264e-06, + "loss": 1.7483, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.2579474449157715, + "learning_rate": 3.950040617384241e-06, + "loss": 1.7379, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 4.039298057556152, + "learning_rate": 3.948009748172218e-06, + "loss": 1.7988, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.3232030868530273, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.7592, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.697164297103882, + "learning_rate": 3.943948009748173e-06, + "loss": 1.7622, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 1.9367746114730835, + "eval_runtime": 43.7013, + "eval_samples_per_second": 22.883, + "eval_steps_per_second": 0.961, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.480297803878784, + "learning_rate": 3.94191714053615e-06, + "loss": 1.8006, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.3624603748321533, + "learning_rate": 3.939886271324127e-06, + "loss": 1.7152, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.417187213897705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.7077, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.6575164794921875, + "learning_rate": 3.935824532900082e-06, + "loss": 1.7744, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.4896557331085205, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8052, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.475693941116333, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.7008, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.3210372924804688, + "learning_rate": 3.929731925264013e-06, + "loss": 1.7339, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.433785915374756, + "learning_rate": 3.927701056051991e-06, + "loss": 1.7544, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 3.181574583053589, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.7229, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.704399347305298, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.7086, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 1.9335075616836548, + "eval_runtime": 47.0248, + "eval_samples_per_second": 21.265, + "eval_steps_per_second": 0.893, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 3.5027623176574707, + "learning_rate": 3.921608448415922e-06, + "loss": 1.7622, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.450943946838379, + "learning_rate": 3.9195775792039e-06, + "loss": 1.7286, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.4603374004364014, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.7725, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.8217649459838867, + "learning_rate": 3.915515840779855e-06, + "loss": 1.7859, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.3308372497558594, + "learning_rate": 3.913484971567831e-06, + "loss": 1.6874, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.624776840209961, + "learning_rate": 3.911454102355808e-06, + "loss": 1.7632, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.4097838401794434, + "learning_rate": 3.909423233143786e-06, + "loss": 1.7576, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.487410545349121, + "learning_rate": 3.907392363931763e-06, + "loss": 1.831, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.647245168685913, + "learning_rate": 3.90536149471974e-06, + "loss": 1.8448, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.397664785385132, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.868, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 1.930379033088684, + "eval_runtime": 51.1662, + "eval_samples_per_second": 19.544, + "eval_steps_per_second": 0.821, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.277433395385742, + "learning_rate": 3.901299756295695e-06, + "loss": 1.7394, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 3.1312496662139893, + "learning_rate": 3.899268887083672e-06, + "loss": 1.7422, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.5011534690856934, + "learning_rate": 3.897238017871649e-06, + "loss": 1.7484, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.8716166019439697, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.7253, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.4346914291381836, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8197, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.348475933074951, + "learning_rate": 3.891145410235581e-06, + "loss": 1.7737, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.3996007442474365, + "learning_rate": 3.889114541023558e-06, + "loss": 1.7333, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.355262279510498, + "learning_rate": 3.887083671811536e-06, + "loss": 1.6902, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.335932731628418, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8206, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.2893190383911133, + "learning_rate": 3.88302193338749e-06, + "loss": 1.7533, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 1.9290764331817627, + "eval_runtime": 46.9373, + "eval_samples_per_second": 21.305, + "eval_steps_per_second": 0.895, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.739842176437378, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.7318, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.587916374206543, + "learning_rate": 3.878960194963445e-06, + "loss": 1.7979, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.406548500061035, + "learning_rate": 3.876929325751422e-06, + "loss": 1.6899, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.440321922302246, + "learning_rate": 3.874898456539399e-06, + "loss": 1.7546, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.320030450820923, + "learning_rate": 3.872867587327377e-06, + "loss": 1.7254, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.3795316219329834, + "learning_rate": 3.870836718115354e-06, + "loss": 1.728, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.299802541732788, + "learning_rate": 3.868805848903331e-06, + "loss": 1.7464, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.469719409942627, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.765, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.612109899520874, + "learning_rate": 3.864744110479286e-06, + "loss": 1.7827, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.623173952102661, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8019, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 1.9268505573272705, + "eval_runtime": 46.77, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.4146616458892822, + "learning_rate": 3.86068237205524e-06, + "loss": 1.7498, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.361994743347168, + "learning_rate": 3.858651502843217e-06, + "loss": 1.6926, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.493215560913086, + "learning_rate": 3.856620633631195e-06, + "loss": 1.736, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.4341812133789062, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8152, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.391331195831299, + "learning_rate": 3.852558895207149e-06, + "loss": 1.7717, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.419492244720459, + "learning_rate": 3.850528025995126e-06, + "loss": 1.7283, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.349090576171875, + "learning_rate": 3.848497156783104e-06, + "loss": 1.7317, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.298647880554199, + "learning_rate": 3.84646628757108e-06, + "loss": 1.7389, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.4612667560577393, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.7333, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.3139944076538086, + "learning_rate": 3.842404549147035e-06, + "loss": 1.6944, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 1.924438714981079, + "eval_runtime": 45.8276, + "eval_samples_per_second": 21.821, + "eval_steps_per_second": 0.916, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 2.17958664894104, + "learning_rate": 3.840373679935013e-06, + "loss": 1.7474, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.337859869003296, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.7345, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4777672290802, + "learning_rate": 3.836311941510967e-06, + "loss": 1.77, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.280158519744873, + "learning_rate": 3.834281072298944e-06, + "loss": 1.7069, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2387664318084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.7889, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.373771905899048, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8268, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.4950132369995117, + "learning_rate": 3.828188464662876e-06, + "loss": 1.7586, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.3296444416046143, + "learning_rate": 3.826157595450853e-06, + "loss": 1.6991, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.6653530597686768, + "learning_rate": 3.82412672623883e-06, + "loss": 1.7505, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.485128402709961, + "learning_rate": 3.822095857026808e-06, + "loss": 1.7196, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 1.9221889972686768, + "eval_runtime": 46.6581, + "eval_samples_per_second": 21.433, + "eval_steps_per_second": 0.9, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.3726181983947754, + "learning_rate": 3.820064987814785e-06, + "loss": 1.7296, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.233992099761963, + "learning_rate": 3.818034118602762e-06, + "loss": 1.7313, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.5388147830963135, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.7435, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.4787333011627197, + "learning_rate": 3.813972380178717e-06, + "loss": 1.7733, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.232564926147461, + "learning_rate": 3.811941510966694e-06, + "loss": 1.693, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.3882386684417725, + "learning_rate": 3.809910641754671e-06, + "loss": 1.6529, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.4785983562469482, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.7456, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.685098886489868, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.7162, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.353523015975952, + "learning_rate": 3.803818034118603e-06, + "loss": 1.7852, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.342163562774658, + "learning_rate": 3.80178716490658e-06, + "loss": 1.7767, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 1.9235820770263672, + "eval_runtime": 47.1327, + "eval_samples_per_second": 21.217, + "eval_steps_per_second": 0.891, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.550818920135498, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.7328, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.3204798698425293, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.7058, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.4752719402313232, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8051, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.5202794075012207, + "learning_rate": 3.793663688058489e-06, + "loss": 1.7966, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.26521635055542, + "learning_rate": 3.791632818846467e-06, + "loss": 1.7879, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.331515312194824, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.7814, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.37178635597229, + "learning_rate": 3.787571080422421e-06, + "loss": 1.7635, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.321417808532715, + "learning_rate": 3.785540211210398e-06, + "loss": 1.7513, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.485989809036255, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.7306, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.7497379779815674, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8409, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 1.9275448322296143, + "eval_runtime": 46.2088, + "eval_samples_per_second": 21.641, + "eval_steps_per_second": 0.909, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 4.9214019775390625, + "learning_rate": 3.77944760357433e-06, + "loss": 1.7834, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 4.505277156829834, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.7179, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.5772030353546143, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.7766, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.254185438156128, + "learning_rate": 3.773354995938262e-06, + "loss": 1.7697, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 3.632233142852783, + "learning_rate": 3.771324126726239e-06, + "loss": 1.7968, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.174461841583252, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.7346, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.3983266353607178, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.6534, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 3.393791913986206, + "learning_rate": 3.765231519090171e-06, + "loss": 1.7561, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.2853987216949463, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.6963, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.649789810180664, + "learning_rate": 3.761169780666125e-06, + "loss": 1.7395, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 1.9242008924484253, + "eval_runtime": 46.6017, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 3.417872190475464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.784, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.316666841506958, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.7154, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.486161231994629, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.7903, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 4.538443565368652, + "learning_rate": 3.753046303818034e-06, + "loss": 1.7711, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.9772889614105225, + "learning_rate": 3.751015434606012e-06, + "loss": 1.8046, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 7.2353715896606445, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.7189, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.736, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.8026, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 12.551762580871582, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.7788, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 3.0898046493530273, + "learning_rate": 3.742891957757921e-06, + "loss": 1.7878, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 1.9646916389465332, + "eval_runtime": 44.1583, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 8.120022773742676, + "learning_rate": 3.740861088545898e-06, + "loss": 1.768, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 6.485774040222168, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.746, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.85135555267334, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.6968, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 5.8551435470581055, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.6796, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 4.45880126953125, + "learning_rate": 3.732737611697807e-06, + "loss": 1.7219, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 3.0435831546783447, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.6959, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.8194847106933594, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.8152, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 3.1266117095947266, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.6937, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 3.642854690551758, + "learning_rate": 3.724614134849716e-06, + "loss": 1.7992, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 3.344496965408325, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.7583, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 1.9293720722198486, + "eval_runtime": 50.8602, + "eval_samples_per_second": 19.662, + "eval_steps_per_second": 0.826, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.7666447162628174, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7441, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.8819239139556885, + "learning_rate": 3.718521527213648e-06, + "loss": 1.6314, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.6030640602111816, + "learning_rate": 3.716490658001625e-06, + "loss": 1.7389, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.705246925354004, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.7408, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 3.181328535079956, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7092, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 3.0449445247650146, + "learning_rate": 3.710398050365557e-06, + "loss": 1.7399, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.650517225265503, + "learning_rate": 3.708367181153534e-06, + "loss": 1.6861, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.7593672275543213, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.6701, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 3.0486466884613037, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.7191, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.502850294113159, + "learning_rate": 3.702274573517466e-06, + "loss": 1.6982, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 1.9228183031082153, + "eval_runtime": 49.565, + "eval_samples_per_second": 20.176, + "eval_steps_per_second": 0.847, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.400770664215088, + "learning_rate": 3.700243704305443e-06, + "loss": 1.6666, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.5583295822143555, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.7971, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.519505739212036, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.7076, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 3.106309413909912, + "learning_rate": 3.694151096669375e-06, + "loss": 1.7279, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.516906499862671, + "learning_rate": 3.692120227457352e-06, + "loss": 1.6265, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.603166103363037, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.7122, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.842773199081421, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.7312, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.8860342502593994, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7229, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.517254114151001, + "learning_rate": 3.683996750609261e-06, + "loss": 1.7342, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.430025100708008, + "learning_rate": 3.681965881397238e-06, + "loss": 1.6817, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 1.9156773090362549, + "eval_runtime": 44.6903, + "eval_samples_per_second": 22.376, + "eval_steps_per_second": 0.94, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.340075969696045, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.7044, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.4758968353271484, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.6935, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.878903865814209, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.7021, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.388366222381592, + "learning_rate": 3.673842404549147e-06, + "loss": 1.7229, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.5789167881011963, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8112, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.743072509765625, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.7239, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.5485000610351562, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.7641, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.53190541267395, + "learning_rate": 3.665718927701056e-06, + "loss": 1.6996, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.3354969024658203, + "learning_rate": 3.663688058489034e-06, + "loss": 1.6623, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.37371826171875, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.7109, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 1.9122159481048584, + "eval_runtime": 43.8844, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.406137228012085, + "learning_rate": 3.659626320064988e-06, + "loss": 1.6958, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.4220657348632812, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7196, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.6203134059906006, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.6014, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.5388472080230713, + "learning_rate": 3.65353371242892e-06, + "loss": 1.7272, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.672658681869507, + "learning_rate": 3.651502843216897e-06, + "loss": 1.6786, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4366531372070312, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7198, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.4872887134552, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.7082, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.371206283569336, + "learning_rate": 3.645410235580829e-06, + "loss": 1.6964, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.496103286743164, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.716, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.658316135406494, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.7263, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 1.9115047454833984, + "eval_runtime": 43.8788, + "eval_samples_per_second": 22.79, + "eval_steps_per_second": 0.957, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.5931882858276367, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.7586, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.4697139263153076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.7232, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.7274153232574463, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.6917, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.683138132095337, + "learning_rate": 3.633225020308692e-06, + "loss": 1.7518, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.4510817527770996, + "learning_rate": 3.63119415109667e-06, + "loss": 1.6533, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.6939656734466553, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.6545, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.572864532470703, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.6965, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.4816951751708984, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.6851, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.640394926071167, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7336, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.6707329750061035, + "learning_rate": 3.621039805036556e-06, + "loss": 1.7182, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 1.91274094581604, + "eval_runtime": 44.35, + "eval_samples_per_second": 22.548, + "eval_steps_per_second": 0.947, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.6947290897369385, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.7255, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.5496068000793457, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.6209, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.6196374893188477, + "learning_rate": 3.614947197400488e-06, + "loss": 1.7277, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.4534807205200195, + "learning_rate": 3.612916328188465e-06, + "loss": 1.7404, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.5428082942962646, + "learning_rate": 3.610885458976442e-06, + "loss": 1.6618, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.5431482791900635, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.6743, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.707956552505493, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.7483, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.5643982887268066, + "learning_rate": 3.604792851340374e-06, + "loss": 1.7304, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.4612927436828613, + "learning_rate": 3.602761982128351e-06, + "loss": 1.6561, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.3140358924865723, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.6834, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 1.9124692678451538, + "eval_runtime": 44.0522, + "eval_samples_per_second": 22.7, + "eval_steps_per_second": 0.953, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4636526107788086, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.7395, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.541180372238159, + "learning_rate": 3.596669374492283e-06, + "loss": 1.6579, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.621335983276367, + "learning_rate": 3.59463850528026e-06, + "loss": 1.7332, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.363450050354004, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.7047, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.8609447479248047, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7416, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.457338571548462, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.7516, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.3943030834198, + "learning_rate": 3.586515028432169e-06, + "loss": 1.6436, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.4719340801239014, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.6988, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3442330360412598, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.7165, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.3391900062561035, + "learning_rate": 3.580422420796101e-06, + "loss": 1.6263, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 1.9100929498672485, + "eval_runtime": 44.2504, + "eval_samples_per_second": 22.599, + "eval_steps_per_second": 0.949, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.533566474914551, + "learning_rate": 3.578391551584078e-06, + "loss": 1.7141, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.464540481567383, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.7335, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.4789979457855225, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7074, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.8252034187316895, + "learning_rate": 3.57229894394801e-06, + "loss": 1.6808, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.5870537757873535, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.7171, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.349393129348755, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.6687, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.3343207836151123, + "learning_rate": 3.566206336311942e-06, + "loss": 1.6351, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.558645248413086, + "learning_rate": 3.564175467099919e-06, + "loss": 1.708, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.7598931789398193, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.6797, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.6048543453216553, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7117, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 1.906927466392517, + "eval_runtime": 44.5189, + "eval_samples_per_second": 22.462, + "eval_steps_per_second": 0.943, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.6124815940856934, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.7328, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.3410215377807617, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.6932, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.5247764587402344, + "learning_rate": 3.554021121039805e-06, + "loss": 1.6628, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.642958164215088, + "learning_rate": 3.551990251827783e-06, + "loss": 1.7035, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.3978517055511475, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.6796, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.3969287872314453, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.6626, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.5723509788513184, + "learning_rate": 3.545897644191714e-06, + "loss": 1.7135, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.545154571533203, + "learning_rate": 3.543866774979692e-06, + "loss": 1.6626, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.501058340072632, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.7112, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.695483684539795, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.6773, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 1.9063464403152466, + "eval_runtime": 44.2034, + "eval_samples_per_second": 22.623, + "eval_steps_per_second": 0.95, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 3.0584375858306885, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.7242, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.445220947265625, + "learning_rate": 3.535743298131601e-06, + "loss": 1.7349, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.427220106124878, + "learning_rate": 3.533712428919578e-06, + "loss": 1.677, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.5148212909698486, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.7007, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.644319534301758, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.7442, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.6118004322052, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.7509, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.5224156379699707, + "learning_rate": 3.525588952071487e-06, + "loss": 1.7279, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.4668405055999756, + "learning_rate": 3.523558082859464e-06, + "loss": 1.6754, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 3.8330910205841064, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.6815, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.569058895111084, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.6518, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 1.906802773475647, + "eval_runtime": 50.8521, + "eval_samples_per_second": 19.665, + "eval_steps_per_second": 0.826, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.705265522003174, + "learning_rate": 3.517465475223396e-06, + "loss": 1.7172, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 3.2667057514190674, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7641, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 3.5119211673736572, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.7178, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.7348034381866455, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.6815, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 3.019892454147339, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.6417, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 3.2486684322357178, + "learning_rate": 3.507311129163282e-06, + "loss": 1.6888, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.897071361541748, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7121, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.7342352867126465, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.6339, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.691178560256958, + "learning_rate": 3.501218521527214e-06, + "loss": 1.7069, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.548682689666748, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7218, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 1.9029825925827026, + "eval_runtime": 48.287, + "eval_samples_per_second": 20.71, + "eval_steps_per_second": 0.87, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.422056198120117, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.6486, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.4367074966430664, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.759, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.48002290725708, + "learning_rate": 3.493095044679123e-06, + "loss": 1.5902, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.5030295848846436, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.7108, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.5892179012298584, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.6154, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.4282920360565186, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.6807, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.4826700687408447, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7284, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.416597843170166, + "learning_rate": 3.482940698619009e-06, + "loss": 1.6972, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.4664933681488037, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.7119, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.541856288909912, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.6752, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 1.8999215364456177, + "eval_runtime": 46.3528, + "eval_samples_per_second": 21.574, + "eval_steps_per_second": 0.906, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4803109169006348, + "learning_rate": 3.476848090982941e-06, + "loss": 1.776, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.4363205432891846, + "learning_rate": 3.474817221770918e-06, + "loss": 1.7196, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.4377009868621826, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.6937, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.425791025161743, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.6547, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3216712474823, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.6789, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.65527081489563, + "learning_rate": 3.466693744922827e-06, + "loss": 1.7231, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.3415145874023438, + "learning_rate": 3.464662875710805e-06, + "loss": 1.6805, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.3864822387695312, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.7153, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.490896224975586, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.6866, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.339711904525757, + "learning_rate": 3.458570268074736e-06, + "loss": 1.6732, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 1.8985264301300049, + "eval_runtime": 46.7854, + "eval_samples_per_second": 21.374, + "eval_steps_per_second": 0.898, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.393709659576416, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.6549, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.564114809036255, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.7019, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3826403617858887, + "learning_rate": 3.452477660438668e-06, + "loss": 1.6427, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.3185465335845947, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.6916, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4503891468048096, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.6793, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.386446714401245, + "learning_rate": 3.4463850528026e-06, + "loss": 1.7186, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.6404714584350586, + "learning_rate": 3.444354183590577e-06, + "loss": 1.6605, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.342374324798584, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.6219, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.7503273487091064, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.693, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.5416243076324463, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7323, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 1.9016659259796143, + "eval_runtime": 48.5456, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.5787603855133057, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.7057, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.649528980255127, + "learning_rate": 3.434199837530463e-06, + "loss": 1.6348, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.504089832305908, + "learning_rate": 3.432168968318441e-06, + "loss": 1.7202, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.5624678134918213, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7351, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.641331911087036, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.7453, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.599500894546509, + "learning_rate": 3.426076360682372e-06, + "loss": 1.6942, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.668599843978882, + "learning_rate": 3.42404549147035e-06, + "loss": 1.6337, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.75667667388916, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.6988, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.641526937484741, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.7641, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.4972548484802246, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.7269, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 1.8979382514953613, + "eval_runtime": 45.6555, + "eval_samples_per_second": 21.903, + "eval_steps_per_second": 0.92, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.5454585552215576, + "learning_rate": 3.415922014622259e-06, + "loss": 1.7294, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.5921928882598877, + "learning_rate": 3.413891145410236e-06, + "loss": 1.7145, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.4028453826904297, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.6985, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.604522705078125, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.6586, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.528970241546631, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.734, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.565030574798584, + "learning_rate": 3.405767668562145e-06, + "loss": 1.6997, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.7735602855682373, + "learning_rate": 3.403736799350122e-06, + "loss": 1.6543, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.5116474628448486, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.7711, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.4827239513397217, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.6831, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.544309377670288, + "learning_rate": 3.397644191714054e-06, + "loss": 1.6905, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 1.8945367336273193, + "eval_runtime": 46.6027, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.456636905670166, + "learning_rate": 3.395613322502031e-06, + "loss": 1.646, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.5900352001190186, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.6967, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 3.3344573974609375, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.6947, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.5621678829193115, + "learning_rate": 3.389520714865963e-06, + "loss": 1.7084, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.5676684379577637, + "learning_rate": 3.38748984565394e-06, + "loss": 1.6788, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.600691318511963, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.6811, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.5324196815490723, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.661, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.7559397220611572, + "learning_rate": 3.381397238017872e-06, + "loss": 1.7473, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.526397466659546, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.3752379417419434, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.689, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 1.8995447158813477, + "eval_runtime": 48.0083, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.5902793407440186, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.6899, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 3.157573938369751, + "learning_rate": 3.373273761169781e-06, + "loss": 1.6792, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.628035545349121, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.7003, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.4810822010040283, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.6489, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.4859938621520996, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.6768, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.6773345470428467, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7046, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.9695260524749756, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.7341, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.8146679401397705, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.7472, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.534985065460205, + "learning_rate": 3.359057676685622e-06, + "loss": 1.6199, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.9351718425750732, + "learning_rate": 3.357026807473599e-06, + "loss": 1.6656, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 1.8962808847427368, + "eval_runtime": 47.0443, + "eval_samples_per_second": 21.257, + "eval_steps_per_second": 0.893, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.4404256343841553, + "learning_rate": 3.354995938261576e-06, + "loss": 1.6207, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.6691436767578125, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.6627, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.5436172485351562, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.6919, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.7167909145355225, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.6434, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.575446367263794, + "learning_rate": 3.346872461413485e-06, + "loss": 1.6554, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.3874504566192627, + "learning_rate": 3.344841592201463e-06, + "loss": 1.6691, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.66317081451416, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.6864, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.4396398067474365, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.7217, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.5002803802490234, + "learning_rate": 3.338748984565394e-06, + "loss": 1.7556, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.679577350616455, + "learning_rate": 3.336718115353372e-06, + "loss": 1.6822, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 1.8942477703094482, + "eval_runtime": 44.3804, + "eval_samples_per_second": 22.532, + "eval_steps_per_second": 0.946, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.6119370460510254, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.662, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.6024818420410156, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.687, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 4.128854751586914, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.6898, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3910412788391113, + "learning_rate": 3.32859463850528e-06, + "loss": 1.6274, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.8630096912384033, + "learning_rate": 3.326563769293258e-06, + "loss": 1.6554, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 4.201050281524658, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.6676, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.492168664932251, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.6319, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 3.6807360649108887, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.6776, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.7271926403045654, + "learning_rate": 3.318440292445167e-06, + "loss": 1.6383, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.5567307472229004, + "learning_rate": 3.316409423233144e-06, + "loss": 1.7325, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 1.892768383026123, + "eval_runtime": 50.909, + "eval_samples_per_second": 19.643, + "eval_steps_per_second": 0.825, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.5373101234436035, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.689, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.602923631668091, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.6861, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.507110118865967, + "learning_rate": 3.310316815597076e-06, + "loss": 1.7575, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.361786127090454, + "learning_rate": 3.308285946385053e-06, + "loss": 1.6267, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5360615253448486, + "learning_rate": 3.30625507717303e-06, + "loss": 1.6911, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.6238744258880615, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.7193, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.5358500480651855, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.6691, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.7285404205322266, + "learning_rate": 3.300162469536962e-06, + "loss": 1.7249, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.346160650253296, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.6958, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.5415399074554443, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.7544, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 1.8925012350082397, + "eval_runtime": 46.4957, + "eval_samples_per_second": 21.507, + "eval_steps_per_second": 0.903, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.5924015045166016, + "learning_rate": 3.294069861900894e-06, + "loss": 1.6901, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.4649693965911865, + "learning_rate": 3.292038992688871e-06, + "loss": 1.7234, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.3648035526275635, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.7324, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.5146145820617676, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.631, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.5919511318206787, + "learning_rate": 3.285946385052803e-06, + "loss": 1.706, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.299527168273926, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.608, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.480637788772583, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.7415, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.458183526992798, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.6806, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.4864983558654785, + "learning_rate": 3.277822908204712e-06, + "loss": 1.6373, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.3432068824768066, + "learning_rate": 3.275792038992689e-06, + "loss": 1.6832, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 1.8899296522140503, + "eval_runtime": 44.3309, + "eval_samples_per_second": 22.558, + "eval_steps_per_second": 0.947, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.365891218185425, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.6492, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.463261842727661, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.6847, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.3977205753326416, + "learning_rate": 3.269699431356621e-06, + "loss": 1.6789, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.4172558784484863, + "learning_rate": 3.267668562144598e-06, + "loss": 1.6495, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.663381814956665, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.7139, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.3702306747436523, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.6876, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 3.1759073734283447, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.7465, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.6344592571258545, + "learning_rate": 3.259545085296507e-06, + "loss": 1.6339, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.364490032196045, + "learning_rate": 3.257514216084484e-06, + "loss": 1.7394, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.5639748573303223, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.6928, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 1.8937300443649292, + "eval_runtime": 43.9356, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 3.152968645095825, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.6693, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.45776104927063, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.6999, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.5289807319641113, + "learning_rate": 3.249390739236393e-06, + "loss": 1.6523, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.4168646335601807, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.6823, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.5115208625793457, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.6816, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.685535192489624, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.6307, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.253084897994995, + "learning_rate": 3.241267262388302e-06, + "loss": 1.662, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.4825408458709717, + "learning_rate": 3.23923639317628e-06, + "loss": 1.6444, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.691168785095215, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.6912, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.978756904602051, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7536, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 1.8903727531433105, + "eval_runtime": 43.4099, + "eval_samples_per_second": 23.036, + "eval_steps_per_second": 0.968, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.317105531692505, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.6441, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 3.1030166149139404, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.7205, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.595792770385742, + "learning_rate": 3.229082047116166e-06, + "loss": 1.677, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.3315582275390625, + "learning_rate": 3.227051177904143e-06, + "loss": 1.6779, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.4262566566467285, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.7003, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.4438798427581787, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.6081, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.573531150817871, + "learning_rate": 3.220958570268075e-06, + "loss": 1.6489, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.477809429168701, + "learning_rate": 3.218927701056052e-06, + "loss": 1.6025, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.5499629974365234, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.7337, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.4699153900146484, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.6443, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 1.8867121934890747, + "eval_runtime": 42.9343, + "eval_samples_per_second": 23.291, + "eval_steps_per_second": 0.978, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.4200708866119385, + "learning_rate": 3.212835093419984e-06, + "loss": 1.6743, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.418069839477539, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.723, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.452284336090088, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.6358, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.4241347312927246, + "learning_rate": 3.206742485783916e-06, + "loss": 1.638, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.461860418319702, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.7017, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.591986894607544, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7065, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.478470802307129, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.6262, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.49831223487854, + "learning_rate": 3.198619008935825e-06, + "loss": 1.6871, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.485332727432251, + "learning_rate": 3.196588139723802e-06, + "loss": 1.6864, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.533710479736328, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.6794, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 1.88768470287323, + "eval_runtime": 43.5948, + "eval_samples_per_second": 22.938, + "eval_steps_per_second": 0.963, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.368609666824341, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.6877, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.4875223636627197, + "learning_rate": 3.190495532087734e-06, + "loss": 1.7406, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.5075156688690186, + "learning_rate": 3.188464662875711e-06, + "loss": 1.6545, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.428731918334961, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.6213, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.467618465423584, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.6263, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.561612129211426, + "learning_rate": 3.182372055239643e-06, + "loss": 1.6987, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.5035271644592285, + "learning_rate": 3.18034118602762e-06, + "loss": 1.6902, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.538311243057251, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.6523, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.486321210861206, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.6665, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.433173894882202, + "learning_rate": 3.174248578391552e-06, + "loss": 1.6605, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 1.8853652477264404, + "eval_runtime": 43.3417, + "eval_samples_per_second": 23.072, + "eval_steps_per_second": 0.969, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.4658470153808594, + "learning_rate": 3.172217709179529e-06, + "loss": 1.6615, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.452450752258301, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.6332, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.5310161113739014, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.6846, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.4504501819610596, + "learning_rate": 3.166125101543461e-06, + "loss": 1.6996, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.6004488468170166, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.6302, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.5000431537628174, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.6742, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.4869067668914795, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.6797, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.5018959045410156, + "learning_rate": 3.15800162469537e-06, + "loss": 1.673, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.6645195484161377, + "learning_rate": 3.155970755483347e-06, + "loss": 1.6908, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.3708560466766357, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.683, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 1.8829350471496582, + "eval_runtime": 43.5383, + "eval_samples_per_second": 22.968, + "eval_steps_per_second": 0.965, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.574108362197876, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.7155, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.4669814109802246, + "learning_rate": 3.149878147847279e-06, + "loss": 1.7529, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4713149070739746, + "learning_rate": 3.147847278635256e-06, + "loss": 1.6491, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.5128800868988037, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.6276, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.740370035171509, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.6778, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.6038050651550293, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.6527, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.782407283782959, + "learning_rate": 3.139723801787165e-06, + "loss": 1.6989, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.7673134803771973, + "learning_rate": 3.137692932575143e-06, + "loss": 1.6291, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.7472870349884033, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.6776, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.4887819290161133, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7009, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.8792160749435425, + "eval_runtime": 43.7334, + "eval_samples_per_second": 22.866, + "eval_steps_per_second": 0.96, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.4528491497039795, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.7245, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.5177690982818604, + "learning_rate": 3.129569455727051e-06, + "loss": 1.6852, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.9238510131835938, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.6994, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.446660041809082, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.6636, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.644211530685425, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.6983, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.697342872619629, + "learning_rate": 3.12144597887896e-06, + "loss": 1.7172, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.714172124862671, + "learning_rate": 3.119415109666938e-06, + "loss": 1.6702, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.471735715866089, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.6945, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.5089669227600098, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.7181, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.349229335784912, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.7297, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.880064845085144, + "eval_runtime": 47.0636, + "eval_samples_per_second": 21.248, + "eval_steps_per_second": 0.892, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.6843974590301514, + "learning_rate": 3.111291632818847e-06, + "loss": 1.6698, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.4325790405273438, + "learning_rate": 3.109260763606824e-06, + "loss": 1.6973, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.488459825515747, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.6552, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.4992082118988037, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.7114, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.4742603302001953, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.7073, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.556330442428589, + "learning_rate": 3.101137286758733e-06, + "loss": 1.7108, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.5711283683776855, + "learning_rate": 3.09910641754671e-06, + "loss": 1.6811, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.6467158794403076, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.7395, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.465963363647461, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.7006, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.622716188430786, + "learning_rate": 3.093013809910642e-06, + "loss": 1.7385, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.8808013200759888, + "eval_runtime": 45.5126, + "eval_samples_per_second": 21.972, + "eval_steps_per_second": 0.923, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.489701509475708, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.6949, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.4398739337921143, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.6802, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.3654873371124268, + "learning_rate": 3.086921202274574e-06, + "loss": 1.6452, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.518768548965454, + "learning_rate": 3.084890333062551e-06, + "loss": 1.6913, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.63173246383667, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.6611, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.531484603881836, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.611, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.4796738624572754, + "learning_rate": 3.078797725426483e-06, + "loss": 1.6764, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.39241099357605, + "learning_rate": 3.07676685621446e-06, + "loss": 1.6417, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3984735012054443, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.6546, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.5202760696411133, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.6879, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.879974365234375, + "eval_runtime": 44.395, + "eval_samples_per_second": 22.525, + "eval_steps_per_second": 0.946, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.5556983947753906, + "learning_rate": 3.070674248578392e-06, + "loss": 1.6733, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.3853955268859863, + "learning_rate": 3.068643379366369e-06, + "loss": 1.6274, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.4503982067108154, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.6896, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.5618021488189697, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.6592, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.6265292167663574, + "learning_rate": 3.062550771730301e-06, + "loss": 1.6634, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.4617390632629395, + "learning_rate": 3.060519902518278e-06, + "loss": 1.6032, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.4191603660583496, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.6871, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.5509119033813477, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.653, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.5399646759033203, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.7046, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.446967124938965, + "learning_rate": 3.052396425670187e-06, + "loss": 1.6547, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.8803874254226685, + "eval_runtime": 45.361, + "eval_samples_per_second": 22.045, + "eval_steps_per_second": 0.926, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.419206142425537, + "learning_rate": 3.050365556458164e-06, + "loss": 1.6517, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.4592065811157227, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.693, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.5429770946502686, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7124, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.5141642093658447, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.7103, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.362539291381836, + "learning_rate": 3.042242079610073e-06, + "loss": 1.6205, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.401175022125244, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.6479, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.5883190631866455, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.6732, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.5273513793945312, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.6673, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.6569600105285645, + "learning_rate": 3.034118602761982e-06, + "loss": 1.6396, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.8843557834625244, + "learning_rate": 3.03208773354996e-06, + "loss": 1.7136, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.8763896226882935, + "eval_runtime": 46.1886, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.6459810733795166, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.6939, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.4619321823120117, + "learning_rate": 3.028025995125914e-06, + "loss": 1.6197, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.390430450439453, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.6805, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.5494728088378906, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.6116, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.5397655963897705, + "learning_rate": 3.021933387489846e-06, + "loss": 1.6753, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.634716510772705, + "learning_rate": 3.019902518277823e-06, + "loss": 1.6892, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.5844202041625977, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.6401, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.577362060546875, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.6405, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.3554489612579346, + "learning_rate": 3.013809910641755e-06, + "loss": 1.6427, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.893918037414551, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.6739, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.8772739171981812, + "eval_runtime": 44.9972, + "eval_samples_per_second": 22.224, + "eval_steps_per_second": 0.933, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.4818382263183594, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.644, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.4990177154541016, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.6962, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.627502918243408, + "learning_rate": 3.005686433793664e-06, + "loss": 1.639, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.5949175357818604, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.7095, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.422765016555786, + "learning_rate": 3.001624695369618e-06, + "loss": 1.6572, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.708702564239502, + "learning_rate": 2.999593826157596e-06, + "loss": 1.6747, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.624866247177124, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.6643, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.5831358432769775, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.7036, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.5799436569213867, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.6628, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.7047722339630127, + "learning_rate": 2.991470349309505e-06, + "loss": 1.6591, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.8736577033996582, + "eval_runtime": 44.7095, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.411628246307373, + "learning_rate": 2.989439480097482e-06, + "loss": 1.6644, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.8774638175964355, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.7192, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.857028007507324, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7079, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.4619641304016113, + "learning_rate": 2.983346872461414e-06, + "loss": 1.6846, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.6566193103790283, + "learning_rate": 2.981316003249391e-06, + "loss": 1.6768, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.681217670440674, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.7002, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.39505672454834, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.65, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.6062734127044678, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.6659, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.6402504444122314, + "learning_rate": 2.9731925264013e-06, + "loss": 1.6491, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.497014284133911, + "learning_rate": 2.971161657189277e-06, + "loss": 1.6383, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.874670147895813, + "eval_runtime": 42.9696, + "eval_samples_per_second": 23.272, + "eval_steps_per_second": 0.977, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.9070584774017334, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.6388, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.580744981765747, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7078, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.4637644290924072, + "learning_rate": 2.965069049553209e-06, + "loss": 1.6266, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.4489941596984863, + "learning_rate": 2.963038180341186e-06, + "loss": 1.705, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.664654016494751, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.6964, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.7017202377319336, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.6356, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.3691024780273438, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.6566, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.540384292602539, + "learning_rate": 2.954914703493095e-06, + "loss": 1.6334, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.5240535736083984, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.6316, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 3.105684518814087, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.6769, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.8741132020950317, + "eval_runtime": 42.6795, + "eval_samples_per_second": 23.43, + "eval_steps_per_second": 0.984, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.494252920150757, + "learning_rate": 2.948822095857027e-06, + "loss": 1.6359, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 4.069019317626953, + "learning_rate": 2.946791226645004e-06, + "loss": 1.621, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.641447067260742, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.6502, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.6700453758239746, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.6593, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.686068296432495, + "learning_rate": 2.940698619008936e-06, + "loss": 1.67, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.409055471420288, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.6809, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.4707298278808594, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.69, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.4735982418060303, + "learning_rate": 2.934606011372868e-06, + "loss": 1.666, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.633183479309082, + "learning_rate": 2.932575142160845e-06, + "loss": 1.6597, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.4808642864227295, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.6667, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.8731157779693604, + "eval_runtime": 41.3423, + "eval_samples_per_second": 24.188, + "eval_steps_per_second": 1.016, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.4044833183288574, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.7028, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.942699909210205, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.6678, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.505645990371704, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.5972, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.700967788696289, + "learning_rate": 2.922420796100731e-06, + "loss": 1.7064, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.4805755615234375, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.6201, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.753181219100952, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.6704, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.488865613937378, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7057, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.5464885234832764, + "learning_rate": 2.91429731925264e-06, + "loss": 1.6678, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.4997060298919678, + "learning_rate": 2.912266450040618e-06, + "loss": 1.662, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.5162436962127686, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.6501, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.869848370552063, + "eval_runtime": 43.3583, + "eval_samples_per_second": 23.064, + "eval_steps_per_second": 0.969, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.616129159927368, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.6206, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.761080503463745, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7131, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.5994064807891846, + "learning_rate": 2.904142973192527e-06, + "loss": 1.6693, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 3.4927968978881836, + "learning_rate": 2.902112103980504e-06, + "loss": 1.6694, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.6050939559936523, + "learning_rate": 2.900081234768481e-06, + "loss": 1.6419, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.796086311340332, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.6354, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.5658411979675293, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.652, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.3802385330200195, + "learning_rate": 2.893988627132413e-06, + "loss": 1.5902, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.479189872741699, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.6617, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.4367568492889404, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.6691, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.868497371673584, + "eval_runtime": 45.172, + "eval_samples_per_second": 22.138, + "eval_steps_per_second": 0.93, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.51888370513916, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.6872, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.5192768573760986, + "learning_rate": 2.885865150284322e-06, + "loss": 1.6584, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.5898501873016357, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.6179, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.698101043701172, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.663, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.889942169189453, + "learning_rate": 2.879772542648254e-06, + "loss": 1.5954, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.7909929752349854, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.7264, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 3.1931703090667725, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7091, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.7934632301330566, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.7204, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.470545530319214, + "learning_rate": 2.871649065800163e-06, + "loss": 1.644, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 3.011662006378174, + "learning_rate": 2.86961819658814e-06, + "loss": 1.6743, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.8678737878799438, + "eval_runtime": 44.7875, + "eval_samples_per_second": 22.328, + "eval_steps_per_second": 0.938, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.60823655128479, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.6467, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.5347230434417725, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.7044, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.481715440750122, + "learning_rate": 2.863525588952072e-06, + "loss": 1.5969, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.4966487884521484, + "learning_rate": 2.861494719740049e-06, + "loss": 1.6231, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.8809802532196045, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.6562, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.570204734802246, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.6511, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.6073813438415527, + "learning_rate": 2.855402112103981e-06, + "loss": 1.6456, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.8125052452087402, + "learning_rate": 2.853371242891958e-06, + "loss": 1.6722, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 5.341272354125977, + "learning_rate": 2.851340373679935e-06, + "loss": 1.7025, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 3.138002395629883, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.6676, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.8733265399932861, + "eval_runtime": 45.4915, + "eval_samples_per_second": 21.982, + "eval_steps_per_second": 0.923, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 7.264789581298828, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.6335, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 3.8533623218536377, + "learning_rate": 2.845247766043867e-06, + "loss": 1.6029, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 4.1789164543151855, + "learning_rate": 2.843216896831844e-06, + "loss": 1.6368, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 3.117408037185669, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.6924, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.721306085586548, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.6914, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.7099668979644775, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.6508, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 3.7823967933654785, + "learning_rate": 2.835093419983753e-06, + "loss": 1.636, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.696370840072632, + "learning_rate": 2.833062550771731e-06, + "loss": 1.6175, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.7303690910339355, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.6654, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.867727041244507, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.6717, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.8720332384109497, + "eval_runtime": 44.164, + "eval_samples_per_second": 22.643, + "eval_steps_per_second": 0.951, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 3.1209630966186523, + "learning_rate": 2.826969943135662e-06, + "loss": 1.6635, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.5057783126831055, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.6309, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.6650168895721436, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.6365, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.371305465698242, + "learning_rate": 2.820877335499594e-06, + "loss": 1.603, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.387404203414917, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.6189, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.6265947818756104, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.6511, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.6108627319335938, + "learning_rate": 2.814784727863526e-06, + "loss": 1.6727, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.7021450996398926, + "learning_rate": 2.812753858651503e-06, + "loss": 1.6334, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.470118999481201, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.6155, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5722644329071045, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.6788, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.8727209568023682, + "eval_runtime": 44.6253, + "eval_samples_per_second": 22.409, + "eval_steps_per_second": 0.941, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.7143876552581787, + "learning_rate": 2.806661251015435e-06, + "loss": 1.6956, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.8522002696990967, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.7558, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.5352110862731934, + "learning_rate": 2.802599512591389e-06, + "loss": 1.6592, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.7065908908843994, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.6121, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.616450786590576, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.6643, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.528763771057129, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7104, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 3.725569009780884, + "learning_rate": 2.794476035743298e-06, + "loss": 1.6713, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.487497091293335, + "learning_rate": 2.792445166531276e-06, + "loss": 1.692, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.534766912460327, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.6257, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.8553240299224854, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.6147, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.8675988912582397, + "eval_runtime": 46.4445, + "eval_samples_per_second": 21.531, + "eval_steps_per_second": 0.904, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.5054166316986084, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.6082, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.4133870601654053, + "learning_rate": 2.784321689683185e-06, + "loss": 1.6148, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.5863375663757324, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7032, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.58491849899292, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.679, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.5225989818573, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.6688, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.366861343383789, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.7076, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.4159810543060303, + "learning_rate": 2.774167343623071e-06, + "loss": 1.6301, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.5394275188446045, + "learning_rate": 2.772136474411048e-06, + "loss": 1.5758, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.409821033477783, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.664, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.4548892974853516, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.643, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.8672850131988525, + "eval_runtime": 46.3149, + "eval_samples_per_second": 21.591, + "eval_steps_per_second": 0.907, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.7483904361724854, + "learning_rate": 2.76604386677498e-06, + "loss": 1.6898, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.538804769515991, + "learning_rate": 2.764012997562957e-06, + "loss": 1.6019, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.7153313159942627, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.691, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.6892242431640625, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.6386, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.550957679748535, + "learning_rate": 2.757920389926889e-06, + "loss": 1.6237, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.3658559322357178, + "learning_rate": 2.755889520714866e-06, + "loss": 1.6015, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.6524102687835693, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.6644, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.7067437171936035, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.687, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.4044861793518066, + "learning_rate": 2.749796913078798e-06, + "loss": 1.6535, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.4933829307556152, + "learning_rate": 2.747766043866775e-06, + "loss": 1.7017, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.8652688264846802, + "eval_runtime": 46.8502, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.7062129974365234, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7398, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.5704996585845947, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.6561, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.5875790119171143, + "learning_rate": 2.741673436230707e-06, + "loss": 1.643, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.7498223781585693, + "learning_rate": 2.739642567018684e-06, + "loss": 1.6652, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.7275257110595703, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.6559, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.6349947452545166, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.6489, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.7058634757995605, + "learning_rate": 2.733549959382616e-06, + "loss": 1.7176, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.481344223022461, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.685, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.468277931213379, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.6305, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.6183228492736816, + "learning_rate": 2.727457351746548e-06, + "loss": 1.6151, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.864329218864441, + "eval_runtime": 43.0443, + "eval_samples_per_second": 23.232, + "eval_steps_per_second": 0.976, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.50679349899292, + "learning_rate": 2.725426482534525e-06, + "loss": 1.6468, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.6565401554107666, + "learning_rate": 2.723395613322502e-06, + "loss": 1.6496, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.5332367420196533, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.6424, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.6645619869232178, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.8012, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.7359142303466797, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.6025, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.660728693008423, + "learning_rate": 2.715272136474411e-06, + "loss": 1.6601, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.5184221267700195, + "learning_rate": 2.713241267262389e-06, + "loss": 1.6383, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.643712043762207, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.6834, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.504753351211548, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.6292, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.3439948558807373, + "learning_rate": 2.70714865962632e-06, + "loss": 1.6558, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.861581563949585, + "eval_runtime": 43.9224, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4820125102996826, + "learning_rate": 2.705117790414298e-06, + "loss": 1.69, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.6156861782073975, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.6304, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.490123987197876, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.6573, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.4439024925231934, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.6368, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.891603708267212, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7111, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.481175661087036, + "learning_rate": 2.694963444354184e-06, + "loss": 1.6408, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.5378847122192383, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.6782, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.592477321624756, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.6085, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.9336471557617188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.6719, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.3115012645721436, + "learning_rate": 2.686839967506093e-06, + "loss": 1.5797, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.8649041652679443, + "eval_runtime": 43.711, + "eval_samples_per_second": 22.878, + "eval_steps_per_second": 0.961, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.627028465270996, + "learning_rate": 2.68480909829407e-06, + "loss": 1.5784, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.917670488357544, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7007, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.418285608291626, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.6613, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.607773780822754, + "learning_rate": 2.678716490658002e-06, + "loss": 1.6561, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.569714307785034, + "learning_rate": 2.676685621445979e-06, + "loss": 1.6541, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.359494209289551, + "learning_rate": 2.674654752233956e-06, + "loss": 1.6467, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.5155694484710693, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.7003, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.439147710800171, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.5679, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.5044116973876953, + "learning_rate": 2.668562144597888e-06, + "loss": 1.6626, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.4475765228271484, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.656, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.8636877536773682, + "eval_runtime": 43.5138, + "eval_samples_per_second": 22.981, + "eval_steps_per_second": 0.965, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4616832733154297, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.6101, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.4125123023986816, + "learning_rate": 2.66246953696182e-06, + "loss": 1.6021, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.571326494216919, + "learning_rate": 2.660438667749797e-06, + "loss": 1.6482, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.424332618713379, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.653, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.3034703731536865, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.6096, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.4758505821228027, + "learning_rate": 2.654346060113729e-06, + "loss": 1.6274, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 3.800152540206909, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.6217, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.479827880859375, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7202, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.604523181915283, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.6778, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 3.595973491668701, + "learning_rate": 2.646222583265638e-06, + "loss": 1.6678, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.8647090196609497, + "eval_runtime": 43.0415, + "eval_samples_per_second": 23.233, + "eval_steps_per_second": 0.976, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 3.297319173812866, + "learning_rate": 2.644191714053615e-06, + "loss": 1.637, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.423567056655884, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.6018, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 4.4547953605651855, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.6594, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 3.0775582790374756, + "learning_rate": 2.638099106417547e-06, + "loss": 1.6259, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.747938871383667, + "learning_rate": 2.636068237205524e-06, + "loss": 1.7208, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.5350255966186523, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.5705, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 3.0630743503570557, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.6507, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 3.5475759506225586, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7045, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.534303903579712, + "learning_rate": 2.627944760357433e-06, + "loss": 1.7173, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.6964452266693115, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.6823, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.865207552909851, + "eval_runtime": 42.9435, + "eval_samples_per_second": 23.286, + "eval_steps_per_second": 0.978, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.924957752227783, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.6036, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.7058141231536865, + "learning_rate": 2.621852152721365e-06, + "loss": 1.6701, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.735339641571045, + "learning_rate": 2.619821283509342e-06, + "loss": 1.607, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.573958396911621, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.667, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5776426792144775, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.6938, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 4.454444408416748, + "learning_rate": 2.613728675873274e-06, + "loss": 1.7139, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.4883337020874023, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.6524, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.672116994857788, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.6197, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 3.511932611465454, + "learning_rate": 2.607636068237206e-06, + "loss": 1.6739, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.5960047245025635, + "learning_rate": 2.605605199025183e-06, + "loss": 1.6516, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.8604435920715332, + "eval_runtime": 42.9247, + "eval_samples_per_second": 23.297, + "eval_steps_per_second": 0.978, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.565376043319702, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.6884, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.676114797592163, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.6693, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.582751750946045, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.6499, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.748908281326294, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.6837, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.5578067302703857, + "learning_rate": 2.595450852965069e-06, + "loss": 1.6959, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.5702805519104004, + "learning_rate": 2.593419983753047e-06, + "loss": 1.6429, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.4748787879943848, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.6878, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.6157331466674805, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.6246, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.8377959728240967, + "learning_rate": 2.587327376116978e-06, + "loss": 1.6451, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.485398292541504, + "learning_rate": 2.585296506904956e-06, + "loss": 1.6288, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.8593864440917969, + "eval_runtime": 44.6341, + "eval_samples_per_second": 22.404, + "eval_steps_per_second": 0.941, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.46850323677063, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.5766, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.673283576965332, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7383, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.935293197631836, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.5742, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.531733751296997, + "learning_rate": 2.577173030056865e-06, + "loss": 1.6714, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.528425455093384, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6802, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.5644803047180176, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.6201, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.382451295852661, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.5824, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.63320255279541, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.6685, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.488844394683838, + "learning_rate": 2.567018683996751e-06, + "loss": 1.6078, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.7845425605773926, + "learning_rate": 2.564987814784728e-06, + "loss": 1.638, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.8572624921798706, + "eval_runtime": 43.2181, + "eval_samples_per_second": 23.138, + "eval_steps_per_second": 0.972, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.709656000137329, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.6425, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.651277542114258, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7401, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.5424256324768066, + "learning_rate": 2.55889520714866e-06, + "loss": 1.6272, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.6508235931396484, + "learning_rate": 2.556864337936637e-06, + "loss": 1.6384, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.882636308670044, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.6965, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.518367290496826, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.6551, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.7977044582366943, + "learning_rate": 2.550771730300569e-06, + "loss": 1.679, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.353041172027588, + "learning_rate": 2.548740861088546e-06, + "loss": 1.5745, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.5718655586242676, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.5655, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.6955313682556152, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.6764, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.8569272756576538, + "eval_runtime": 45.3844, + "eval_samples_per_second": 22.034, + "eval_steps_per_second": 0.925, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4756696224212646, + "learning_rate": 2.542648253452478e-06, + "loss": 1.6522, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.573193311691284, + "learning_rate": 2.540617384240455e-06, + "loss": 1.6079, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.5967671871185303, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.6341, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.7896342277526855, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.6449, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.519728183746338, + "learning_rate": 2.534524776604387e-06, + "loss": 1.549, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.543139934539795, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.6335, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.5580177307128906, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6402, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.443547010421753, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.6421, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.7075212001800537, + "learning_rate": 2.526401299756296e-06, + "loss": 1.6283, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.6505324840545654, + "learning_rate": 2.524370430544273e-06, + "loss": 1.6028, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.8556184768676758, + "eval_runtime": 43.731, + "eval_samples_per_second": 22.867, + "eval_steps_per_second": 0.96, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.597172260284424, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.6631, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 3.1529037952423096, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.6219, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.5374975204467773, + "learning_rate": 2.518277822908205e-06, + "loss": 1.6476, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 3.0557401180267334, + "learning_rate": 2.516246953696182e-06, + "loss": 1.6627, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 6.759350776672363, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.6532, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.670427083969116, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.658, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 4.269686698913574, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.5873, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 13.37558650970459, + "learning_rate": 2.508123476848091e-06, + "loss": 1.6872, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 7.060877323150635, + "learning_rate": 2.506092607636069e-06, + "loss": 1.6406, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 11.259824752807617, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.6421, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.861360788345337, + "eval_runtime": 45.4438, + "eval_samples_per_second": 22.005, + "eval_steps_per_second": 0.924, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 5.861843585968018, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.701, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 3.9136719703674316, + "learning_rate": 2.5e-06, + "loss": 1.6289, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 12.720739364624023, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.6532, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 5.970798492431641, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7035, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 3.490950584411621, + "learning_rate": 2.493907392363932e-06, + "loss": 1.6565, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 5.826146602630615, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.655, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 9.610876083374023, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.7295, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 3.754833221435547, + "learning_rate": 2.487814784727864e-06, + "loss": 1.6406, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.7952232360839844, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.6463, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 6.3178558349609375, + "learning_rate": 2.483753046303818e-06, + "loss": 1.6474, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.8672106266021729, + "eval_runtime": 50.0163, + "eval_samples_per_second": 19.993, + "eval_steps_per_second": 0.84, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 7.4854960441589355, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.6798, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.8903026580810547, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.667, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.481947898864746, + "learning_rate": 2.47766043866775e-06, + "loss": 1.6677, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.8459818363189697, + "learning_rate": 2.475629569455727e-06, + "loss": 1.6129, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 5.00568962097168, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.6233, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 3.7508153915405273, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.6414, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.7895889282226562, + "learning_rate": 2.469536961819659e-06, + "loss": 1.65, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.532818555831909, + "learning_rate": 2.467506092607636e-06, + "loss": 1.6174, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.699639320373535, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7128, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.5636072158813477, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.6416, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.8608473539352417, + "eval_runtime": 43.8315, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.7466976642608643, + "learning_rate": 2.461413484971568e-06, + "loss": 1.7246, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.690990686416626, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.596, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.868340492248535, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.622, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.4456427097320557, + "learning_rate": 2.4553208773355e-06, + "loss": 1.6446, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.5499041080474854, + "learning_rate": 2.453290008123477e-06, + "loss": 1.6587, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.68770170211792, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.6612, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.589557647705078, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.6428, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.7014052867889404, + "learning_rate": 2.447197400487409e-06, + "loss": 1.6605, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.6242196559906006, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.6897, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.5285110473632812, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.6636, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.8544306755065918, + "eval_runtime": 45.6109, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.484952449798584, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.6636, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.5793070793151855, + "learning_rate": 2.439073923639318e-06, + "loss": 1.6044, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.5558605194091797, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.5885, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.6365678310394287, + "learning_rate": 2.435012185215272e-06, + "loss": 1.6236, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.428422451019287, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.6446, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.535776138305664, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.6252, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.743934154510498, + "learning_rate": 2.428919577579204e-06, + "loss": 1.729, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.4823739528656006, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.6251, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.4650003910064697, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.6191, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.4338109493255615, + "learning_rate": 2.422826969943136e-06, + "loss": 1.5919, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.8501240015029907, + "eval_runtime": 43.8558, + "eval_samples_per_second": 22.802, + "eval_steps_per_second": 0.958, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.6054813861846924, + "learning_rate": 2.420796100731113e-06, + "loss": 1.6485, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.390143394470215, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.661, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.467470645904541, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.6308, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.360908269882202, + "learning_rate": 2.414703493095045e-06, + "loss": 1.6394, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.5966176986694336, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.6138, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.528402328491211, + "learning_rate": 2.410641754670999e-06, + "loss": 1.6817, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.5319254398345947, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.6189, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.559260129928589, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.6476, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.5595016479492188, + "learning_rate": 2.404549147034931e-06, + "loss": 1.6878, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.488196849822998, + "learning_rate": 2.402518277822908e-06, + "loss": 1.6095, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.8484610319137573, + "eval_runtime": 43.4829, + "eval_samples_per_second": 22.998, + "eval_steps_per_second": 0.966, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6928117275238037, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.6533, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.485538959503174, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.6174, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.4725208282470703, + "learning_rate": 2.39642567018684e-06, + "loss": 1.6306, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.456028461456299, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.6586, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3731250762939453, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.6009, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.4567086696624756, + "learning_rate": 2.390333062550772e-06, + "loss": 1.5257, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.577141284942627, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7127, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.4612233638763428, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.638, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.463212728500366, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.5933, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.5121495723724365, + "learning_rate": 2.382209585702681e-06, + "loss": 1.68, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.8499811887741089, + "eval_runtime": 45.066, + "eval_samples_per_second": 22.19, + "eval_steps_per_second": 0.932, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.658245801925659, + "learning_rate": 2.380178716490658e-06, + "loss": 1.6354, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.6265509128570557, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.6755, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 3.016807794570923, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.6764, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.6552720069885254, + "learning_rate": 2.37408610885459e-06, + "loss": 1.6281, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.520078659057617, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.7077, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.528329849243164, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.6494, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.623776912689209, + "learning_rate": 2.367993501218522e-06, + "loss": 1.6436, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.56669020652771, + "learning_rate": 2.365962632006499e-06, + "loss": 1.6762, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.5415968894958496, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.6711, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.5164198875427246, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.6725, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.8490949869155884, + "eval_runtime": 43.612, + "eval_samples_per_second": 22.929, + "eval_steps_per_second": 0.963, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.626594066619873, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.6637, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3480069637298584, + "learning_rate": 2.357839155158408e-06, + "loss": 1.6561, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.644103527069092, + "learning_rate": 2.355808285946385e-06, + "loss": 1.6763, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.5539023876190186, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.653, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.5055317878723145, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.6643, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.4204092025756836, + "learning_rate": 2.349715678310317e-06, + "loss": 1.6133, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.5867843627929688, + "learning_rate": 2.347684809098294e-06, + "loss": 1.6804, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.5993685722351074, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.6101, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.565439462661743, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.6221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.5296084880828857, + "learning_rate": 2.341592201462226e-06, + "loss": 1.6234, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.8482695817947388, + "eval_runtime": 45.1972, + "eval_samples_per_second": 22.125, + "eval_steps_per_second": 0.929, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.6142499446868896, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.5886, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.486616849899292, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.6097, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.5778701305389404, + "learning_rate": 2.335499593826158e-06, + "loss": 1.6585, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.6165313720703125, + "learning_rate": 2.333468724614135e-06, + "loss": 1.663, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.4976565837860107, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.6083, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.485017776489258, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.6006, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.620819330215454, + "learning_rate": 2.327376116978067e-06, + "loss": 1.6641, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.528599262237549, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.6128, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.541442632675171, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.6545, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.4722139835357666, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.6146, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.848878264427185, + "eval_runtime": 43.7961, + "eval_samples_per_second": 22.833, + "eval_steps_per_second": 0.959, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.6106903553009033, + "learning_rate": 2.319252640129976e-06, + "loss": 1.7086, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.5255045890808105, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.6738, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.43829083442688, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.5861, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.527411460876465, + "learning_rate": 2.313160032493908e-06, + "loss": 1.6634, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.4998528957366943, + "learning_rate": 2.311129163281885e-06, + "loss": 1.6187, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.4838368892669678, + "learning_rate": 2.309098294069862e-06, + "loss": 1.5999, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.466900110244751, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.6446, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.583103895187378, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.6658, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.478386163711548, + "learning_rate": 2.303005686433794e-06, + "loss": 1.6261, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.5724942684173584, + "learning_rate": 2.300974817221771e-06, + "loss": 1.5955, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.8444501161575317, + "eval_runtime": 50.0263, + "eval_samples_per_second": 19.989, + "eval_steps_per_second": 0.84, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.601644277572632, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.6493, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.5638768672943115, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.6702, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.4802567958831787, + "learning_rate": 2.294882209585703e-06, + "loss": 1.6036, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.5114285945892334, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.642, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.487563133239746, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.673, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.6014463901519775, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.6261, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.53680157661438, + "learning_rate": 2.286758732737612e-06, + "loss": 1.6364, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.764408588409424, + "learning_rate": 2.284727863525589e-06, + "loss": 1.6791, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.8998825550079346, + "learning_rate": 2.282696994313566e-06, + "loss": 1.644, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.5797383785247803, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.6267, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.8467868566513062, + "eval_runtime": 49.4778, + "eval_samples_per_second": 20.211, + "eval_steps_per_second": 0.849, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.447178602218628, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.5955, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.6275408267974854, + "learning_rate": 2.276604386677498e-06, + "loss": 1.6117, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.6952037811279297, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.6894, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.5299718379974365, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.6361, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.837455987930298, + "learning_rate": 2.27051177904143e-06, + "loss": 1.6269, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.5406227111816406, + "learning_rate": 2.268480909829407e-06, + "loss": 1.6538, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.716038227081299, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.6691, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.883439064025879, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.6298, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.7631118297576904, + "learning_rate": 2.262388302193339e-06, + "loss": 1.6245, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.6777117252349854, + "learning_rate": 2.260357432981316e-06, + "loss": 1.6446, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.845224142074585, + "eval_runtime": 46.8476, + "eval_samples_per_second": 21.346, + "eval_steps_per_second": 0.897, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.6280243396759033, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.6398, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.471925973892212, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.6403, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.6293203830718994, + "learning_rate": 2.254264825345248e-06, + "loss": 1.642, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.651144027709961, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.6445, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.6489877700805664, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.6761, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.5869088172912598, + "learning_rate": 2.24817221770918e-06, + "loss": 1.6593, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.411802053451538, + "learning_rate": 2.246141348497157e-06, + "loss": 1.6232, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.522610902786255, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.6358, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.5430991649627686, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.6605, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.4440951347351074, + "learning_rate": 2.240048740861089e-06, + "loss": 1.5354, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.8440407514572144, + "eval_runtime": 48.0239, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.5492775440216064, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.668, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.7511909008026123, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6424, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.5176775455474854, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.6336, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.5408577919006348, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.5843, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.4791393280029297, + "learning_rate": 2.229894394800975e-06, + "loss": 1.555, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.383148670196533, + "learning_rate": 2.227863525588952e-06, + "loss": 1.598, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.429348945617676, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.5943, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.708796501159668, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.6382, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 3.0029661655426025, + "learning_rate": 2.221770917952884e-06, + "loss": 1.6699, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.5410728454589844, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.5965, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.8443207740783691, + "eval_runtime": 47.8671, + "eval_samples_per_second": 20.891, + "eval_steps_per_second": 0.877, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.5718178749084473, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.6214, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.474177837371826, + "learning_rate": 2.215678310316816e-06, + "loss": 1.6498, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.5960052013397217, + "learning_rate": 2.213647441104793e-06, + "loss": 1.6156, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.6217992305755615, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.6311, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.5601158142089844, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.6008, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.7786355018615723, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.62, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.548356771469116, + "learning_rate": 2.205523964256702e-06, + "loss": 1.6473, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.6265289783477783, + "learning_rate": 2.203493095044679e-06, + "loss": 1.6565, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.522379159927368, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.6764, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.5479297637939453, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.6194, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.8434110879898071, + "eval_runtime": 46.6043, + "eval_samples_per_second": 21.457, + "eval_steps_per_second": 0.901, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.5401806831359863, + "learning_rate": 2.197400487408611e-06, + "loss": 1.5911, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.466893434524536, + "learning_rate": 2.195369618196588e-06, + "loss": 1.6145, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.566653251647949, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.6558, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.664053201675415, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7089, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 3.6594645977020264, + "learning_rate": 2.18927701056052e-06, + "loss": 1.6549, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.68450927734375, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.6306, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.7094614505767822, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.6147, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.481332540512085, + "learning_rate": 2.183184402924452e-06, + "loss": 1.6612, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5651051998138428, + "learning_rate": 2.181153533712429e-06, + "loss": 1.6394, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.5570919513702393, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.6162, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.8433003425598145, + "eval_runtime": 46.8673, + "eval_samples_per_second": 21.337, + "eval_steps_per_second": 0.896, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.6206037998199463, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.6869, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.8962907791137695, + "learning_rate": 2.175060926076361e-06, + "loss": 1.7022, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.6211588382720947, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.6249, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.7770566940307617, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.5821, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.543658494949341, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.6587, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.508188486099243, + "learning_rate": 2.16693744922827e-06, + "loss": 1.6114, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.703335762023926, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.6152, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.469688892364502, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.6121, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.4836199283599854, + "learning_rate": 2.160844841592202e-06, + "loss": 1.5918, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.521851062774658, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.6228, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.845065712928772, + "eval_runtime": 45.3003, + "eval_samples_per_second": 22.075, + "eval_steps_per_second": 0.927, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.7808425426483154, + "learning_rate": 2.156783103168156e-06, + "loss": 1.5887, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.696708917617798, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.6517, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.7222893238067627, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.6817, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.504417896270752, + "learning_rate": 2.150690495532088e-06, + "loss": 1.647, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.574836015701294, + "learning_rate": 2.148659626320065e-06, + "loss": 1.6554, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.5004355907440186, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.6089, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.6979293823242188, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.6226, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.5239524841308594, + "learning_rate": 2.142567018683997e-06, + "loss": 1.6875, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.5218467712402344, + "learning_rate": 2.140536149471974e-06, + "loss": 1.6095, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.5778310298919678, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.6155, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.8443745374679565, + "eval_runtime": 46.6273, + "eval_samples_per_second": 21.447, + "eval_steps_per_second": 0.901, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.671973705291748, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.6696, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.5520615577697754, + "learning_rate": 2.134443541835906e-06, + "loss": 1.6291, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.39286208152771, + "learning_rate": 2.132412672623883e-06, + "loss": 1.6012, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.8277461528778076, + "learning_rate": 2.13038180341186e-06, + "loss": 1.6398, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.7147881984710693, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.6685, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.499044418334961, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.6128, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.514784574508667, + "learning_rate": 2.124289195775792e-06, + "loss": 1.6263, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.505342483520508, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.5532, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.608139753341675, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.5946, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.5578057765960693, + "learning_rate": 2.118196588139724e-06, + "loss": 1.6103, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.8425958156585693, + "eval_runtime": 44.206, + "eval_samples_per_second": 22.621, + "eval_steps_per_second": 0.95, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.4881014823913574, + "learning_rate": 2.116165718927701e-06, + "loss": 1.6364, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.4141550064086914, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.5943, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.649827718734741, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.5994, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.547238826751709, + "learning_rate": 2.110073111291633e-06, + "loss": 1.6247, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.6515395641326904, + "learning_rate": 2.10804224207961e-06, + "loss": 1.6543, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.5575809478759766, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.6618, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.466344118118286, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.5437, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4897878170013428, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6199, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.6578829288482666, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.5989, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.6173501014709473, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.6392, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.838721752166748, + "eval_runtime": 49.4357, + "eval_samples_per_second": 20.228, + "eval_steps_per_second": 0.85, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.5655064582824707, + "learning_rate": 2.095857026807474e-06, + "loss": 1.6276, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.685790777206421, + "learning_rate": 2.093826157595451e-06, + "loss": 1.6142, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.7278990745544434, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.6292, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.5769667625427246, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.6774, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.5120608806610107, + "learning_rate": 2.087733549959383e-06, + "loss": 1.5972, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.4658212661743164, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.5683, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.526524782180786, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.6687, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.3989250659942627, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.6184, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.589115858078003, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.6153, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.4985275268554688, + "learning_rate": 2.077579203899269e-06, + "loss": 1.5682, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.8374594449996948, + "eval_runtime": 47.0329, + "eval_samples_per_second": 21.262, + "eval_steps_per_second": 0.893, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5770835876464844, + "learning_rate": 2.075548334687246e-06, + "loss": 1.6437, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.5111560821533203, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.5711, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.6110589504241943, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.6125, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.6229705810546875, + "learning_rate": 2.069455727051178e-06, + "loss": 1.595, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.648716926574707, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.5908, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.673701763153076, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.6652, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.667478561401367, + "learning_rate": 2.06336311941511e-06, + "loss": 1.6088, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 3.5688586235046387, + "learning_rate": 2.061332250203087e-06, + "loss": 1.5852, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.512603282928467, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.6218, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.5296075344085693, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.6382, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.8406264781951904, + "eval_runtime": 43.8351, + "eval_samples_per_second": 22.813, + "eval_steps_per_second": 0.958, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.598994255065918, + "learning_rate": 2.055239642567019e-06, + "loss": 1.6263, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.93025279045105, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.6502, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 3.3655412197113037, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.5872, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.955120325088501, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.6404, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.5226638317108154, + "learning_rate": 2.047116165718928e-06, + "loss": 1.6102, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.5462093353271484, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.6757, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 4.066190242767334, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.6072, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.8548827171325684, + "learning_rate": 2.04102355808286e-06, + "loss": 1.5717, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.6206634044647217, + "learning_rate": 2.038992688870837e-06, + "loss": 1.5943, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.8793396949768066, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.6121, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.83925199508667, + "eval_runtime": 44.747, + "eval_samples_per_second": 22.348, + "eval_steps_per_second": 0.939, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.6215157508850098, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.6492, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.6103456020355225, + "learning_rate": 2.032900081234769e-06, + "loss": 1.6531, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.485775947570801, + "learning_rate": 2.030869212022746e-06, + "loss": 1.6086, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.399557113647461, + "learning_rate": 2.028838342810723e-06, + "loss": 1.6126, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.4711368083953857, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.5985, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.654580593109131, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.6557, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.5091657638549805, + "learning_rate": 2.022745735174655e-06, + "loss": 1.6181, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.5723490715026855, + "learning_rate": 2.020714865962632e-06, + "loss": 1.6421, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4211535453796387, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.6226, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.5550055503845215, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.5729, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.8397995233535767, + "eval_runtime": 44.4762, + "eval_samples_per_second": 22.484, + "eval_steps_per_second": 0.944, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.618758201599121, + "learning_rate": 2.014622258326564e-06, + "loss": 1.6356, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.6717913150787354, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.6246, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.5464835166931152, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.6027, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.557669162750244, + "learning_rate": 2.008529650690496e-06, + "loss": 1.6531, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.581970453262329, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.6076, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.609236478805542, + "learning_rate": 2.00446791226645e-06, + "loss": 1.5932, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.4249160289764404, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.6372, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.5093257427215576, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.6627, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.3989412784576416, + "learning_rate": 1.998375304630382e-06, + "loss": 1.6292, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4299519062042236, + "learning_rate": 1.996344435418359e-06, + "loss": 1.6385, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.8367400169372559, + "eval_runtime": 45.921, + "eval_samples_per_second": 21.777, + "eval_steps_per_second": 0.915, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.509530782699585, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.6441, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.591426134109497, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.6382, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.6569297313690186, + "learning_rate": 1.990251827782291e-06, + "loss": 1.6469, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.6453206539154053, + "learning_rate": 1.988220958570268e-06, + "loss": 1.6348, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.5522994995117188, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.6311, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.5981009006500244, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.585, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.617323637008667, + "learning_rate": 1.9821283509342e-06, + "loss": 1.584, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.6743686199188232, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.6334, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.418383836746216, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.5934, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.7952582836151123, + "learning_rate": 1.976035743298132e-06, + "loss": 1.6779, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.8379877805709839, + "eval_runtime": 46.545, + "eval_samples_per_second": 21.485, + "eval_steps_per_second": 0.902, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.5364768505096436, + "learning_rate": 1.974004874086109e-06, + "loss": 1.5802, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.446056365966797, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6141, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.527529239654541, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.6042, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.5953001976013184, + "learning_rate": 1.967912266450041e-06, + "loss": 1.5959, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.545994758605957, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.7091, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.3900482654571533, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.5946, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.51353120803833, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.6166, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.550389528274536, + "learning_rate": 1.95978878960195e-06, + "loss": 1.6817, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.6815762519836426, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.6594, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.591909885406494, + "learning_rate": 1.955727051177904e-06, + "loss": 1.5991, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.8390737771987915, + "eval_runtime": 46.8495, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.5111982822418213, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.5993, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.655891180038452, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.5915, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.471090316772461, + "learning_rate": 1.949634443541836e-06, + "loss": 1.5867, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.8599209785461426, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.6485, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4986093044281006, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.6084, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.6345789432525635, + "learning_rate": 1.943541835905768e-06, + "loss": 1.6719, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.5636799335479736, + "learning_rate": 1.941510966693745e-06, + "loss": 1.5817, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.5867199897766113, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.6498, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.6369481086730957, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.6791, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.4554147720336914, + "learning_rate": 1.935418359057677e-06, + "loss": 1.6319, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.8408559560775757, + "eval_runtime": 49.3515, + "eval_samples_per_second": 20.263, + "eval_steps_per_second": 0.851, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.5412509441375732, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.6673, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.5755672454833984, + "learning_rate": 1.931356620633631e-06, + "loss": 1.5636, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.602530002593994, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.6261, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.755247116088867, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.5817, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.528991460800171, + "learning_rate": 1.925264012997563e-06, + "loss": 1.6173, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.473057508468628, + "learning_rate": 1.92323314378554e-06, + "loss": 1.648, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.62630033493042, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.5864, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.7876687049865723, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.5804, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.491473436355591, + "learning_rate": 1.917140536149472e-06, + "loss": 1.5716, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.656989336013794, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.6595, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.839701771736145, + "eval_runtime": 44.5976, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.6800942420959473, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.6415, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.4725751876831055, + "learning_rate": 1.911047928513404e-06, + "loss": 1.5901, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.552624464035034, + "learning_rate": 1.909017059301381e-06, + "loss": 1.5956, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.6042184829711914, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.6328, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.5159223079681396, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.604, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.488041639328003, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.5539, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.659151077270508, + "learning_rate": 1.90089358245329e-06, + "loss": 1.5999, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.700378179550171, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.6485, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.3943607807159424, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.5736, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.8171095848083496, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6259, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.8379307985305786, + "eval_runtime": 48.6918, + "eval_samples_per_second": 20.537, + "eval_steps_per_second": 0.863, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.562655210494995, + "learning_rate": 1.892770105605199e-06, + "loss": 1.6439, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.5640337467193604, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.6113, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.5324809551239014, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.6331, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.393224000930786, + "learning_rate": 1.886677497969131e-06, + "loss": 1.5228, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 3.073945999145508, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.632, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.7066094875335693, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.5508, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.981933832168579, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6973, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.903444528579712, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.6036, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.8616178035736084, + "learning_rate": 1.876523151909017e-06, + "loss": 1.5723, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.8243541717529297, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.555, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.8437565565109253, + "eval_runtime": 48.5449, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.803142547607422, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.5861, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 3.146677017211914, + "learning_rate": 1.870430544272949e-06, + "loss": 1.6791, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 3.2554781436920166, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.5626, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.955395460128784, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.5893, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.941377878189087, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.5962, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.765620708465576, + "learning_rate": 1.862307067424858e-06, + "loss": 1.595, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.8529839515686035, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.6296, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.7770590782165527, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.5711, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.827406883239746, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.5966, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.927119016647339, + "learning_rate": 1.854183590576767e-06, + "loss": 1.6463, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.841707468032837, + "eval_runtime": 49.0965, + "eval_samples_per_second": 20.368, + "eval_steps_per_second": 0.855, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.8223392963409424, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.559, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.862473726272583, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.5189, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.9060091972351074, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.5631, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.8727498054504395, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6013, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.862298011779785, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.549, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.8221330642700195, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.6341, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.938336133956909, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.602, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.719489812850952, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.5626, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.6615874767303467, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.5535, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.835413932800293, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.5742, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.8375873565673828, + "eval_runtime": 45.731, + "eval_samples_per_second": 21.867, + "eval_steps_per_second": 0.918, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.773707389831543, + "learning_rate": 1.831844029244517e-06, + "loss": 1.5499, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 3.0959930419921875, + "learning_rate": 1.829813160032494e-06, + "loss": 1.5656, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.8521792888641357, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.5982, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.970618486404419, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.5731, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 3.0716512203216553, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.5824, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.8776299953460693, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.6283, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.72183895111084, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.5422, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.853234052658081, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.5533, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.7193424701690674, + "learning_rate": 1.815597075548335e-06, + "loss": 1.5923, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.9768450260162354, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.6277, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.8401803970336914, + "eval_runtime": 45.8062, + "eval_samples_per_second": 21.831, + "eval_steps_per_second": 0.917, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.752772569656372, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.5241, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.784980297088623, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.5657, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.7485857009887695, + "learning_rate": 1.807473598700244e-06, + "loss": 1.5501, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.7191033363342285, + "learning_rate": 1.805442729488221e-06, + "loss": 1.5718, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.697540044784546, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.5804, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.7175710201263428, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.5809, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.764824151992798, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.545, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.946669816970825, + "learning_rate": 1.79731925264013e-06, + "loss": 1.595, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.689326524734497, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.5676, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.759997606277466, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.6032, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.8379274606704712, + "eval_runtime": 45.5941, + "eval_samples_per_second": 21.933, + "eval_steps_per_second": 0.921, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.8656060695648193, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.6086, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.7155873775482178, + "learning_rate": 1.789195775792039e-06, + "loss": 1.5218, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.8487389087677, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6285, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.8070144653320312, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.5479, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.856282949447632, + "learning_rate": 1.783103168155971e-06, + "loss": 1.5689, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.7465922832489014, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.5733, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.824997901916504, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.5583, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.8030786514282227, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.5383, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.911377429962158, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.586, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.7452991008758545, + "learning_rate": 1.772948822095857e-06, + "loss": 1.6409, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.8399248123168945, + "eval_runtime": 46.1241, + "eval_samples_per_second": 21.681, + "eval_steps_per_second": 0.911, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.8255481719970703, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.6134, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.865100622177124, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.5791, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.740133047103882, + "learning_rate": 1.766856214459789e-06, + "loss": 1.5651, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.7219033241271973, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.5453, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.7266135215759277, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.6375, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.8476688861846924, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.5877, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.810105085372925, + "learning_rate": 1.758732737611698e-06, + "loss": 1.5796, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.78798246383667, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.5708, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.750525951385498, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.5567, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.818673849105835, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.5661, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.8374614715576172, + "eval_runtime": 43.5618, + "eval_samples_per_second": 22.956, + "eval_steps_per_second": 0.964, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.7967586517333984, + "learning_rate": 1.750609260763607e-06, + "loss": 1.5846, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.766272783279419, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.5721, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.8006162643432617, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.5261, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 3.714587688446045, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.6599, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 3.143052339553833, + "learning_rate": 1.742485783915516e-06, + "loss": 1.659, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.8256232738494873, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.5076, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.8843302726745605, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.5862, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 6.692182540893555, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.6144, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.928053617477417, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.571, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.8071024417877197, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.5861, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.837381362915039, + "eval_runtime": 44.7089, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 3.070399522781372, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.5636, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 3.298100709915161, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.5717, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 5.891775131225586, + "learning_rate": 1.726238830219334e-06, + "loss": 1.5757, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.838676691055298, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.6369, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.8816466331481934, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.5885, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 3.1703031063079834, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.5966, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 3.504777669906616, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.4932, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 3.1451492309570312, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.5071, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.7158446311950684, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.5725, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.8391170501708984, + "learning_rate": 1.712022745735175e-06, + "loss": 1.5742, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.8371202945709229, + "eval_runtime": 43.2464, + "eval_samples_per_second": 23.123, + "eval_steps_per_second": 0.971, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.8346052169799805, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.5808, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.767813205718994, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.5209, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.7955846786499023, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.581, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.8616271018981934, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.5758, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.7975544929504395, + "learning_rate": 1.701868399675061e-06, + "loss": 1.5912, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 3.095912456512451, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.6522, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 3.102691888809204, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.6004, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.8405165672302246, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.5649, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.8038063049316406, + "learning_rate": 1.69374492282697e-06, + "loss": 1.545, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.8661468029022217, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6297, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.8386919498443604, + "eval_runtime": 46.0334, + "eval_samples_per_second": 21.723, + "eval_steps_per_second": 0.912, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.8217391967773438, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.5694, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 3.099724769592285, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.5986, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.9951272010803223, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.5429, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.9366989135742188, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.5666, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 3.0171356201171875, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.6212, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.952259063720703, + "learning_rate": 1.679528838342811e-06, + "loss": 1.5611, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.667905330657959, + "learning_rate": 1.677497969130788e-06, + "loss": 1.5556, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.8001160621643066, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.5854, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.6954760551452637, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.5278, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.8026583194732666, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.5103, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.8384881019592285, + "eval_runtime": 46.1899, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.898869752883911, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6188, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 3.0432069301605225, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.5945, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.7744672298431396, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.5642, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.8844289779663086, + "learning_rate": 1.663281884646629e-06, + "loss": 1.5657, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 3.0022218227386475, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.6297, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.918320417404175, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.5928, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.913848876953125, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.6219, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.954702615737915, + "learning_rate": 1.655158407798538e-06, + "loss": 1.5812, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.7966420650482178, + "learning_rate": 1.653127538586515e-06, + "loss": 1.6034, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 3.1328368186950684, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.6168, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.836020827293396, + "eval_runtime": 46.1979, + "eval_samples_per_second": 21.646, + "eval_steps_per_second": 0.909, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.8109259605407715, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.6091, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.823570728302002, + "learning_rate": 1.647034930950447e-06, + "loss": 1.5504, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.885016918182373, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.5784, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 3.612942934036255, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.6014, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.8981637954711914, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.5552, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.6703410148620605, + "learning_rate": 1.638911454102356e-06, + "loss": 1.5567, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.7087886333465576, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.613, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.7711687088012695, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.5586, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.8473854064941406, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.565, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 3.783824920654297, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.5909, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.8384798765182495, + "eval_runtime": 46.2339, + "eval_samples_per_second": 21.629, + "eval_steps_per_second": 0.908, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.9971532821655273, + "learning_rate": 1.628757108042242e-06, + "loss": 1.6251, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.8581976890563965, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.5632, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.9119839668273926, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.5806, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.840101718902588, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.6706, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.8574070930480957, + "learning_rate": 1.620633631194151e-06, + "loss": 1.6138, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.820692539215088, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.5534, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7690017223358154, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.5922, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 3.062229871749878, + "learning_rate": 1.614541023558083e-06, + "loss": 1.5893, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.727062225341797, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.5851, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.7693302631378174, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.6031, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.8332058191299438, + "eval_runtime": 45.0301, + "eval_samples_per_second": 22.207, + "eval_steps_per_second": 0.933, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.803758382797241, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.5504, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.775770902633667, + "learning_rate": 1.606417546709992e-06, + "loss": 1.5707, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.7698636054992676, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.5763, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.739940881729126, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.5386, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.7594099044799805, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.5805, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.853273391723633, + "learning_rate": 1.598294069861901e-06, + "loss": 1.5548, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.9859204292297363, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6278, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.742898941040039, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.555, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.769071340560913, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.5504, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.7191832065582275, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6014, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.8336049318313599, + "eval_runtime": 47.0078, + "eval_samples_per_second": 21.273, + "eval_steps_per_second": 0.893, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.842529773712158, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.5723, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.768066883087158, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.5272, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7481656074523926, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.5843, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.806687593460083, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.5598, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.8214614391326904, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.5972, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.8541247844696045, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.5988, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 3.000246286392212, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.583, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.9164586067199707, + "learning_rate": 1.573923639317628e-06, + "loss": 1.5949, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.8099935054779053, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.5959, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.88322377204895, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.6062, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.832599401473999, + "eval_runtime": 45.9576, + "eval_samples_per_second": 21.759, + "eval_steps_per_second": 0.914, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.8019371032714844, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.5445, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.8585665225982666, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6079, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.7423956394195557, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.5537, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.7420766353607178, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.5846, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.8864388465881348, + "learning_rate": 1.559707554833469e-06, + "loss": 1.6213, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.7483463287353516, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.5012, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.7340080738067627, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.555, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.8000667095184326, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.5259, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.9176089763641357, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.5726, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.9147658348083496, + "learning_rate": 1.549553208773355e-06, + "loss": 1.5685, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.834657907485962, + "eval_runtime": 45.8709, + "eval_samples_per_second": 21.8, + "eval_steps_per_second": 0.916, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.9109392166137695, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.5533, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 3.0263078212738037, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.6277, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.9347317218780518, + "learning_rate": 1.543460601137287e-06, + "loss": 1.5738, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.8229548931121826, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.5969, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.906954288482666, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.5772, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.9214975833892822, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.5147, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.844081163406372, + "learning_rate": 1.535337124289196e-06, + "loss": 1.5635, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 3.0214555263519287, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.6466, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.9254050254821777, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.5837, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.8986024856567383, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.5997, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.8318188190460205, + "eval_runtime": 46.7695, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.974231719970703, + "learning_rate": 1.5272136474411047e-06, + "loss": 1.6143, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.7890946865081787, + "learning_rate": 1.525182778229082e-06, + "loss": 1.5956, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.820800304412842, + "learning_rate": 1.5231519090170593e-06, + "loss": 1.5626, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.84753680229187, + "learning_rate": 1.5211210398050365e-06, + "loss": 1.5473, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.829115152359009, + "learning_rate": 1.5190901705930138e-06, + "loss": 1.591, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.786787509918213, + "learning_rate": 1.517059301380991e-06, + "loss": 1.565, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.7778282165527344, + "learning_rate": 1.5150284321689684e-06, + "loss": 1.5666, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.9727330207824707, + "learning_rate": 1.5129975629569456e-06, + "loss": 1.577, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.876600980758667, + "learning_rate": 1.510966693744923e-06, + "loss": 1.5742, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.7339680194854736, + "learning_rate": 1.5089358245329002e-06, + "loss": 1.5378, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 1.834323763847351, + "eval_runtime": 44.2017, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.9201254844665527, + "learning_rate": 1.5069049553208775e-06, + "loss": 1.5962, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.8632285594940186, + "learning_rate": 1.5048740861088547e-06, + "loss": 1.589, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.9003729820251465, + "learning_rate": 1.502843216896832e-06, + "loss": 1.5125, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.81120228767395, + "learning_rate": 1.500812347684809e-06, + "loss": 1.5809, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.747528553009033, + "learning_rate": 1.4987814784727864e-06, + "loss": 1.5611, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.6891186237335205, + "learning_rate": 1.4967506092607636e-06, + "loss": 1.5387, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.9105594158172607, + "learning_rate": 1.494719740048741e-06, + "loss": 1.5921, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.8186421394348145, + "learning_rate": 1.4926888708367182e-06, + "loss": 1.6075, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 3.0198328495025635, + "learning_rate": 1.4906580016246955e-06, + "loss": 1.611, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 3.0839414596557617, + "learning_rate": 1.4886271324126727e-06, + "loss": 1.6298, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 1.8325375318527222, + "eval_runtime": 53.1022, + "eval_samples_per_second": 18.832, + "eval_steps_per_second": 0.791, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.8375303745269775, + "learning_rate": 1.48659626320065e-06, + "loss": 1.4942, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.857344627380371, + "learning_rate": 1.4845653939886273e-06, + "loss": 1.5182, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.7296979427337646, + "learning_rate": 1.4825345247766046e-06, + "loss": 1.5081, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.9989287853240967, + "learning_rate": 1.4805036555645818e-06, + "loss": 1.5577, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.9123456478118896, + "learning_rate": 1.4784727863525591e-06, + "loss": 1.5339, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.845898151397705, + "learning_rate": 1.4764419171405362e-06, + "loss": 1.6084, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 3.089202404022217, + "learning_rate": 1.4744110479285135e-06, + "loss": 1.6133, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.9908018112182617, + "learning_rate": 1.4723801787164907e-06, + "loss": 1.5984, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.838611125946045, + "learning_rate": 1.470349309504468e-06, + "loss": 1.5971, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.902926445007324, + "learning_rate": 1.4683184402924453e-06, + "loss": 1.6029, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 1.8312546014785767, + "eval_runtime": 46.3457, + "eval_samples_per_second": 21.577, + "eval_steps_per_second": 0.906, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.8279829025268555, + "learning_rate": 1.4662875710804226e-06, + "loss": 1.65, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.7380146980285645, + "learning_rate": 1.4642567018683998e-06, + "loss": 1.5673, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.723890781402588, + "learning_rate": 1.4622258326563771e-06, + "loss": 1.5573, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.8826022148132324, + "learning_rate": 1.4601949634443544e-06, + "loss": 1.5648, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.948598861694336, + "learning_rate": 1.4581640942323317e-06, + "loss": 1.6603, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.8282432556152344, + "learning_rate": 1.456133225020309e-06, + "loss": 1.5875, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.874748468399048, + "learning_rate": 1.4541023558082862e-06, + "loss": 1.5713, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.894249677658081, + "learning_rate": 1.4520714865962635e-06, + "loss": 1.6063, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.907095193862915, + "learning_rate": 1.4500406173842406e-06, + "loss": 1.5744, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.892090082168579, + "learning_rate": 1.4480097481722178e-06, + "loss": 1.5877, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 1.8305796384811401, + "eval_runtime": 48.3583, + "eval_samples_per_second": 20.679, + "eval_steps_per_second": 0.869, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.7991275787353516, + "learning_rate": 1.4459788789601951e-06, + "loss": 1.5475, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.9188716411590576, + "learning_rate": 1.4439480097481724e-06, + "loss": 1.5385, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.8575944900512695, + "learning_rate": 1.4419171405361497e-06, + "loss": 1.5855, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.814990520477295, + "learning_rate": 1.439886271324127e-06, + "loss": 1.5923, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.7953567504882812, + "learning_rate": 1.4378554021121042e-06, + "loss": 1.5724, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.882002115249634, + "learning_rate": 1.4358245329000815e-06, + "loss": 1.6086, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.910799503326416, + "learning_rate": 1.4337936636880588e-06, + "loss": 1.5933, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.911015748977661, + "learning_rate": 1.431762794476036e-06, + "loss": 1.5356, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.9788129329681396, + "learning_rate": 1.4297319252640133e-06, + "loss": 1.5546, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.8637843132019043, + "learning_rate": 1.4277010560519906e-06, + "loss": 1.5423, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 1.8307991027832031, + "eval_runtime": 45.4486, + "eval_samples_per_second": 22.003, + "eval_steps_per_second": 0.924, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 3.007390260696411, + "learning_rate": 1.4256701868399674e-06, + "loss": 1.6244, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.908496856689453, + "learning_rate": 1.4236393176279447e-06, + "loss": 1.5767, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.9650135040283203, + "learning_rate": 1.421608448415922e-06, + "loss": 1.6639, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.8085806369781494, + "learning_rate": 1.4195775792038993e-06, + "loss": 1.5486, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.7759034633636475, + "learning_rate": 1.4175467099918765e-06, + "loss": 1.5847, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.9309468269348145, + "learning_rate": 1.4155158407798538e-06, + "loss": 1.6014, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.8542673587799072, + "learning_rate": 1.413484971567831e-06, + "loss": 1.5183, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 3.0022106170654297, + "learning_rate": 1.4114541023558084e-06, + "loss": 1.5511, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.826852321624756, + "learning_rate": 1.4094232331437856e-06, + "loss": 1.6079, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.8682327270507812, + "learning_rate": 1.407392363931763e-06, + "loss": 1.6579, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 1.8299224376678467, + "eval_runtime": 44.6401, + "eval_samples_per_second": 22.401, + "eval_steps_per_second": 0.941, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.8816471099853516, + "learning_rate": 1.4053614947197402e-06, + "loss": 1.5825, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.696047306060791, + "learning_rate": 1.4033306255077175e-06, + "loss": 1.5068, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.9243481159210205, + "learning_rate": 1.4012997562956945e-06, + "loss": 1.6172, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.810744285583496, + "learning_rate": 1.3992688870836718e-06, + "loss": 1.5976, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.7787344455718994, + "learning_rate": 1.397238017871649e-06, + "loss": 1.5829, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.817336082458496, + "learning_rate": 1.3952071486596264e-06, + "loss": 1.5482, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.7691237926483154, + "learning_rate": 1.3931762794476036e-06, + "loss": 1.6098, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.7974281311035156, + "learning_rate": 1.391145410235581e-06, + "loss": 1.5591, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.8233275413513184, + "learning_rate": 1.3891145410235582e-06, + "loss": 1.4975, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.7465834617614746, + "learning_rate": 1.3870836718115355e-06, + "loss": 1.5682, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 1.8317815065383911, + "eval_runtime": 44.8326, + "eval_samples_per_second": 22.305, + "eval_steps_per_second": 0.937, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.9384610652923584, + "learning_rate": 1.3850528025995127e-06, + "loss": 1.5627, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.6946020126342773, + "learning_rate": 1.38302193338749e-06, + "loss": 1.4851, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.875818967819214, + "learning_rate": 1.3809910641754673e-06, + "loss": 1.595, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.871393918991089, + "learning_rate": 1.3789601949634446e-06, + "loss": 1.5297, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.8451645374298096, + "learning_rate": 1.3769293257514218e-06, + "loss": 1.5912, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.808422803878784, + "learning_rate": 1.374898456539399e-06, + "loss": 1.5316, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.866929292678833, + "learning_rate": 1.3728675873273762e-06, + "loss": 1.6223, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.969078302383423, + "learning_rate": 1.3708367181153535e-06, + "loss": 1.5855, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.739272117614746, + "learning_rate": 1.3688058489033307e-06, + "loss": 1.5915, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.9254791736602783, + "learning_rate": 1.366774979691308e-06, + "loss": 1.5989, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 1.8284611701965332, + "eval_runtime": 45.0516, + "eval_samples_per_second": 22.197, + "eval_steps_per_second": 0.932, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.7639904022216797, + "learning_rate": 1.3647441104792853e-06, + "loss": 1.5735, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.782290458679199, + "learning_rate": 1.3627132412672626e-06, + "loss": 1.5365, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.8718478679656982, + "learning_rate": 1.3606823720552398e-06, + "loss": 1.5776, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.770900249481201, + "learning_rate": 1.3586515028432171e-06, + "loss": 1.5242, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.7527379989624023, + "learning_rate": 1.3566206336311944e-06, + "loss": 1.5783, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.724205732345581, + "learning_rate": 1.3545897644191717e-06, + "loss": 1.5556, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.7595510482788086, + "learning_rate": 1.352558895207149e-06, + "loss": 1.6084, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.832383394241333, + "learning_rate": 1.3505280259951258e-06, + "loss": 1.5755, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.8154635429382324, + "learning_rate": 1.348497156783103e-06, + "loss": 1.6091, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.808480739593506, + "learning_rate": 1.3464662875710804e-06, + "loss": 1.5403, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 1.8285717964172363, + "eval_runtime": 44.0301, + "eval_samples_per_second": 22.712, + "eval_steps_per_second": 0.954, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.7877376079559326, + "learning_rate": 1.3444354183590576e-06, + "loss": 1.5985, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.854957103729248, + "learning_rate": 1.342404549147035e-06, + "loss": 1.5576, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.8522772789001465, + "learning_rate": 1.3403736799350122e-06, + "loss": 1.5914, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.866199493408203, + "learning_rate": 1.3383428107229895e-06, + "loss": 1.6094, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.9277379512786865, + "learning_rate": 1.3363119415109667e-06, + "loss": 1.6065, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.8854689598083496, + "learning_rate": 1.334281072298944e-06, + "loss": 1.5892, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.7135169506073, + "learning_rate": 1.3322502030869213e-06, + "loss": 1.5733, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.8702995777130127, + "learning_rate": 1.3302193338748986e-06, + "loss": 1.6054, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.8330469131469727, + "learning_rate": 1.3281884646628758e-06, + "loss": 1.5902, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 3.016582727432251, + "learning_rate": 1.3261575954508531e-06, + "loss": 1.5931, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 1.82985520362854, + "eval_runtime": 45.8601, + "eval_samples_per_second": 21.805, + "eval_steps_per_second": 0.916, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.8396756649017334, + "learning_rate": 1.3241267262388302e-06, + "loss": 1.5882, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.7104387283325195, + "learning_rate": 1.3220958570268074e-06, + "loss": 1.5223, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.787862777709961, + "learning_rate": 1.3200649878147847e-06, + "loss": 1.5549, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.9957401752471924, + "learning_rate": 1.318034118602762e-06, + "loss": 1.6036, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.916990280151367, + "learning_rate": 1.3160032493907393e-06, + "loss": 1.5947, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.972477912902832, + "learning_rate": 1.3139723801787166e-06, + "loss": 1.624, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.84188175201416, + "learning_rate": 1.3119415109666938e-06, + "loss": 1.5708, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.908295154571533, + "learning_rate": 1.309910641754671e-06, + "loss": 1.5412, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.8915605545043945, + "learning_rate": 1.3078797725426484e-06, + "loss": 1.5463, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.7805707454681396, + "learning_rate": 1.3058489033306257e-06, + "loss": 1.5555, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 1.8281073570251465, + "eval_runtime": 44.7855, + "eval_samples_per_second": 22.329, + "eval_steps_per_second": 0.938, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.933260679244995, + "learning_rate": 1.303818034118603e-06, + "loss": 1.578, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.970050096511841, + "learning_rate": 1.3017871649065802e-06, + "loss": 1.5606, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.717027187347412, + "learning_rate": 1.2997562956945573e-06, + "loss": 1.5446, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.966905355453491, + "learning_rate": 1.2977254264825345e-06, + "loss": 1.549, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.847358465194702, + "learning_rate": 1.2956945572705118e-06, + "loss": 1.581, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.814897060394287, + "learning_rate": 1.293663688058489e-06, + "loss": 1.5416, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.788320302963257, + "learning_rate": 1.2916328188464664e-06, + "loss": 1.5018, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.944441556930542, + "learning_rate": 1.2896019496344436e-06, + "loss": 1.5318, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.8705525398254395, + "learning_rate": 1.287571080422421e-06, + "loss": 1.5953, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.7703897953033447, + "learning_rate": 1.2855402112103982e-06, + "loss": 1.5287, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 1.8304927349090576, + "eval_runtime": 47.1736, + "eval_samples_per_second": 21.198, + "eval_steps_per_second": 0.89, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.862663745880127, + "learning_rate": 1.2835093419983755e-06, + "loss": 1.541, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.8288817405700684, + "learning_rate": 1.2814784727863528e-06, + "loss": 1.5466, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 3.0720043182373047, + "learning_rate": 1.27944760357433e-06, + "loss": 1.6106, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.9464635848999023, + "learning_rate": 1.2774167343623073e-06, + "loss": 1.6051, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.9039576053619385, + "learning_rate": 1.2753858651502846e-06, + "loss": 1.5525, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.8839735984802246, + "learning_rate": 1.2733549959382616e-06, + "loss": 1.5551, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.915053129196167, + "learning_rate": 1.271324126726239e-06, + "loss": 1.5853, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.862844467163086, + "learning_rate": 1.2692932575142162e-06, + "loss": 1.5233, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.717557430267334, + "learning_rate": 1.2672623883021935e-06, + "loss": 1.5138, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 3.1381704807281494, + "learning_rate": 1.2652315190901707e-06, + "loss": 1.6073, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 1.82700514793396, + "eval_runtime": 51.3349, + "eval_samples_per_second": 19.48, + "eval_steps_per_second": 0.818, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.8603286743164062, + "learning_rate": 1.263200649878148e-06, + "loss": 1.5713, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.9590084552764893, + "learning_rate": 1.2611697806661253e-06, + "loss": 1.5952, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.8419439792633057, + "learning_rate": 1.2591389114541026e-06, + "loss": 1.5729, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.7028636932373047, + "learning_rate": 1.2571080422420798e-06, + "loss": 1.5427, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.819112539291382, + "learning_rate": 1.2550771730300571e-06, + "loss": 1.6221, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.821469783782959, + "learning_rate": 1.2530463038180344e-06, + "loss": 1.5756, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.979747772216797, + "learning_rate": 1.2510154346060117e-06, + "loss": 1.5662, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 3.176698684692383, + "learning_rate": 1.2489845653939887e-06, + "loss": 1.591, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.8962442874908447, + "learning_rate": 1.246953696181966e-06, + "loss": 1.6549, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.840876817703247, + "learning_rate": 1.2449228269699433e-06, + "loss": 1.5636, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 1.8279746770858765, + "eval_runtime": 45.3966, + "eval_samples_per_second": 22.028, + "eval_steps_per_second": 0.925, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.8487303256988525, + "learning_rate": 1.2428919577579204e-06, + "loss": 1.5345, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.8009743690490723, + "learning_rate": 1.2408610885458976e-06, + "loss": 1.5673, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.7863199710845947, + "learning_rate": 1.238830219333875e-06, + "loss": 1.5351, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 3.4834156036376953, + "learning_rate": 1.2367993501218522e-06, + "loss": 1.6171, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.9101381301879883, + "learning_rate": 1.2347684809098295e-06, + "loss": 1.6026, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.750434160232544, + "learning_rate": 1.2327376116978067e-06, + "loss": 1.5199, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.8388402462005615, + "learning_rate": 1.230706742485784e-06, + "loss": 1.5253, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 3.000380754470825, + "learning_rate": 1.2286758732737613e-06, + "loss": 1.6463, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 3.0284345149993896, + "learning_rate": 1.2266450040617386e-06, + "loss": 1.637, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.9016034603118896, + "learning_rate": 1.2246141348497158e-06, + "loss": 1.5794, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 1.8288570642471313, + "eval_runtime": 47.0365, + "eval_samples_per_second": 21.26, + "eval_steps_per_second": 0.893, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 3.27822208404541, + "learning_rate": 1.2225832656376931e-06, + "loss": 1.5734, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.8195855617523193, + "learning_rate": 1.2205523964256704e-06, + "loss": 1.5512, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.7829315662384033, + "learning_rate": 1.2185215272136477e-06, + "loss": 1.5804, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.8760147094726562, + "learning_rate": 1.2164906580016247e-06, + "loss": 1.5858, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.8194808959960938, + "learning_rate": 1.214459788789602e-06, + "loss": 1.485, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.901355266571045, + "learning_rate": 1.2124289195775793e-06, + "loss": 1.5287, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 3.0580921173095703, + "learning_rate": 1.2103980503655566e-06, + "loss": 1.5596, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 3.0313403606414795, + "learning_rate": 1.2083671811535338e-06, + "loss": 1.5536, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.7663581371307373, + "learning_rate": 1.2063363119415111e-06, + "loss": 1.5479, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.7729651927948, + "learning_rate": 1.2043054427294882e-06, + "loss": 1.56, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 1.8264360427856445, + "eval_runtime": 45.8995, + "eval_samples_per_second": 21.787, + "eval_steps_per_second": 0.915, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 3.0862157344818115, + "learning_rate": 1.2022745735174654e-06, + "loss": 1.5869, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.9891650676727295, + "learning_rate": 1.2002437043054427e-06, + "loss": 1.6249, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.8348777294158936, + "learning_rate": 1.19821283509342e-06, + "loss": 1.6116, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.915382146835327, + "learning_rate": 1.1961819658813973e-06, + "loss": 1.5783, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 3.086989402770996, + "learning_rate": 1.1941510966693746e-06, + "loss": 1.5473, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.8165829181671143, + "learning_rate": 1.1921202274573518e-06, + "loss": 1.5617, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 3.121267795562744, + "learning_rate": 1.190089358245329e-06, + "loss": 1.6125, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.889802932739258, + "learning_rate": 1.1880584890333064e-06, + "loss": 1.5637, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.8893845081329346, + "learning_rate": 1.1860276198212837e-06, + "loss": 1.5929, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.900737762451172, + "learning_rate": 1.183996750609261e-06, + "loss": 1.5798, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 1.826514720916748, + "eval_runtime": 46.0672, + "eval_samples_per_second": 21.707, + "eval_steps_per_second": 0.912, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.8097946643829346, + "learning_rate": 1.1819658813972382e-06, + "loss": 1.5653, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.82616925239563, + "learning_rate": 1.1799350121852153e-06, + "loss": 1.612, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 3.3667945861816406, + "learning_rate": 1.1779041429731925e-06, + "loss": 1.6107, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.9167160987854004, + "learning_rate": 1.1758732737611698e-06, + "loss": 1.5424, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.9599924087524414, + "learning_rate": 1.173842404549147e-06, + "loss": 1.6195, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.8982062339782715, + "learning_rate": 1.1718115353371244e-06, + "loss": 1.5579, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.915224075317383, + "learning_rate": 1.1697806661251016e-06, + "loss": 1.6204, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.9213180541992188, + "learning_rate": 1.167749796913079e-06, + "loss": 1.5988, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.9090816974639893, + "learning_rate": 1.1657189277010562e-06, + "loss": 1.5773, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.8555307388305664, + "learning_rate": 1.1636880584890335e-06, + "loss": 1.5687, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 1.8246150016784668, + "eval_runtime": 45.9157, + "eval_samples_per_second": 21.779, + "eval_steps_per_second": 0.915, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.8956985473632812, + "learning_rate": 1.1616571892770108e-06, + "loss": 1.5878, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 4.177687644958496, + "learning_rate": 1.159626320064988e-06, + "loss": 1.579, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.9070162773132324, + "learning_rate": 1.1575954508529653e-06, + "loss": 1.5926, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.7478721141815186, + "learning_rate": 1.1555645816409426e-06, + "loss": 1.5898, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.6728436946868896, + "learning_rate": 1.1535337124289196e-06, + "loss": 1.5345, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.809030532836914, + "learning_rate": 1.151502843216897e-06, + "loss": 1.5174, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 3.078012466430664, + "learning_rate": 1.1494719740048742e-06, + "loss": 1.5914, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 4.21104621887207, + "learning_rate": 1.1474411047928515e-06, + "loss": 1.5694, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 3.0975046157836914, + "learning_rate": 1.1454102355808287e-06, + "loss": 1.545, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.925477981567383, + "learning_rate": 1.143379366368806e-06, + "loss": 1.5396, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 1.8294752836227417, + "eval_runtime": 47.0872, + "eval_samples_per_second": 21.237, + "eval_steps_per_second": 0.892, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 3.391608238220215, + "learning_rate": 1.141348497156783e-06, + "loss": 1.6501, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 3.053389549255371, + "learning_rate": 1.1393176279447604e-06, + "loss": 1.6376, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.8790252208709717, + "learning_rate": 1.1372867587327376e-06, + "loss": 1.5002, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 3.0458412170410156, + "learning_rate": 1.135255889520715e-06, + "loss": 1.5955, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.8105313777923584, + "learning_rate": 1.1332250203086922e-06, + "loss": 1.5849, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.893143892288208, + "learning_rate": 1.1311941510966695e-06, + "loss": 1.5554, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.86181640625, + "learning_rate": 1.1291632818846467e-06, + "loss": 1.5409, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.749263286590576, + "learning_rate": 1.127132412672624e-06, + "loss": 1.544, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.759432554244995, + "learning_rate": 1.1251015434606013e-06, + "loss": 1.5787, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.8302371501922607, + "learning_rate": 1.1230706742485786e-06, + "loss": 1.6204, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 1.8242546319961548, + "eval_runtime": 45.7883, + "eval_samples_per_second": 21.84, + "eval_steps_per_second": 0.917, + "step": 2200 + }, + { + "epoch": 2.412054794520548, + "grad_norm": 2.8081696033477783, + "learning_rate": 1.1210398050365558e-06, + "loss": 1.5497, + "step": 2201 + }, + { + "epoch": 2.413150684931507, + "grad_norm": 2.777566432952881, + "learning_rate": 1.1190089358245331e-06, + "loss": 1.5249, + "step": 2202 + }, + { + "epoch": 2.4142465753424656, + "grad_norm": 2.851879835128784, + "learning_rate": 1.1169780666125102e-06, + "loss": 1.5885, + "step": 2203 + }, + { + "epoch": 2.4153424657534246, + "grad_norm": 2.859072685241699, + "learning_rate": 1.1149471974004875e-06, + "loss": 1.5438, + "step": 2204 + }, + { + "epoch": 2.4164383561643836, + "grad_norm": 2.908538818359375, + "learning_rate": 1.1129163281884647e-06, + "loss": 1.6276, + "step": 2205 + }, + { + "epoch": 2.417534246575342, + "grad_norm": 2.7648227214813232, + "learning_rate": 1.110885458976442e-06, + "loss": 1.5367, + "step": 2206 + }, + { + "epoch": 2.4186301369863012, + "grad_norm": 2.9282727241516113, + "learning_rate": 1.1088545897644193e-06, + "loss": 1.6055, + "step": 2207 + }, + { + "epoch": 2.4197260273972603, + "grad_norm": 2.7551443576812744, + "learning_rate": 1.1068237205523966e-06, + "loss": 1.5984, + "step": 2208 + }, + { + "epoch": 2.4208219178082193, + "grad_norm": 2.88720440864563, + "learning_rate": 1.1047928513403736e-06, + "loss": 1.5988, + "step": 2209 + }, + { + "epoch": 2.421917808219178, + "grad_norm": 2.75961971282959, + "learning_rate": 1.102761982128351e-06, + "loss": 1.5291, + "step": 2210 + }, + { + "epoch": 2.421917808219178, + "eval_loss": 1.8230210542678833, + "eval_runtime": 45.9261, + "eval_samples_per_second": 21.774, + "eval_steps_per_second": 0.915, + "step": 2210 + }, + { + "epoch": 2.423013698630137, + "grad_norm": 2.7809903621673584, + "learning_rate": 1.1007311129163282e-06, + "loss": 1.5563, + "step": 2211 + }, + { + "epoch": 2.424109589041096, + "grad_norm": 2.7741317749023438, + "learning_rate": 1.0987002437043055e-06, + "loss": 1.5594, + "step": 2212 + }, + { + "epoch": 2.425205479452055, + "grad_norm": 2.840890884399414, + "learning_rate": 1.0966693744922827e-06, + "loss": 1.52, + "step": 2213 + }, + { + "epoch": 2.4263013698630136, + "grad_norm": 2.9647538661956787, + "learning_rate": 1.09463850528026e-06, + "loss": 1.5447, + "step": 2214 + }, + { + "epoch": 2.4273972602739726, + "grad_norm": 2.9884531497955322, + "learning_rate": 1.0926076360682373e-06, + "loss": 1.5426, + "step": 2215 + }, + { + "epoch": 2.4284931506849317, + "grad_norm": 2.8884270191192627, + "learning_rate": 1.0905767668562146e-06, + "loss": 1.5546, + "step": 2216 + }, + { + "epoch": 2.4295890410958902, + "grad_norm": 2.97591495513916, + "learning_rate": 1.0885458976441918e-06, + "loss": 1.632, + "step": 2217 + }, + { + "epoch": 2.4306849315068493, + "grad_norm": 2.944126605987549, + "learning_rate": 1.0865150284321691e-06, + "loss": 1.5845, + "step": 2218 + }, + { + "epoch": 2.4317808219178083, + "grad_norm": 2.9660258293151855, + "learning_rate": 1.0844841592201464e-06, + "loss": 1.5939, + "step": 2219 + }, + { + "epoch": 2.432876712328767, + "grad_norm": 2.8848304748535156, + "learning_rate": 1.0824532900081237e-06, + "loss": 1.5848, + "step": 2220 + }, + { + "epoch": 2.432876712328767, + "eval_loss": 1.8266901969909668, + "eval_runtime": 51.9703, + "eval_samples_per_second": 19.242, + "eval_steps_per_second": 0.808, + "step": 2220 + }, + { + "epoch": 2.433972602739726, + "grad_norm": 2.9656991958618164, + "learning_rate": 1.080422420796101e-06, + "loss": 1.568, + "step": 2221 + }, + { + "epoch": 2.435068493150685, + "grad_norm": 2.9461898803710938, + "learning_rate": 1.078391551584078e-06, + "loss": 1.5922, + "step": 2222 + }, + { + "epoch": 2.436164383561644, + "grad_norm": 2.937289237976074, + "learning_rate": 1.0763606823720553e-06, + "loss": 1.6414, + "step": 2223 + }, + { + "epoch": 2.4372602739726026, + "grad_norm": 2.8320538997650146, + "learning_rate": 1.0743298131600326e-06, + "loss": 1.5995, + "step": 2224 + }, + { + "epoch": 2.4383561643835616, + "grad_norm": 2.8221585750579834, + "learning_rate": 1.0722989439480098e-06, + "loss": 1.549, + "step": 2225 + }, + { + "epoch": 2.4394520547945207, + "grad_norm": 2.854454517364502, + "learning_rate": 1.070268074735987e-06, + "loss": 1.5474, + "step": 2226 + }, + { + "epoch": 2.4405479452054797, + "grad_norm": 2.7813878059387207, + "learning_rate": 1.0682372055239644e-06, + "loss": 1.5638, + "step": 2227 + }, + { + "epoch": 2.4416438356164383, + "grad_norm": 2.985661029815674, + "learning_rate": 1.0662063363119414e-06, + "loss": 1.5457, + "step": 2228 + }, + { + "epoch": 2.4427397260273973, + "grad_norm": 2.863631010055542, + "learning_rate": 1.0641754670999187e-06, + "loss": 1.5824, + "step": 2229 + }, + { + "epoch": 2.4438356164383563, + "grad_norm": 2.8328969478607178, + "learning_rate": 1.062144597887896e-06, + "loss": 1.5708, + "step": 2230 + }, + { + "epoch": 2.4438356164383563, + "eval_loss": 1.8261159658432007, + "eval_runtime": 50.0095, + "eval_samples_per_second": 19.996, + "eval_steps_per_second": 0.84, + "step": 2230 + }, + { + "epoch": 2.444931506849315, + "grad_norm": 2.9791696071624756, + "learning_rate": 1.0601137286758733e-06, + "loss": 1.6242, + "step": 2231 + }, + { + "epoch": 2.446027397260274, + "grad_norm": 2.7748308181762695, + "learning_rate": 1.0580828594638505e-06, + "loss": 1.499, + "step": 2232 + }, + { + "epoch": 2.447123287671233, + "grad_norm": 2.8511860370635986, + "learning_rate": 1.0560519902518278e-06, + "loss": 1.5787, + "step": 2233 + }, + { + "epoch": 2.4482191780821916, + "grad_norm": 2.802671194076538, + "learning_rate": 1.054021121039805e-06, + "loss": 1.4995, + "step": 2234 + }, + { + "epoch": 2.4493150684931506, + "grad_norm": 2.8005664348602295, + "learning_rate": 1.0519902518277824e-06, + "loss": 1.5083, + "step": 2235 + }, + { + "epoch": 2.4504109589041096, + "grad_norm": 2.92136287689209, + "learning_rate": 1.0499593826157596e-06, + "loss": 1.5807, + "step": 2236 + }, + { + "epoch": 2.4515068493150687, + "grad_norm": 2.850313186645508, + "learning_rate": 1.047928513403737e-06, + "loss": 1.5592, + "step": 2237 + }, + { + "epoch": 2.4526027397260273, + "grad_norm": 2.745370864868164, + "learning_rate": 1.0458976441917142e-06, + "loss": 1.495, + "step": 2238 + }, + { + "epoch": 2.4536986301369863, + "grad_norm": 2.794156789779663, + "learning_rate": 1.0438667749796915e-06, + "loss": 1.5024, + "step": 2239 + }, + { + "epoch": 2.4547945205479453, + "grad_norm": 2.8183324337005615, + "learning_rate": 1.0418359057676688e-06, + "loss": 1.5461, + "step": 2240 + }, + { + "epoch": 2.4547945205479453, + "eval_loss": 1.8265395164489746, + "eval_runtime": 47.1962, + "eval_samples_per_second": 21.188, + "eval_steps_per_second": 0.89, + "step": 2240 + }, + { + "epoch": 2.455890410958904, + "grad_norm": 2.863630771636963, + "learning_rate": 1.0398050365556458e-06, + "loss": 1.5977, + "step": 2241 + }, + { + "epoch": 2.456986301369863, + "grad_norm": 2.7688169479370117, + "learning_rate": 1.037774167343623e-06, + "loss": 1.5399, + "step": 2242 + }, + { + "epoch": 2.458082191780822, + "grad_norm": 2.8812880516052246, + "learning_rate": 1.0357432981316004e-06, + "loss": 1.5722, + "step": 2243 + }, + { + "epoch": 2.459178082191781, + "grad_norm": 3.2114691734313965, + "learning_rate": 1.0337124289195776e-06, + "loss": 1.6012, + "step": 2244 + }, + { + "epoch": 2.4602739726027396, + "grad_norm": 2.961271047592163, + "learning_rate": 1.031681559707555e-06, + "loss": 1.606, + "step": 2245 + }, + { + "epoch": 2.4613698630136986, + "grad_norm": 2.835495710372925, + "learning_rate": 1.0296506904955322e-06, + "loss": 1.5518, + "step": 2246 + }, + { + "epoch": 2.4624657534246577, + "grad_norm": 2.8796403408050537, + "learning_rate": 1.0276198212835095e-06, + "loss": 1.573, + "step": 2247 + }, + { + "epoch": 2.4635616438356163, + "grad_norm": 2.774409770965576, + "learning_rate": 1.0255889520714867e-06, + "loss": 1.5485, + "step": 2248 + }, + { + "epoch": 2.4646575342465753, + "grad_norm": 2.8847763538360596, + "learning_rate": 1.023558082859464e-06, + "loss": 1.5168, + "step": 2249 + }, + { + "epoch": 2.4657534246575343, + "grad_norm": 2.9533803462982178, + "learning_rate": 1.0215272136474413e-06, + "loss": 1.6179, + "step": 2250 + }, + { + "epoch": 2.4657534246575343, + "eval_loss": 1.8261420726776123, + "eval_runtime": 46.1356, + "eval_samples_per_second": 21.675, + "eval_steps_per_second": 0.91, + "step": 2250 + }, + { + "epoch": 2.466849315068493, + "grad_norm": 2.8762729167938232, + "learning_rate": 1.0194963444354186e-06, + "loss": 1.5109, + "step": 2251 + }, + { + "epoch": 2.467945205479452, + "grad_norm": 2.9228789806365967, + "learning_rate": 1.0174654752233958e-06, + "loss": 1.5861, + "step": 2252 + }, + { + "epoch": 2.469041095890411, + "grad_norm": 2.9467368125915527, + "learning_rate": 1.015434606011373e-06, + "loss": 1.5533, + "step": 2253 + }, + { + "epoch": 2.47013698630137, + "grad_norm": 2.9693222045898438, + "learning_rate": 1.0134037367993502e-06, + "loss": 1.5741, + "step": 2254 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 2.933368682861328, + "learning_rate": 1.0113728675873275e-06, + "loss": 1.5555, + "step": 2255 + }, + { + "epoch": 2.4723287671232876, + "grad_norm": 2.9675047397613525, + "learning_rate": 1.0093419983753047e-06, + "loss": 1.5803, + "step": 2256 + }, + { + "epoch": 2.4734246575342467, + "grad_norm": 2.931260108947754, + "learning_rate": 1.007311129163282e-06, + "loss": 1.5263, + "step": 2257 + }, + { + "epoch": 2.4745205479452057, + "grad_norm": 2.8160653114318848, + "learning_rate": 1.0052802599512593e-06, + "loss": 1.5281, + "step": 2258 + }, + { + "epoch": 2.4756164383561643, + "grad_norm": 3.0168352127075195, + "learning_rate": 1.0032493907392364e-06, + "loss": 1.5746, + "step": 2259 + }, + { + "epoch": 2.4767123287671233, + "grad_norm": 2.7623541355133057, + "learning_rate": 1.0012185215272136e-06, + "loss": 1.5181, + "step": 2260 + }, + { + "epoch": 2.4767123287671233, + "eval_loss": 1.8249253034591675, + "eval_runtime": 47.2027, + "eval_samples_per_second": 21.185, + "eval_steps_per_second": 0.89, + "step": 2260 + }, + { + "epoch": 2.4778082191780824, + "grad_norm": 2.8098902702331543, + "learning_rate": 9.99187652315191e-07, + "loss": 1.6099, + "step": 2261 + }, + { + "epoch": 2.478904109589041, + "grad_norm": 2.8562352657318115, + "learning_rate": 9.971567831031682e-07, + "loss": 1.5795, + "step": 2262 + }, + { + "epoch": 2.48, + "grad_norm": 2.7867422103881836, + "learning_rate": 9.951259138911455e-07, + "loss": 1.5104, + "step": 2263 + }, + { + "epoch": 2.481095890410959, + "grad_norm": 2.9658195972442627, + "learning_rate": 9.930950446791227e-07, + "loss": 1.5891, + "step": 2264 + }, + { + "epoch": 2.4821917808219176, + "grad_norm": 2.824355363845825, + "learning_rate": 9.910641754671e-07, + "loss": 1.6385, + "step": 2265 + }, + { + "epoch": 2.4832876712328766, + "grad_norm": 2.833772897720337, + "learning_rate": 9.890333062550773e-07, + "loss": 1.5759, + "step": 2266 + }, + { + "epoch": 2.4843835616438357, + "grad_norm": 2.9420840740203857, + "learning_rate": 9.870024370430546e-07, + "loss": 1.573, + "step": 2267 + }, + { + "epoch": 2.4854794520547947, + "grad_norm": 2.84629225730896, + "learning_rate": 9.849715678310318e-07, + "loss": 1.5606, + "step": 2268 + }, + { + "epoch": 2.4865753424657533, + "grad_norm": 2.7983481884002686, + "learning_rate": 9.829406986190091e-07, + "loss": 1.5718, + "step": 2269 + }, + { + "epoch": 2.4876712328767123, + "grad_norm": 2.8418283462524414, + "learning_rate": 9.809098294069864e-07, + "loss": 1.5052, + "step": 2270 + }, + { + "epoch": 2.4876712328767123, + "eval_loss": 1.8237050771713257, + "eval_runtime": 53.2157, + "eval_samples_per_second": 18.791, + "eval_steps_per_second": 0.789, + "step": 2270 + }, + { + "epoch": 2.4887671232876714, + "grad_norm": 2.8386685848236084, + "learning_rate": 9.788789601949637e-07, + "loss": 1.5638, + "step": 2271 + }, + { + "epoch": 2.48986301369863, + "grad_norm": 2.8796494007110596, + "learning_rate": 9.768480909829407e-07, + "loss": 1.6258, + "step": 2272 + }, + { + "epoch": 2.490958904109589, + "grad_norm": 2.8450164794921875, + "learning_rate": 9.74817221770918e-07, + "loss": 1.5567, + "step": 2273 + }, + { + "epoch": 2.492054794520548, + "grad_norm": 2.8501436710357666, + "learning_rate": 9.727863525588953e-07, + "loss": 1.5731, + "step": 2274 + }, + { + "epoch": 2.493150684931507, + "grad_norm": 2.8499205112457275, + "learning_rate": 9.707554833468726e-07, + "loss": 1.5383, + "step": 2275 + }, + { + "epoch": 2.4942465753424656, + "grad_norm": 2.904446601867676, + "learning_rate": 9.687246141348498e-07, + "loss": 1.5423, + "step": 2276 + }, + { + "epoch": 2.4953424657534247, + "grad_norm": 2.7885360717773438, + "learning_rate": 9.666937449228271e-07, + "loss": 1.5583, + "step": 2277 + }, + { + "epoch": 2.4964383561643837, + "grad_norm": 2.8230857849121094, + "learning_rate": 9.646628757108042e-07, + "loss": 1.531, + "step": 2278 + }, + { + "epoch": 2.4975342465753423, + "grad_norm": 2.9093189239501953, + "learning_rate": 9.626320064987814e-07, + "loss": 1.5901, + "step": 2279 + }, + { + "epoch": 2.4986301369863013, + "grad_norm": 2.853430986404419, + "learning_rate": 9.606011372867587e-07, + "loss": 1.5425, + "step": 2280 + }, + { + "epoch": 2.4986301369863013, + "eval_loss": 1.824331283569336, + "eval_runtime": 51.6196, + "eval_samples_per_second": 19.372, + "eval_steps_per_second": 0.814, + "step": 2280 + }, + { + "epoch": 2.4997260273972604, + "grad_norm": 2.857558488845825, + "learning_rate": 9.58570268074736e-07, + "loss": 1.555, + "step": 2281 + }, + { + "epoch": 2.500821917808219, + "grad_norm": 2.758384943008423, + "learning_rate": 9.565393988627133e-07, + "loss": 1.5546, + "step": 2282 + }, + { + "epoch": 2.501917808219178, + "grad_norm": 2.8295326232910156, + "learning_rate": 9.545085296506906e-07, + "loss": 1.5842, + "step": 2283 + }, + { + "epoch": 2.503013698630137, + "grad_norm": 2.855534315109253, + "learning_rate": 9.524776604386677e-07, + "loss": 1.5636, + "step": 2284 + }, + { + "epoch": 2.504109589041096, + "grad_norm": 2.7737905979156494, + "learning_rate": 9.50446791226645e-07, + "loss": 1.5101, + "step": 2285 + }, + { + "epoch": 2.5052054794520546, + "grad_norm": 2.7556989192962646, + "learning_rate": 9.484159220146223e-07, + "loss": 1.6216, + "step": 2286 + }, + { + "epoch": 2.5063013698630137, + "grad_norm": 2.8598365783691406, + "learning_rate": 9.463850528025995e-07, + "loss": 1.5818, + "step": 2287 + }, + { + "epoch": 2.5073972602739727, + "grad_norm": 3.0465996265411377, + "learning_rate": 9.443541835905768e-07, + "loss": 1.6237, + "step": 2288 + }, + { + "epoch": 2.5084931506849317, + "grad_norm": 3.067974328994751, + "learning_rate": 9.423233143785541e-07, + "loss": 1.6346, + "step": 2289 + }, + { + "epoch": 2.5095890410958903, + "grad_norm": 2.9208505153656006, + "learning_rate": 9.402924451665313e-07, + "loss": 1.543, + "step": 2290 + }, + { + "epoch": 2.5095890410958903, + "eval_loss": 1.8239408731460571, + "eval_runtime": 52.1634, + "eval_samples_per_second": 19.171, + "eval_steps_per_second": 0.805, + "step": 2290 + }, + { + "epoch": 2.5106849315068493, + "grad_norm": 3.0587332248687744, + "learning_rate": 9.382615759545085e-07, + "loss": 1.5814, + "step": 2291 + }, + { + "epoch": 2.5117808219178084, + "grad_norm": 2.9887471199035645, + "learning_rate": 9.362307067424858e-07, + "loss": 1.5481, + "step": 2292 + }, + { + "epoch": 2.512876712328767, + "grad_norm": 3.127073049545288, + "learning_rate": 9.341998375304631e-07, + "loss": 1.6121, + "step": 2293 + }, + { + "epoch": 2.513972602739726, + "grad_norm": 2.7613704204559326, + "learning_rate": 9.321689683184404e-07, + "loss": 1.5337, + "step": 2294 + }, + { + "epoch": 2.515068493150685, + "grad_norm": 2.9512479305267334, + "learning_rate": 9.301380991064176e-07, + "loss": 1.5161, + "step": 2295 + }, + { + "epoch": 2.5161643835616436, + "grad_norm": 2.9375245571136475, + "learning_rate": 9.281072298943949e-07, + "loss": 1.5538, + "step": 2296 + }, + { + "epoch": 2.5172602739726027, + "grad_norm": 3.8778791427612305, + "learning_rate": 9.260763606823721e-07, + "loss": 1.5736, + "step": 2297 + }, + { + "epoch": 2.5183561643835617, + "grad_norm": 3.176246404647827, + "learning_rate": 9.240454914703494e-07, + "loss": 1.5855, + "step": 2298 + }, + { + "epoch": 2.5194520547945203, + "grad_norm": 2.826871395111084, + "learning_rate": 9.220146222583266e-07, + "loss": 1.526, + "step": 2299 + }, + { + "epoch": 2.5205479452054793, + "grad_norm": 2.7983620166778564, + "learning_rate": 9.199837530463039e-07, + "loss": 1.5656, + "step": 2300 + }, + { + "epoch": 2.5205479452054793, + "eval_loss": 1.8233634233474731, + "eval_runtime": 51.1613, + "eval_samples_per_second": 19.546, + "eval_steps_per_second": 0.821, + "step": 2300 + }, + { + "epoch": 2.5216438356164383, + "grad_norm": 2.8449342250823975, + "learning_rate": 9.179528838342812e-07, + "loss": 1.5574, + "step": 2301 + }, + { + "epoch": 2.5227397260273974, + "grad_norm": 2.8809049129486084, + "learning_rate": 9.159220146222585e-07, + "loss": 1.5828, + "step": 2302 + }, + { + "epoch": 2.5238356164383564, + "grad_norm": 2.7337660789489746, + "learning_rate": 9.138911454102356e-07, + "loss": 1.5006, + "step": 2303 + }, + { + "epoch": 2.524931506849315, + "grad_norm": 2.7337660789489746, + "learning_rate": 9.138911454102356e-07, + "loss": 1.5592, + "step": 2304 + }, + { + "epoch": 2.526027397260274, + "grad_norm": 3.1104655265808105, + "learning_rate": 9.118602761982129e-07, + "loss": 1.5642, + "step": 2305 + }, + { + "epoch": 2.527123287671233, + "grad_norm": 4.62161922454834, + "learning_rate": 9.098294069861902e-07, + "loss": 1.5333, + "step": 2306 + }, + { + "epoch": 2.5282191780821917, + "grad_norm": 2.879892110824585, + "learning_rate": 9.077985377741675e-07, + "loss": 1.504, + "step": 2307 + }, + { + "epoch": 2.5293150684931507, + "grad_norm": 2.9683990478515625, + "learning_rate": 9.057676685621447e-07, + "loss": 1.6184, + "step": 2308 + }, + { + "epoch": 2.5304109589041097, + "grad_norm": 2.919980049133301, + "learning_rate": 9.03736799350122e-07, + "loss": 1.5908, + "step": 2309 + }, + { + "epoch": 2.5315068493150683, + "grad_norm": 3.0819506645202637, + "learning_rate": 9.017059301380991e-07, + "loss": 1.6252, + "step": 2310 + }, + { + "epoch": 2.5315068493150683, + "eval_loss": 1.8249709606170654, + "eval_runtime": 44.4295, + "eval_samples_per_second": 22.508, + "eval_steps_per_second": 0.945, + "step": 2310 + }, + { + "epoch": 2.5326027397260273, + "grad_norm": 3.0078251361846924, + "learning_rate": 8.996750609260764e-07, + "loss": 1.5653, + "step": 2311 + }, + { + "epoch": 2.5336986301369864, + "grad_norm": 3.0950827598571777, + "learning_rate": 8.976441917140536e-07, + "loss": 1.6341, + "step": 2312 + }, + { + "epoch": 2.534794520547945, + "grad_norm": 2.9090089797973633, + "learning_rate": 8.956133225020309e-07, + "loss": 1.5441, + "step": 2313 + }, + { + "epoch": 2.535890410958904, + "grad_norm": 2.8889381885528564, + "learning_rate": 8.935824532900082e-07, + "loss": 1.5371, + "step": 2314 + }, + { + "epoch": 2.536986301369863, + "grad_norm": 3.3039140701293945, + "learning_rate": 8.915515840779855e-07, + "loss": 1.5936, + "step": 2315 + }, + { + "epoch": 2.538082191780822, + "grad_norm": 2.9188146591186523, + "learning_rate": 8.895207148659626e-07, + "loss": 1.5787, + "step": 2316 + }, + { + "epoch": 2.539178082191781, + "grad_norm": 2.8546876907348633, + "learning_rate": 8.874898456539399e-07, + "loss": 1.58, + "step": 2317 + }, + { + "epoch": 2.5402739726027397, + "grad_norm": 2.8007283210754395, + "learning_rate": 8.854589764419172e-07, + "loss": 1.518, + "step": 2318 + }, + { + "epoch": 2.5413698630136987, + "grad_norm": 2.8198978900909424, + "learning_rate": 8.834281072298945e-07, + "loss": 1.5695, + "step": 2319 + }, + { + "epoch": 2.5424657534246577, + "grad_norm": 3.3483028411865234, + "learning_rate": 8.813972380178717e-07, + "loss": 1.5814, + "step": 2320 + }, + { + "epoch": 2.5424657534246577, + "eval_loss": 1.8233474493026733, + "eval_runtime": 49.2421, + "eval_samples_per_second": 20.308, + "eval_steps_per_second": 0.853, + "step": 2320 + }, + { + "epoch": 2.5435616438356163, + "grad_norm": 2.838481903076172, + "learning_rate": 8.79366368805849e-07, + "loss": 1.5311, + "step": 2321 + }, + { + "epoch": 2.5446575342465754, + "grad_norm": 2.8281280994415283, + "learning_rate": 8.773354995938262e-07, + "loss": 1.5235, + "step": 2322 + }, + { + "epoch": 2.5457534246575344, + "grad_norm": 2.8211021423339844, + "learning_rate": 8.753046303818035e-07, + "loss": 1.5484, + "step": 2323 + }, + { + "epoch": 2.546849315068493, + "grad_norm": 2.865307331085205, + "learning_rate": 8.732737611697807e-07, + "loss": 1.5892, + "step": 2324 + }, + { + "epoch": 2.547945205479452, + "grad_norm": 2.87202787399292, + "learning_rate": 8.71242891957758e-07, + "loss": 1.5961, + "step": 2325 + }, + { + "epoch": 2.549041095890411, + "grad_norm": 2.9286715984344482, + "learning_rate": 8.692120227457353e-07, + "loss": 1.5597, + "step": 2326 + }, + { + "epoch": 2.5501369863013696, + "grad_norm": 2.965538740158081, + "learning_rate": 8.671811535337126e-07, + "loss": 1.6278, + "step": 2327 + }, + { + "epoch": 2.5512328767123287, + "grad_norm": 2.991199254989624, + "learning_rate": 8.651502843216898e-07, + "loss": 1.5904, + "step": 2328 + }, + { + "epoch": 2.5523287671232877, + "grad_norm": 2.902418613433838, + "learning_rate": 8.63119415109667e-07, + "loss": 1.584, + "step": 2329 + }, + { + "epoch": 2.5534246575342463, + "grad_norm": 2.8904943466186523, + "learning_rate": 8.610885458976443e-07, + "loss": 1.5178, + "step": 2330 + }, + { + "epoch": 2.5534246575342463, + "eval_loss": 1.8210104703903198, + "eval_runtime": 49.456, + "eval_samples_per_second": 20.22, + "eval_steps_per_second": 0.849, + "step": 2330 + }, + { + "epoch": 2.5545205479452053, + "grad_norm": 2.778411865234375, + "learning_rate": 8.590576766856216e-07, + "loss": 1.5318, + "step": 2331 + }, + { + "epoch": 2.5556164383561644, + "grad_norm": 2.8281290531158447, + "learning_rate": 8.570268074735988e-07, + "loss": 1.5399, + "step": 2332 + }, + { + "epoch": 2.5567123287671234, + "grad_norm": 2.8783960342407227, + "learning_rate": 8.549959382615761e-07, + "loss": 1.5653, + "step": 2333 + }, + { + "epoch": 2.5578082191780824, + "grad_norm": 2.7188220024108887, + "learning_rate": 8.529650690495534e-07, + "loss": 1.5113, + "step": 2334 + }, + { + "epoch": 2.558904109589041, + "grad_norm": 2.705050468444824, + "learning_rate": 8.509341998375305e-07, + "loss": 1.4955, + "step": 2335 + }, + { + "epoch": 2.56, + "grad_norm": 2.798614740371704, + "learning_rate": 8.489033306255077e-07, + "loss": 1.529, + "step": 2336 + }, + { + "epoch": 2.561095890410959, + "grad_norm": 2.795304298400879, + "learning_rate": 8.46872461413485e-07, + "loss": 1.5431, + "step": 2337 + }, + { + "epoch": 2.5621917808219177, + "grad_norm": 2.838188648223877, + "learning_rate": 8.448415922014623e-07, + "loss": 1.5772, + "step": 2338 + }, + { + "epoch": 2.5632876712328767, + "grad_norm": 2.9334065914154053, + "learning_rate": 8.428107229894396e-07, + "loss": 1.5621, + "step": 2339 + }, + { + "epoch": 2.5643835616438357, + "grad_norm": 2.9324002265930176, + "learning_rate": 8.407798537774168e-07, + "loss": 1.5422, + "step": 2340 + }, + { + "epoch": 2.5643835616438357, + "eval_loss": 1.8223708868026733, + "eval_runtime": 50.7517, + "eval_samples_per_second": 19.704, + "eval_steps_per_second": 0.828, + "step": 2340 + }, + { + "epoch": 2.5654794520547943, + "grad_norm": 2.7721550464630127, + "learning_rate": 8.38748984565394e-07, + "loss": 1.5507, + "step": 2341 + }, + { + "epoch": 2.5665753424657534, + "grad_norm": 2.918529510498047, + "learning_rate": 8.367181153533713e-07, + "loss": 1.5703, + "step": 2342 + }, + { + "epoch": 2.5676712328767124, + "grad_norm": 2.926311492919922, + "learning_rate": 8.346872461413486e-07, + "loss": 1.5611, + "step": 2343 + }, + { + "epoch": 2.568767123287671, + "grad_norm": 2.9318113327026367, + "learning_rate": 8.326563769293258e-07, + "loss": 1.6348, + "step": 2344 + }, + { + "epoch": 2.56986301369863, + "grad_norm": 2.8003222942352295, + "learning_rate": 8.306255077173031e-07, + "loss": 1.5064, + "step": 2345 + }, + { + "epoch": 2.570958904109589, + "grad_norm": 2.797013998031616, + "learning_rate": 8.285946385052804e-07, + "loss": 1.5907, + "step": 2346 + }, + { + "epoch": 2.572054794520548, + "grad_norm": 2.7971713542938232, + "learning_rate": 8.265637692932575e-07, + "loss": 1.5844, + "step": 2347 + }, + { + "epoch": 2.573150684931507, + "grad_norm": 2.8959572315216064, + "learning_rate": 8.245329000812348e-07, + "loss": 1.5173, + "step": 2348 + }, + { + "epoch": 2.5742465753424657, + "grad_norm": 2.8146274089813232, + "learning_rate": 8.225020308692121e-07, + "loss": 1.5615, + "step": 2349 + }, + { + "epoch": 2.5753424657534247, + "grad_norm": 2.817338705062866, + "learning_rate": 8.204711616571894e-07, + "loss": 1.5076, + "step": 2350 + }, + { + "epoch": 2.5753424657534247, + "eval_loss": 1.8196357488632202, + "eval_runtime": 48.5561, + "eval_samples_per_second": 20.595, + "eval_steps_per_second": 0.865, + "step": 2350 + }, + { + "epoch": 2.5764383561643838, + "grad_norm": 2.9276442527770996, + "learning_rate": 8.184402924451667e-07, + "loss": 1.5776, + "step": 2351 + }, + { + "epoch": 2.5775342465753424, + "grad_norm": 2.860095739364624, + "learning_rate": 8.164094232331439e-07, + "loss": 1.5677, + "step": 2352 + }, + { + "epoch": 2.5786301369863014, + "grad_norm": 2.917368173599243, + "learning_rate": 8.14378554021121e-07, + "loss": 1.5489, + "step": 2353 + }, + { + "epoch": 2.5797260273972604, + "grad_norm": 2.8872427940368652, + "learning_rate": 8.123476848090983e-07, + "loss": 1.5458, + "step": 2354 + }, + { + "epoch": 2.580821917808219, + "grad_norm": 2.8507397174835205, + "learning_rate": 8.103168155970755e-07, + "loss": 1.5662, + "step": 2355 + }, + { + "epoch": 2.581917808219178, + "grad_norm": 2.8826792240142822, + "learning_rate": 8.082859463850528e-07, + "loss": 1.5574, + "step": 2356 + }, + { + "epoch": 2.583013698630137, + "grad_norm": 2.8728246688842773, + "learning_rate": 8.062550771730301e-07, + "loss": 1.5516, + "step": 2357 + }, + { + "epoch": 2.5841095890410957, + "grad_norm": 2.9101784229278564, + "learning_rate": 8.042242079610074e-07, + "loss": 1.5472, + "step": 2358 + }, + { + "epoch": 2.5852054794520547, + "grad_norm": 2.791224718093872, + "learning_rate": 8.021933387489846e-07, + "loss": 1.5663, + "step": 2359 + }, + { + "epoch": 2.5863013698630137, + "grad_norm": 2.983250856399536, + "learning_rate": 8.001624695369618e-07, + "loss": 1.5908, + "step": 2360 + }, + { + "epoch": 2.5863013698630137, + "eval_loss": 1.8226059675216675, + "eval_runtime": 48.2046, + "eval_samples_per_second": 20.745, + "eval_steps_per_second": 0.871, + "step": 2360 + }, + { + "epoch": 2.5873972602739728, + "grad_norm": 2.920762538909912, + "learning_rate": 7.981316003249391e-07, + "loss": 1.5111, + "step": 2361 + }, + { + "epoch": 2.5884931506849314, + "grad_norm": 2.983764171600342, + "learning_rate": 7.961007311129164e-07, + "loss": 1.566, + "step": 2362 + }, + { + "epoch": 2.5895890410958904, + "grad_norm": 2.976851463317871, + "learning_rate": 7.940698619008936e-07, + "loss": 1.6571, + "step": 2363 + }, + { + "epoch": 2.5906849315068494, + "grad_norm": 2.856072187423706, + "learning_rate": 7.920389926888709e-07, + "loss": 1.5176, + "step": 2364 + }, + { + "epoch": 2.5917808219178085, + "grad_norm": 2.8299734592437744, + "learning_rate": 7.900081234768482e-07, + "loss": 1.587, + "step": 2365 + }, + { + "epoch": 2.592876712328767, + "grad_norm": 2.937560558319092, + "learning_rate": 7.879772542648254e-07, + "loss": 1.5769, + "step": 2366 + }, + { + "epoch": 2.593972602739726, + "grad_norm": 2.8437840938568115, + "learning_rate": 7.859463850528026e-07, + "loss": 1.5665, + "step": 2367 + }, + { + "epoch": 2.595068493150685, + "grad_norm": 3.0254647731781006, + "learning_rate": 7.839155158407799e-07, + "loss": 1.5783, + "step": 2368 + }, + { + "epoch": 2.5961643835616437, + "grad_norm": 2.895967960357666, + "learning_rate": 7.818846466287572e-07, + "loss": 1.5783, + "step": 2369 + }, + { + "epoch": 2.5972602739726027, + "grad_norm": 2.8868582248687744, + "learning_rate": 7.798537774167345e-07, + "loss": 1.527, + "step": 2370 + }, + { + "epoch": 2.5972602739726027, + "eval_loss": 1.8213152885437012, + "eval_runtime": 51.0508, + "eval_samples_per_second": 19.588, + "eval_steps_per_second": 0.823, + "step": 2370 + }, + { + "epoch": 2.5983561643835618, + "grad_norm": 2.906033992767334, + "learning_rate": 7.778229082047117e-07, + "loss": 1.5688, + "step": 2371 + }, + { + "epoch": 2.5994520547945204, + "grad_norm": 2.846966505050659, + "learning_rate": 7.757920389926889e-07, + "loss": 1.5184, + "step": 2372 + }, + { + "epoch": 2.6005479452054794, + "grad_norm": 3.124784231185913, + "learning_rate": 7.737611697806662e-07, + "loss": 1.5953, + "step": 2373 + }, + { + "epoch": 2.6016438356164384, + "grad_norm": 3.014098644256592, + "learning_rate": 7.717303005686435e-07, + "loss": 1.6318, + "step": 2374 + }, + { + "epoch": 2.602739726027397, + "grad_norm": 2.82450532913208, + "learning_rate": 7.696994313566207e-07, + "loss": 1.5106, + "step": 2375 + }, + { + "epoch": 2.603835616438356, + "grad_norm": 2.8513712882995605, + "learning_rate": 7.67668562144598e-07, + "loss": 1.5053, + "step": 2376 + }, + { + "epoch": 2.604931506849315, + "grad_norm": 2.99210786819458, + "learning_rate": 7.656376929325753e-07, + "loss": 1.5844, + "step": 2377 + }, + { + "epoch": 2.606027397260274, + "grad_norm": 2.858902931213379, + "learning_rate": 7.636068237205524e-07, + "loss": 1.5156, + "step": 2378 + }, + { + "epoch": 2.607123287671233, + "grad_norm": 2.9171338081359863, + "learning_rate": 7.615759545085296e-07, + "loss": 1.5474, + "step": 2379 + }, + { + "epoch": 2.6082191780821917, + "grad_norm": 2.868882417678833, + "learning_rate": 7.595450852965069e-07, + "loss": 1.5826, + "step": 2380 + }, + { + "epoch": 2.6082191780821917, + "eval_loss": 1.8230814933776855, + "eval_runtime": 46.2884, + "eval_samples_per_second": 21.604, + "eval_steps_per_second": 0.907, + "step": 2380 + }, + { + "epoch": 2.6093150684931508, + "grad_norm": 2.8466811180114746, + "learning_rate": 7.575142160844842e-07, + "loss": 1.5272, + "step": 2381 + }, + { + "epoch": 2.61041095890411, + "grad_norm": 2.8146281242370605, + "learning_rate": 7.554833468724615e-07, + "loss": 1.5303, + "step": 2382 + }, + { + "epoch": 2.6115068493150684, + "grad_norm": 2.8843190670013428, + "learning_rate": 7.534524776604387e-07, + "loss": 1.5767, + "step": 2383 + }, + { + "epoch": 2.6126027397260274, + "grad_norm": 2.988900899887085, + "learning_rate": 7.51421608448416e-07, + "loss": 1.5335, + "step": 2384 + }, + { + "epoch": 2.6136986301369864, + "grad_norm": 3.4128847122192383, + "learning_rate": 7.493907392363932e-07, + "loss": 1.485, + "step": 2385 + }, + { + "epoch": 2.614794520547945, + "grad_norm": 2.8596994876861572, + "learning_rate": 7.473598700243705e-07, + "loss": 1.5597, + "step": 2386 + }, + { + "epoch": 2.615890410958904, + "grad_norm": 2.923269748687744, + "learning_rate": 7.453290008123477e-07, + "loss": 1.5736, + "step": 2387 + }, + { + "epoch": 2.616986301369863, + "grad_norm": 2.8989574909210205, + "learning_rate": 7.43298131600325e-07, + "loss": 1.5458, + "step": 2388 + }, + { + "epoch": 2.6180821917808217, + "grad_norm": 2.935607671737671, + "learning_rate": 7.412672623883023e-07, + "loss": 1.4983, + "step": 2389 + }, + { + "epoch": 2.6191780821917807, + "grad_norm": 3.215050220489502, + "learning_rate": 7.392363931762796e-07, + "loss": 1.5571, + "step": 2390 + }, + { + "epoch": 2.6191780821917807, + "eval_loss": 1.8221906423568726, + "eval_runtime": 47.9911, + "eval_samples_per_second": 20.837, + "eval_steps_per_second": 0.875, + "step": 2390 + }, + { + "epoch": 2.6202739726027398, + "grad_norm": 3.0844898223876953, + "learning_rate": 7.372055239642567e-07, + "loss": 1.5767, + "step": 2391 + }, + { + "epoch": 2.621369863013699, + "grad_norm": 3.113297700881958, + "learning_rate": 7.35174654752234e-07, + "loss": 1.5895, + "step": 2392 + }, + { + "epoch": 2.6224657534246574, + "grad_norm": 2.904573917388916, + "learning_rate": 7.331437855402113e-07, + "loss": 1.5162, + "step": 2393 + }, + { + "epoch": 2.6235616438356164, + "grad_norm": 2.8379499912261963, + "learning_rate": 7.311129163281886e-07, + "loss": 1.5901, + "step": 2394 + }, + { + "epoch": 2.6246575342465754, + "grad_norm": 2.9121007919311523, + "learning_rate": 7.290820471161658e-07, + "loss": 1.5908, + "step": 2395 + }, + { + "epoch": 2.6257534246575345, + "grad_norm": 2.8151888847351074, + "learning_rate": 7.270511779041431e-07, + "loss": 1.5163, + "step": 2396 + }, + { + "epoch": 2.626849315068493, + "grad_norm": 2.9227914810180664, + "learning_rate": 7.250203086921203e-07, + "loss": 1.5466, + "step": 2397 + }, + { + "epoch": 2.627945205479452, + "grad_norm": 2.834669589996338, + "learning_rate": 7.229894394800976e-07, + "loss": 1.5724, + "step": 2398 + }, + { + "epoch": 2.629041095890411, + "grad_norm": 2.7271687984466553, + "learning_rate": 7.209585702680748e-07, + "loss": 1.5099, + "step": 2399 + }, + { + "epoch": 2.6301369863013697, + "grad_norm": 2.7215335369110107, + "learning_rate": 7.189277010560521e-07, + "loss": 1.5419, + "step": 2400 + }, + { + "epoch": 2.6301369863013697, + "eval_loss": 1.8201849460601807, + "eval_runtime": 50.7567, + "eval_samples_per_second": 19.702, + "eval_steps_per_second": 0.827, + "step": 2400 + }, + { + "epoch": 2.6312328767123287, + "grad_norm": 2.840486764907837, + "learning_rate": 7.168968318440294e-07, + "loss": 1.5203, + "step": 2401 + }, + { + "epoch": 2.632328767123288, + "grad_norm": 2.885180711746216, + "learning_rate": 7.148659626320067e-07, + "loss": 1.6037, + "step": 2402 + }, + { + "epoch": 2.6334246575342464, + "grad_norm": 2.9077260494232178, + "learning_rate": 7.128350934199837e-07, + "loss": 1.587, + "step": 2403 + }, + { + "epoch": 2.6345205479452054, + "grad_norm": 2.781118154525757, + "learning_rate": 7.10804224207961e-07, + "loss": 1.545, + "step": 2404 + }, + { + "epoch": 2.6356164383561644, + "grad_norm": 2.815735101699829, + "learning_rate": 7.087733549959383e-07, + "loss": 1.5327, + "step": 2405 + }, + { + "epoch": 2.636712328767123, + "grad_norm": 2.9639029502868652, + "learning_rate": 7.067424857839155e-07, + "loss": 1.5591, + "step": 2406 + }, + { + "epoch": 2.637808219178082, + "grad_norm": 2.9713387489318848, + "learning_rate": 7.047116165718928e-07, + "loss": 1.5416, + "step": 2407 + }, + { + "epoch": 2.638904109589041, + "grad_norm": 3.1526248455047607, + "learning_rate": 7.026807473598701e-07, + "loss": 1.607, + "step": 2408 + }, + { + "epoch": 2.64, + "grad_norm": 3.009585380554199, + "learning_rate": 7.006498781478473e-07, + "loss": 1.5223, + "step": 2409 + }, + { + "epoch": 2.641095890410959, + "grad_norm": 2.7750561237335205, + "learning_rate": 6.986190089358245e-07, + "loss": 1.5663, + "step": 2410 + }, + { + "epoch": 2.641095890410959, + "eval_loss": 1.8215185403823853, + "eval_runtime": 44.6832, + "eval_samples_per_second": 22.38, + "eval_steps_per_second": 0.94, + "step": 2410 + }, + { + "epoch": 2.6421917808219177, + "grad_norm": 3.1443002223968506, + "learning_rate": 6.965881397238018e-07, + "loss": 1.5878, + "step": 2411 + }, + { + "epoch": 2.643287671232877, + "grad_norm": 2.8203115463256836, + "learning_rate": 6.945572705117791e-07, + "loss": 1.5749, + "step": 2412 + }, + { + "epoch": 2.644383561643836, + "grad_norm": 2.878829002380371, + "learning_rate": 6.925264012997564e-07, + "loss": 1.5788, + "step": 2413 + }, + { + "epoch": 2.6454794520547944, + "grad_norm": 2.9213924407958984, + "learning_rate": 6.904955320877336e-07, + "loss": 1.57, + "step": 2414 + }, + { + "epoch": 2.6465753424657534, + "grad_norm": 3.0402517318725586, + "learning_rate": 6.884646628757109e-07, + "loss": 1.6245, + "step": 2415 + }, + { + "epoch": 2.6476712328767125, + "grad_norm": 2.874117612838745, + "learning_rate": 6.864337936636881e-07, + "loss": 1.6213, + "step": 2416 + }, + { + "epoch": 2.648767123287671, + "grad_norm": 2.841010570526123, + "learning_rate": 6.844029244516654e-07, + "loss": 1.5244, + "step": 2417 + }, + { + "epoch": 2.64986301369863, + "grad_norm": 2.8916752338409424, + "learning_rate": 6.823720552396426e-07, + "loss": 1.5824, + "step": 2418 + }, + { + "epoch": 2.650958904109589, + "grad_norm": 2.8366785049438477, + "learning_rate": 6.803411860276199e-07, + "loss": 1.5663, + "step": 2419 + }, + { + "epoch": 2.6520547945205477, + "grad_norm": 2.881715774536133, + "learning_rate": 6.783103168155972e-07, + "loss": 1.5599, + "step": 2420 + }, + { + "epoch": 2.6520547945205477, + "eval_loss": 1.8196642398834229, + "eval_runtime": 50.2214, + "eval_samples_per_second": 19.912, + "eval_steps_per_second": 0.836, + "step": 2420 + }, + { + "epoch": 2.6531506849315067, + "grad_norm": 2.83880615234375, + "learning_rate": 6.762794476035745e-07, + "loss": 1.5839, + "step": 2421 + }, + { + "epoch": 2.6542465753424658, + "grad_norm": 2.8980135917663574, + "learning_rate": 6.742485783915515e-07, + "loss": 1.5898, + "step": 2422 + }, + { + "epoch": 2.655342465753425, + "grad_norm": 2.82328724861145, + "learning_rate": 6.722177091795288e-07, + "loss": 1.5778, + "step": 2423 + }, + { + "epoch": 2.6564383561643834, + "grad_norm": 2.7781100273132324, + "learning_rate": 6.701868399675061e-07, + "loss": 1.5381, + "step": 2424 + }, + { + "epoch": 2.6575342465753424, + "grad_norm": 2.94514536857605, + "learning_rate": 6.681559707554834e-07, + "loss": 1.5305, + "step": 2425 + }, + { + "epoch": 2.6586301369863015, + "grad_norm": 2.9485371112823486, + "learning_rate": 6.661251015434606e-07, + "loss": 1.5348, + "step": 2426 + }, + { + "epoch": 2.6597260273972605, + "grad_norm": 2.8444228172302246, + "learning_rate": 6.640942323314379e-07, + "loss": 1.5651, + "step": 2427 + }, + { + "epoch": 2.660821917808219, + "grad_norm": 2.923813819885254, + "learning_rate": 6.620633631194151e-07, + "loss": 1.5827, + "step": 2428 + }, + { + "epoch": 2.661917808219178, + "grad_norm": 2.8742222785949707, + "learning_rate": 6.600324939073924e-07, + "loss": 1.5464, + "step": 2429 + }, + { + "epoch": 2.663013698630137, + "grad_norm": 2.963705539703369, + "learning_rate": 6.580016246953696e-07, + "loss": 1.615, + "step": 2430 + }, + { + "epoch": 2.663013698630137, + "eval_loss": 1.8202346563339233, + "eval_runtime": 53.0891, + "eval_samples_per_second": 18.836, + "eval_steps_per_second": 0.791, + "step": 2430 + }, + { + "epoch": 2.6641095890410957, + "grad_norm": 2.9104220867156982, + "learning_rate": 6.559707554833469e-07, + "loss": 1.5432, + "step": 2431 + }, + { + "epoch": 2.6652054794520548, + "grad_norm": 2.9200713634490967, + "learning_rate": 6.539398862713242e-07, + "loss": 1.5988, + "step": 2432 + }, + { + "epoch": 2.666301369863014, + "grad_norm": 2.861328363418579, + "learning_rate": 6.519090170593015e-07, + "loss": 1.5721, + "step": 2433 + }, + { + "epoch": 2.6673972602739724, + "grad_norm": 2.9656898975372314, + "learning_rate": 6.498781478472786e-07, + "loss": 1.598, + "step": 2434 + }, + { + "epoch": 2.6684931506849314, + "grad_norm": 2.9453625679016113, + "learning_rate": 6.478472786352559e-07, + "loss": 1.564, + "step": 2435 + }, + { + "epoch": 2.6695890410958905, + "grad_norm": 2.8961451053619385, + "learning_rate": 6.458164094232332e-07, + "loss": 1.5021, + "step": 2436 + }, + { + "epoch": 2.670684931506849, + "grad_norm": 2.8768153190612793, + "learning_rate": 6.437855402112105e-07, + "loss": 1.5226, + "step": 2437 + }, + { + "epoch": 2.671780821917808, + "grad_norm": 2.942457914352417, + "learning_rate": 6.417546709991877e-07, + "loss": 1.5903, + "step": 2438 + }, + { + "epoch": 2.672876712328767, + "grad_norm": 2.82218074798584, + "learning_rate": 6.39723801787165e-07, + "loss": 1.5596, + "step": 2439 + }, + { + "epoch": 2.673972602739726, + "grad_norm": 2.828130006790161, + "learning_rate": 6.376929325751423e-07, + "loss": 1.5491, + "step": 2440 + }, + { + "epoch": 2.673972602739726, + "eval_loss": 1.820644736289978, + "eval_runtime": 48.3133, + "eval_samples_per_second": 20.698, + "eval_steps_per_second": 0.869, + "step": 2440 + }, + { + "epoch": 2.675068493150685, + "grad_norm": 2.8218741416931152, + "learning_rate": 6.356620633631195e-07, + "loss": 1.5274, + "step": 2441 + }, + { + "epoch": 2.6761643835616438, + "grad_norm": 3.038788080215454, + "learning_rate": 6.336311941510967e-07, + "loss": 1.5708, + "step": 2442 + }, + { + "epoch": 2.677260273972603, + "grad_norm": 3.0761072635650635, + "learning_rate": 6.31600324939074e-07, + "loss": 1.6051, + "step": 2443 + }, + { + "epoch": 2.678356164383562, + "grad_norm": 2.881915807723999, + "learning_rate": 6.295694557270513e-07, + "loss": 1.5661, + "step": 2444 + }, + { + "epoch": 2.6794520547945204, + "grad_norm": 2.8945391178131104, + "learning_rate": 6.275385865150286e-07, + "loss": 1.5515, + "step": 2445 + }, + { + "epoch": 2.6805479452054795, + "grad_norm": 2.8491926193237305, + "learning_rate": 6.255077173030058e-07, + "loss": 1.5675, + "step": 2446 + }, + { + "epoch": 2.6816438356164385, + "grad_norm": 2.860481023788452, + "learning_rate": 6.23476848090983e-07, + "loss": 1.5485, + "step": 2447 + }, + { + "epoch": 2.682739726027397, + "grad_norm": 2.8958401679992676, + "learning_rate": 6.214459788789602e-07, + "loss": 1.5927, + "step": 2448 + }, + { + "epoch": 2.683835616438356, + "grad_norm": 2.8343098163604736, + "learning_rate": 6.194151096669375e-07, + "loss": 1.5443, + "step": 2449 + }, + { + "epoch": 2.684931506849315, + "grad_norm": 2.8697097301483154, + "learning_rate": 6.173842404549147e-07, + "loss": 1.5521, + "step": 2450 + }, + { + "epoch": 2.684931506849315, + "eval_loss": 1.8198002576828003, + "eval_runtime": 50.7198, + "eval_samples_per_second": 19.716, + "eval_steps_per_second": 0.828, + "step": 2450 + }, + { + "epoch": 2.6860273972602737, + "grad_norm": 2.8190839290618896, + "learning_rate": 6.15353371242892e-07, + "loss": 1.4845, + "step": 2451 + }, + { + "epoch": 2.6871232876712328, + "grad_norm": 2.7570931911468506, + "learning_rate": 6.133225020308693e-07, + "loss": 1.5452, + "step": 2452 + }, + { + "epoch": 2.688219178082192, + "grad_norm": 2.960784673690796, + "learning_rate": 6.112916328188466e-07, + "loss": 1.6249, + "step": 2453 + }, + { + "epoch": 2.689315068493151, + "grad_norm": 2.7679786682128906, + "learning_rate": 6.092607636068238e-07, + "loss": 1.5327, + "step": 2454 + }, + { + "epoch": 2.69041095890411, + "grad_norm": 2.8044891357421875, + "learning_rate": 6.07229894394801e-07, + "loss": 1.5435, + "step": 2455 + }, + { + "epoch": 2.6915068493150685, + "grad_norm": 2.8200135231018066, + "learning_rate": 6.051990251827783e-07, + "loss": 1.5759, + "step": 2456 + }, + { + "epoch": 2.6926027397260275, + "grad_norm": 2.904179096221924, + "learning_rate": 6.031681559707556e-07, + "loss": 1.5597, + "step": 2457 + }, + { + "epoch": 2.6936986301369865, + "grad_norm": 2.8314926624298096, + "learning_rate": 6.011372867587327e-07, + "loss": 1.5394, + "step": 2458 + }, + { + "epoch": 2.694794520547945, + "grad_norm": 2.8240511417388916, + "learning_rate": 5.9910641754671e-07, + "loss": 1.5517, + "step": 2459 + }, + { + "epoch": 2.695890410958904, + "grad_norm": 2.844630718231201, + "learning_rate": 5.970755483346873e-07, + "loss": 1.5426, + "step": 2460 + }, + { + "epoch": 2.695890410958904, + "eval_loss": 1.8208848237991333, + "eval_runtime": 52.7964, + "eval_samples_per_second": 18.941, + "eval_steps_per_second": 0.796, + "step": 2460 + }, + { + "epoch": 2.696986301369863, + "grad_norm": 2.926901340484619, + "learning_rate": 5.950446791226646e-07, + "loss": 1.5457, + "step": 2461 + }, + { + "epoch": 2.6980821917808218, + "grad_norm": 2.845421314239502, + "learning_rate": 5.930138099106418e-07, + "loss": 1.5397, + "step": 2462 + }, + { + "epoch": 2.699178082191781, + "grad_norm": 2.984661817550659, + "learning_rate": 5.909829406986191e-07, + "loss": 1.5649, + "step": 2463 + }, + { + "epoch": 2.70027397260274, + "grad_norm": 2.8373823165893555, + "learning_rate": 5.889520714865963e-07, + "loss": 1.5255, + "step": 2464 + }, + { + "epoch": 2.7013698630136984, + "grad_norm": 2.9435815811157227, + "learning_rate": 5.869212022745735e-07, + "loss": 1.5705, + "step": 2465 + }, + { + "epoch": 2.7024657534246574, + "grad_norm": 2.8383870124816895, + "learning_rate": 5.848903330625508e-07, + "loss": 1.5457, + "step": 2466 + }, + { + "epoch": 2.7035616438356165, + "grad_norm": 2.899071216583252, + "learning_rate": 5.828594638505281e-07, + "loss": 1.5474, + "step": 2467 + }, + { + "epoch": 2.704657534246575, + "grad_norm": 2.9021337032318115, + "learning_rate": 5.808285946385054e-07, + "loss": 1.5549, + "step": 2468 + }, + { + "epoch": 2.705753424657534, + "grad_norm": 2.9504072666168213, + "learning_rate": 5.787977254264827e-07, + "loss": 1.6014, + "step": 2469 + }, + { + "epoch": 2.706849315068493, + "grad_norm": 2.9223573207855225, + "learning_rate": 5.767668562144598e-07, + "loss": 1.5574, + "step": 2470 + }, + { + "epoch": 2.706849315068493, + "eval_loss": 1.8203133344650269, + "eval_runtime": 49.5424, + "eval_samples_per_second": 20.185, + "eval_steps_per_second": 0.848, + "step": 2470 + }, + { + "epoch": 2.707945205479452, + "grad_norm": 2.8095791339874268, + "learning_rate": 5.747359870024371e-07, + "loss": 1.5462, + "step": 2471 + }, + { + "epoch": 2.709041095890411, + "grad_norm": 2.8600597381591797, + "learning_rate": 5.727051177904144e-07, + "loss": 1.5872, + "step": 2472 + }, + { + "epoch": 2.71013698630137, + "grad_norm": 2.8269565105438232, + "learning_rate": 5.706742485783915e-07, + "loss": 1.5218, + "step": 2473 + }, + { + "epoch": 2.711232876712329, + "grad_norm": 2.7480826377868652, + "learning_rate": 5.686433793663688e-07, + "loss": 1.5049, + "step": 2474 + }, + { + "epoch": 2.712328767123288, + "grad_norm": 2.9086508750915527, + "learning_rate": 5.666125101543461e-07, + "loss": 1.604, + "step": 2475 + }, + { + "epoch": 2.7134246575342464, + "grad_norm": 2.791057825088501, + "learning_rate": 5.645816409423234e-07, + "loss": 1.5647, + "step": 2476 + }, + { + "epoch": 2.7145205479452055, + "grad_norm": 2.8584253787994385, + "learning_rate": 5.625507717303006e-07, + "loss": 1.5477, + "step": 2477 + }, + { + "epoch": 2.7156164383561645, + "grad_norm": 2.8320658206939697, + "learning_rate": 5.605199025182779e-07, + "loss": 1.5245, + "step": 2478 + }, + { + "epoch": 2.716712328767123, + "grad_norm": 2.867785930633545, + "learning_rate": 5.584890333062551e-07, + "loss": 1.6039, + "step": 2479 + }, + { + "epoch": 2.717808219178082, + "grad_norm": 2.942857027053833, + "learning_rate": 5.564581640942324e-07, + "loss": 1.5313, + "step": 2480 + }, + { + "epoch": 2.717808219178082, + "eval_loss": 1.821783423423767, + "eval_runtime": 51.124, + "eval_samples_per_second": 19.56, + "eval_steps_per_second": 0.822, + "step": 2480 + }, + { + "epoch": 2.718904109589041, + "grad_norm": 2.8085427284240723, + "learning_rate": 5.544272948822096e-07, + "loss": 1.5575, + "step": 2481 + }, + { + "epoch": 2.7199999999999998, + "grad_norm": 2.8479437828063965, + "learning_rate": 5.523964256701868e-07, + "loss": 1.5816, + "step": 2482 + }, + { + "epoch": 2.721095890410959, + "grad_norm": 2.8424737453460693, + "learning_rate": 5.503655564581641e-07, + "loss": 1.5434, + "step": 2483 + }, + { + "epoch": 2.722191780821918, + "grad_norm": 2.8704795837402344, + "learning_rate": 5.483346872461414e-07, + "loss": 1.5565, + "step": 2484 + }, + { + "epoch": 2.723287671232877, + "grad_norm": 2.868283987045288, + "learning_rate": 5.463038180341186e-07, + "loss": 1.5173, + "step": 2485 + }, + { + "epoch": 2.724383561643836, + "grad_norm": 2.8845019340515137, + "learning_rate": 5.442729488220959e-07, + "loss": 1.5581, + "step": 2486 + }, + { + "epoch": 2.7254794520547945, + "grad_norm": 2.83331036567688, + "learning_rate": 5.422420796100732e-07, + "loss": 1.5502, + "step": 2487 + }, + { + "epoch": 2.7265753424657535, + "grad_norm": 2.8796298503875732, + "learning_rate": 5.402112103980505e-07, + "loss": 1.5737, + "step": 2488 + }, + { + "epoch": 2.7276712328767125, + "grad_norm": 2.939396381378174, + "learning_rate": 5.381803411860276e-07, + "loss": 1.6211, + "step": 2489 + }, + { + "epoch": 2.728767123287671, + "grad_norm": 2.8413333892822266, + "learning_rate": 5.361494719740049e-07, + "loss": 1.5263, + "step": 2490 + }, + { + "epoch": 2.728767123287671, + "eval_loss": 1.819288969039917, + "eval_runtime": 50.1547, + "eval_samples_per_second": 19.938, + "eval_steps_per_second": 0.837, + "step": 2490 + }, + { + "epoch": 2.72986301369863, + "grad_norm": 2.9400246143341064, + "learning_rate": 5.341186027619822e-07, + "loss": 1.5439, + "step": 2491 + }, + { + "epoch": 2.730958904109589, + "grad_norm": 2.8259570598602295, + "learning_rate": 5.320877335499594e-07, + "loss": 1.6046, + "step": 2492 + }, + { + "epoch": 2.732054794520548, + "grad_norm": 2.9316084384918213, + "learning_rate": 5.300568643379366e-07, + "loss": 1.569, + "step": 2493 + }, + { + "epoch": 2.733150684931507, + "grad_norm": 2.8864996433258057, + "learning_rate": 5.280259951259139e-07, + "loss": 1.5384, + "step": 2494 + }, + { + "epoch": 2.734246575342466, + "grad_norm": 3.0581302642822266, + "learning_rate": 5.259951259138912e-07, + "loss": 1.5318, + "step": 2495 + }, + { + "epoch": 2.7353424657534244, + "grad_norm": 2.8676114082336426, + "learning_rate": 5.239642567018685e-07, + "loss": 1.543, + "step": 2496 + }, + { + "epoch": 2.7364383561643835, + "grad_norm": 2.935861110687256, + "learning_rate": 5.219333874898457e-07, + "loss": 1.5603, + "step": 2497 + }, + { + "epoch": 2.7375342465753425, + "grad_norm": 2.963350296020508, + "learning_rate": 5.199025182778229e-07, + "loss": 1.5479, + "step": 2498 + }, + { + "epoch": 2.7386301369863015, + "grad_norm": 2.9598515033721924, + "learning_rate": 5.178716490658002e-07, + "loss": 1.5116, + "step": 2499 + }, + { + "epoch": 2.73972602739726, + "grad_norm": 2.7877373695373535, + "learning_rate": 5.158407798537775e-07, + "loss": 1.4805, + "step": 2500 + }, + { + "epoch": 2.73972602739726, + "eval_loss": 1.8200916051864624, + "eval_runtime": 50.4013, + "eval_samples_per_second": 19.841, + "eval_steps_per_second": 0.833, + "step": 2500 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 7.7932735234048e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2500/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/tokenizer.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..ef619d58a1c1ba0b3e2a2cf98779f6ec025df2cb --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2600/trainer_state.json @@ -0,0 +1,20313 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.8493150684931505, + "eval_steps": 10, + "global_step": 2600, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.5818793773651123, + "learning_rate": 4.569455727051178e-06, + "loss": 1.85, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 3.0734691619873047, + "learning_rate": 4.567424857839156e-06, + "loss": 1.8682, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.496253252029419, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9007, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.360891342163086, + "learning_rate": 4.5633631194151105e-06, + "loss": 1.8262, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.5731117725372314, + "learning_rate": 4.561332250203087e-06, + "loss": 1.8369, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.700558662414551, + "learning_rate": 4.559301380991065e-06, + "loss": 1.7825, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.561306953430176, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.8174, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.618889570236206, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.8812, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.4671308994293213, + "learning_rate": 4.553208773354996e-06, + "loss": 1.7959, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.4571878910064697, + "learning_rate": 4.551177904142974e-06, + "loss": 1.8943, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.010777711868286, + "eval_runtime": 43.7424, + "eval_samples_per_second": 22.861, + "eval_steps_per_second": 0.96, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.4786667823791504, + "learning_rate": 4.5491470349309505e-06, + "loss": 1.8147, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.2955000400543213, + "learning_rate": 4.547116165718929e-06, + "loss": 1.8547, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.3873913288116455, + "learning_rate": 4.545085296506905e-06, + "loss": 1.8156, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.2449734210968018, + "learning_rate": 4.543054427294882e-06, + "loss": 1.8053, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.3969857692718506, + "learning_rate": 4.54102355808286e-06, + "loss": 1.8085, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.3446357250213623, + "learning_rate": 4.538992688870837e-06, + "loss": 1.8355, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.408762216567993, + "learning_rate": 4.536961819658814e-06, + "loss": 1.8319, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.362962245941162, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.8422, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.402984857559204, + "learning_rate": 4.532900081234769e-06, + "loss": 1.8125, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.3964452743530273, + "learning_rate": 4.530869212022746e-06, + "loss": 1.8417, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.006908416748047, + "eval_runtime": 44.2701, + "eval_samples_per_second": 22.589, + "eval_steps_per_second": 0.949, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.595233201980591, + "learning_rate": 4.528838342810723e-06, + "loss": 1.8275, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.326117753982544, + "learning_rate": 4.526807473598701e-06, + "loss": 1.8405, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.4904465675354004, + "learning_rate": 4.524776604386678e-06, + "loss": 1.8172, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.4591610431671143, + "learning_rate": 4.522745735174655e-06, + "loss": 1.876, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.319800853729248, + "learning_rate": 4.520714865962632e-06, + "loss": 1.8294, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.28790020942688, + "learning_rate": 4.51868399675061e-06, + "loss": 1.8537, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.284822702407837, + "learning_rate": 4.516653127538587e-06, + "loss": 1.8087, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.273820638656616, + "learning_rate": 4.514622258326564e-06, + "loss": 1.7992, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.2025654315948486, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.8248, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 2.3218843936920166, + "learning_rate": 4.510560519902519e-06, + "loss": 1.8647, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.0048956871032715, + "eval_runtime": 44.5972, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.515766143798828, + "learning_rate": 4.508529650690496e-06, + "loss": 1.7673, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.4376487731933594, + "learning_rate": 4.506498781478473e-06, + "loss": 1.7755, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.2801272869110107, + "learning_rate": 4.504467912266451e-06, + "loss": 1.8273, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.4707891941070557, + "learning_rate": 4.502437043054428e-06, + "loss": 1.8971, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.301301956176758, + "learning_rate": 4.500406173842405e-06, + "loss": 1.8345, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.2317373752593994, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.8248, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.447373867034912, + "learning_rate": 4.49634443541836e-06, + "loss": 1.8148, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.337245225906372, + "learning_rate": 4.494313566206336e-06, + "loss": 1.8811, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3717031478881836, + "learning_rate": 4.492282696994314e-06, + "loss": 1.8311, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.287677526473999, + "learning_rate": 4.490251827782291e-06, + "loss": 1.7728, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.004889965057373, + "eval_runtime": 43.8132, + "eval_samples_per_second": 22.824, + "eval_steps_per_second": 0.959, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.441445827484131, + "learning_rate": 4.488220958570269e-06, + "loss": 1.7927, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.3191535472869873, + "learning_rate": 4.486190089358245e-06, + "loss": 1.759, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.3118820190429688, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8509, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.4711859226226807, + "learning_rate": 4.4821283509342e-06, + "loss": 1.8402, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.3142173290252686, + "learning_rate": 4.480097481722178e-06, + "loss": 1.8238, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.2665774822235107, + "learning_rate": 4.478066612510154e-06, + "loss": 1.7826, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.1692490577697754, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.8138, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.412783145904541, + "learning_rate": 4.474004874086109e-06, + "loss": 1.9421, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3692686557769775, + "learning_rate": 4.471974004874087e-06, + "loss": 1.7722, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.3224151134490967, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.8753, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.0016305446624756, + "eval_runtime": 45.564, + "eval_samples_per_second": 21.947, + "eval_steps_per_second": 0.922, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.51043438911438, + "learning_rate": 4.467912266450041e-06, + "loss": 1.8325, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.4568657875061035, + "learning_rate": 4.465881397238018e-06, + "loss": 1.8081, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.3279001712799072, + "learning_rate": 4.463850528025995e-06, + "loss": 1.8439, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.6041171550750732, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.8623, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.3780531883239746, + "learning_rate": 4.45978878960195e-06, + "loss": 1.8269, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.511500597000122, + "learning_rate": 4.457757920389927e-06, + "loss": 1.7749, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.707789659500122, + "learning_rate": 4.455727051177904e-06, + "loss": 1.8157, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.5495495796203613, + "learning_rate": 4.453696181965882e-06, + "loss": 1.8534, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.255876302719116, + "learning_rate": 4.451665312753859e-06, + "loss": 1.8377, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.415097951889038, + "learning_rate": 4.449634443541836e-06, + "loss": 1.8764, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 1.9943474531173706, + "eval_runtime": 42.2985, + "eval_samples_per_second": 23.642, + "eval_steps_per_second": 0.993, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.2604713439941406, + "learning_rate": 4.4476035743298135e-06, + "loss": 1.7898, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.5643842220306396, + "learning_rate": 4.445572705117791e-06, + "loss": 1.7953, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.412548780441284, + "learning_rate": 4.443541835905768e-06, + "loss": 1.8611, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.5423877239227295, + "learning_rate": 4.441510966693745e-06, + "loss": 1.8049, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.4986631870269775, + "learning_rate": 4.439480097481723e-06, + "loss": 1.8759, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.8349273204803467, + "learning_rate": 4.4374492282697e-06, + "loss": 1.7416, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.452017068862915, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8268, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.6425273418426514, + "learning_rate": 4.433387489845654e-06, + "loss": 1.8334, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.328740358352661, + "learning_rate": 4.431356620633632e-06, + "loss": 1.7931, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.262209415435791, + "learning_rate": 4.429325751421609e-06, + "loss": 1.8151, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 1.9917123317718506, + "eval_runtime": 42.4309, + "eval_samples_per_second": 23.568, + "eval_steps_per_second": 0.99, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.2984049320220947, + "learning_rate": 4.427294882209586e-06, + "loss": 1.8187, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.3910019397735596, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.7512, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.4494471549987793, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8202, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.3857171535491943, + "learning_rate": 4.421202274573518e-06, + "loss": 1.8279, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.4526500701904297, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.7136, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.2216501235961914, + "learning_rate": 4.417140536149473e-06, + "loss": 1.7574, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.7639758586883545, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8403, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.3177521228790283, + "learning_rate": 4.413078797725427e-06, + "loss": 1.7898, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.527902364730835, + "learning_rate": 4.411047928513404e-06, + "loss": 1.8184, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.34061336517334, + "learning_rate": 4.409017059301382e-06, + "loss": 1.811, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 1.9882216453552246, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.6462934017181396, + "learning_rate": 4.406986190089358e-06, + "loss": 1.8501, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.2725913524627686, + "learning_rate": 4.404955320877336e-06, + "loss": 1.8262, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.390084981918335, + "learning_rate": 4.402924451665313e-06, + "loss": 1.7842, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.314305305480957, + "learning_rate": 4.400893582453291e-06, + "loss": 1.7692, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.2828257083892822, + "learning_rate": 4.398862713241267e-06, + "loss": 1.8031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.3493599891662598, + "learning_rate": 4.396831844029245e-06, + "loss": 1.8183, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.423365831375122, + "learning_rate": 4.394800974817222e-06, + "loss": 1.8012, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.438015937805176, + "learning_rate": 4.392770105605199e-06, + "loss": 1.8217, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.357172727584839, + "learning_rate": 4.390739236393176e-06, + "loss": 1.8382, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.3383474349975586, + "learning_rate": 4.388708367181154e-06, + "loss": 1.841, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 1.9878787994384766, + "eval_runtime": 44.0694, + "eval_samples_per_second": 22.691, + "eval_steps_per_second": 0.953, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.459385633468628, + "learning_rate": 4.386677497969131e-06, + "loss": 1.8377, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.356788158416748, + "learning_rate": 4.384646628757108e-06, + "loss": 1.8718, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.3217222690582275, + "learning_rate": 4.3826157595450855e-06, + "loss": 1.8073, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.3134331703186035, + "learning_rate": 4.380584890333063e-06, + "loss": 1.7675, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.3273653984069824, + "learning_rate": 4.37855402112104e-06, + "loss": 1.7169, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.611034393310547, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8702, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.2077300548553467, + "learning_rate": 4.3744922826969946e-06, + "loss": 1.7951, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.3393311500549316, + "learning_rate": 4.372461413484972e-06, + "loss": 1.8136, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.3306660652160645, + "learning_rate": 4.370430544272949e-06, + "loss": 1.8372, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.4440250396728516, + "learning_rate": 4.368399675060926e-06, + "loss": 1.844, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 1.98439621925354, + "eval_runtime": 44.4595, + "eval_samples_per_second": 22.492, + "eval_steps_per_second": 0.945, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.205542802810669, + "learning_rate": 4.366368805848904e-06, + "loss": 1.8182, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.6030094623565674, + "learning_rate": 4.364337936636881e-06, + "loss": 1.7524, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.4000444412231445, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8269, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.319526433944702, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.7722, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.1624808311462402, + "learning_rate": 4.358245329000813e-06, + "loss": 1.8081, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.26615309715271, + "learning_rate": 4.35621445978879e-06, + "loss": 1.8049, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.4895036220550537, + "learning_rate": 4.354183590576767e-06, + "loss": 1.7932, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.437161445617676, + "learning_rate": 4.352152721364745e-06, + "loss": 1.772, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.344069480895996, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8213, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.360734224319458, + "learning_rate": 4.348090982940699e-06, + "loss": 1.8268, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 1.9846479892730713, + "eval_runtime": 44.947, + "eval_samples_per_second": 22.248, + "eval_steps_per_second": 0.934, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.59487247467041, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.8303, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.6512107849121094, + "learning_rate": 4.344029244516654e-06, + "loss": 1.7979, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.2900476455688477, + "learning_rate": 4.341998375304631e-06, + "loss": 1.8707, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.409494638442993, + "learning_rate": 4.339967506092607e-06, + "loss": 1.8075, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.527205467224121, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.7884, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.33461332321167, + "learning_rate": 4.335905767668562e-06, + "loss": 1.7839, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.3555047512054443, + "learning_rate": 4.33387489845654e-06, + "loss": 1.8245, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.397148609161377, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.8115, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.232139825820923, + "learning_rate": 4.329813160032495e-06, + "loss": 1.7747, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.1684389114379883, + "learning_rate": 4.327782290820471e-06, + "loss": 1.748, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 1.980297565460205, + "eval_runtime": 42.9183, + "eval_samples_per_second": 23.3, + "eval_steps_per_second": 0.979, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.2454638481140137, + "learning_rate": 4.325751421608449e-06, + "loss": 1.83, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.3264827728271484, + "learning_rate": 4.323720552396426e-06, + "loss": 1.7856, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.298804998397827, + "learning_rate": 4.321689683184404e-06, + "loss": 1.78, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2202727794647217, + "learning_rate": 4.31965881397238e-06, + "loss": 1.7875, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.494479179382324, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.8363, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3520524501800537, + "learning_rate": 4.315597075548335e-06, + "loss": 1.8266, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.373878002166748, + "learning_rate": 4.313566206336312e-06, + "loss": 1.8348, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.4232168197631836, + "learning_rate": 4.311535337124289e-06, + "loss": 1.77, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.2572121620178223, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.7914, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.5817952156066895, + "learning_rate": 4.307473598700244e-06, + "loss": 1.8099, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 1.9784427881240845, + "eval_runtime": 43.2981, + "eval_samples_per_second": 23.096, + "eval_steps_per_second": 0.97, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.646040916442871, + "learning_rate": 4.305442729488221e-06, + "loss": 1.7452, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.370220184326172, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8399, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.4599130153656006, + "learning_rate": 4.301380991064176e-06, + "loss": 1.8273, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 2.4135093688964844, + "learning_rate": 4.299350121852153e-06, + "loss": 1.852, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.371021270751953, + "learning_rate": 4.29731925264013e-06, + "loss": 1.8339, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.6858229637145996, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.792, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.4101266860961914, + "learning_rate": 4.293257514216085e-06, + "loss": 1.7556, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.769761800765991, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8223, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.5777804851531982, + "learning_rate": 4.289195775792039e-06, + "loss": 1.8406, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.376864433288574, + "learning_rate": 4.287164906580017e-06, + "loss": 1.7846, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 1.9758412837982178, + "eval_runtime": 43.9353, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.7461960315704346, + "learning_rate": 4.285134037367994e-06, + "loss": 1.776, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.549576759338379, + "learning_rate": 4.283103168155971e-06, + "loss": 1.7644, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.568074941635132, + "learning_rate": 4.281072298943948e-06, + "loss": 1.7571, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.8195722103118896, + "learning_rate": 4.279041429731926e-06, + "loss": 1.7959, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.471623182296753, + "learning_rate": 4.277010560519903e-06, + "loss": 1.7584, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.664689302444458, + "learning_rate": 4.27497969130788e-06, + "loss": 1.7909, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.684504985809326, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.8205, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.5247528553009033, + "learning_rate": 4.270917952883835e-06, + "loss": 1.8365, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.757972478866577, + "learning_rate": 4.268887083671812e-06, + "loss": 1.7665, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.6032614707946777, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8702, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 1.9716291427612305, + "eval_runtime": 43.5482, + "eval_samples_per_second": 22.963, + "eval_steps_per_second": 0.964, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.5324370861053467, + "learning_rate": 4.264825345247766e-06, + "loss": 1.738, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.730712890625, + "learning_rate": 4.262794476035744e-06, + "loss": 1.7918, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.5554966926574707, + "learning_rate": 4.26076360682372e-06, + "loss": 1.7218, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.4774436950683594, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.7823, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.77431058883667, + "learning_rate": 4.256701868399675e-06, + "loss": 1.794, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.258828639984131, + "learning_rate": 4.254670999187653e-06, + "loss": 1.805, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.384399175643921, + "learning_rate": 4.252640129975629e-06, + "loss": 1.818, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3278932571411133, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.7719, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.207348108291626, + "learning_rate": 4.248578391551584e-06, + "loss": 1.7362, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.4525978565216064, + "learning_rate": 4.246547522339562e-06, + "loss": 1.778, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 1.9694374799728394, + "eval_runtime": 42.5977, + "eval_samples_per_second": 23.475, + "eval_steps_per_second": 0.986, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.377835273742676, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.7926, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.4623866081237793, + "learning_rate": 4.242485783915517e-06, + "loss": 1.7366, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.542219400405884, + "learning_rate": 4.240454914703493e-06, + "loss": 1.7409, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.1955819129943848, + "learning_rate": 4.23842404549147e-06, + "loss": 1.7596, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.5834269523620605, + "learning_rate": 4.236393176279448e-06, + "loss": 1.7343, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.219660520553589, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8089, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.283200740814209, + "learning_rate": 4.232331437855402e-06, + "loss": 1.8248, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.274127244949341, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.825, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.3550686836242676, + "learning_rate": 4.228269699431357e-06, + "loss": 1.7957, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.588027238845825, + "learning_rate": 4.226238830219334e-06, + "loss": 1.7865, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 1.9662212133407593, + "eval_runtime": 43.2661, + "eval_samples_per_second": 23.113, + "eval_steps_per_second": 0.971, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.2445170879364014, + "learning_rate": 4.224207961007311e-06, + "loss": 1.791, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.399174690246582, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.7944, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.586184501647949, + "learning_rate": 4.220146222583266e-06, + "loss": 1.7856, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.4174695014953613, + "learning_rate": 4.218115353371243e-06, + "loss": 1.7842, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2684640884399414, + "learning_rate": 4.21608448415922e-06, + "loss": 1.7482, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 3.1782288551330566, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8286, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.4046335220336914, + "learning_rate": 4.212022745735175e-06, + "loss": 1.7228, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 3.3722083568573, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8044, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.293466329574585, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.7415, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.28663969039917, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8282, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 1.9627619981765747, + "eval_runtime": 43.6092, + "eval_samples_per_second": 22.931, + "eval_steps_per_second": 0.963, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.255870819091797, + "learning_rate": 4.203899268887084e-06, + "loss": 1.7983, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.4964773654937744, + "learning_rate": 4.201868399675061e-06, + "loss": 1.8141, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.3372740745544434, + "learning_rate": 4.199837530463039e-06, + "loss": 1.7142, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.3506968021392822, + "learning_rate": 4.197806661251016e-06, + "loss": 1.7898, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.242410898208618, + "learning_rate": 4.195775792038993e-06, + "loss": 1.7677, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.3864502906799316, + "learning_rate": 4.19374492282697e-06, + "loss": 1.7097, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.482349157333374, + "learning_rate": 4.191714053614948e-06, + "loss": 1.8052, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.317652940750122, + "learning_rate": 4.189683184402924e-06, + "loss": 1.7692, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.2777392864227295, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8246, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.3506035804748535, + "learning_rate": 4.185621445978879e-06, + "loss": 1.7576, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 1.9610412120819092, + "eval_runtime": 43.3119, + "eval_samples_per_second": 23.088, + "eval_steps_per_second": 0.97, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.3011536598205566, + "learning_rate": 4.183590576766857e-06, + "loss": 1.7891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.2614853382110596, + "learning_rate": 4.181559707554833e-06, + "loss": 1.8258, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.219175100326538, + "learning_rate": 4.179528838342811e-06, + "loss": 1.7494, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.4239423274993896, + "learning_rate": 4.177497969130788e-06, + "loss": 1.7611, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.470712423324585, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8281, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.504060745239258, + "learning_rate": 4.173436230706742e-06, + "loss": 1.7641, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.326601982116699, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.7671, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.4280030727386475, + "learning_rate": 4.169374492282697e-06, + "loss": 1.7573, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.455716371536255, + "learning_rate": 4.167343623070675e-06, + "loss": 1.809, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.3029844760894775, + "learning_rate": 4.165312753858652e-06, + "loss": 1.77, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 1.9604709148406982, + "eval_runtime": 44.4627, + "eval_samples_per_second": 22.491, + "eval_steps_per_second": 0.945, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.3833982944488525, + "learning_rate": 4.163281884646629e-06, + "loss": 1.7652, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.2392966747283936, + "learning_rate": 4.161251015434607e-06, + "loss": 1.7904, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.4546353816986084, + "learning_rate": 4.159220146222583e-06, + "loss": 1.7324, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.13592529296875, + "learning_rate": 4.157189277010561e-06, + "loss": 1.759, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.381657361984253, + "learning_rate": 4.155158407798538e-06, + "loss": 1.7871, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.4142940044403076, + "learning_rate": 4.153127538586516e-06, + "loss": 1.7832, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.305697441101074, + "learning_rate": 4.151096669374492e-06, + "loss": 1.7619, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.193429708480835, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.7738, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.3007619380950928, + "learning_rate": 4.147034930950447e-06, + "loss": 1.7839, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.6337807178497314, + "learning_rate": 4.145004061738425e-06, + "loss": 1.7748, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 1.9597856998443604, + "eval_runtime": 43.8644, + "eval_samples_per_second": 22.798, + "eval_steps_per_second": 0.957, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.4294793605804443, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.7001, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.512155532836914, + "learning_rate": 4.140942323314379e-06, + "loss": 1.7334, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.290808916091919, + "learning_rate": 4.138911454102356e-06, + "loss": 1.7522, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.483161211013794, + "learning_rate": 4.136880584890333e-06, + "loss": 1.798, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2973921298980713, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.7847, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.253084659576416, + "learning_rate": 4.132818846466288e-06, + "loss": 1.7812, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.306342840194702, + "learning_rate": 4.130787977254265e-06, + "loss": 1.7886, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.5153164863586426, + "learning_rate": 4.128757108042242e-06, + "loss": 1.7954, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.4122278690338135, + "learning_rate": 4.12672623883022e-06, + "loss": 1.8525, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.4520716667175293, + "learning_rate": 4.124695369618197e-06, + "loss": 1.7648, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 1.9547661542892456, + "eval_runtime": 45.3774, + "eval_samples_per_second": 22.037, + "eval_steps_per_second": 0.926, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2959611415863037, + "learning_rate": 4.122664500406174e-06, + "loss": 1.7626, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.2280471324920654, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.7954, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.3870065212249756, + "learning_rate": 4.118602761982129e-06, + "loss": 1.7377, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.292612075805664, + "learning_rate": 4.116571892770106e-06, + "loss": 1.7, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.4607021808624268, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8338, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.532824993133545, + "learning_rate": 4.112510154346061e-06, + "loss": 1.7549, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2139623165130615, + "learning_rate": 4.110479285134038e-06, + "loss": 1.78, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.4749462604522705, + "learning_rate": 4.108448415922015e-06, + "loss": 1.8034, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.609265089035034, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.7941, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.367753028869629, + "learning_rate": 4.10438667749797e-06, + "loss": 1.6653, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 1.9539440870285034, + "eval_runtime": 48.8438, + "eval_samples_per_second": 20.473, + "eval_steps_per_second": 0.86, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.9723567962646484, + "learning_rate": 4.102355808285947e-06, + "loss": 1.8317, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.388806104660034, + "learning_rate": 4.100324939073924e-06, + "loss": 1.7597, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4882986545562744, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.7708, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.7586395740509033, + "learning_rate": 4.096263200649879e-06, + "loss": 1.7804, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.237025260925293, + "learning_rate": 4.094232331437856e-06, + "loss": 1.7415, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.421330213546753, + "learning_rate": 4.092201462225833e-06, + "loss": 1.734, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.7281360626220703, + "learning_rate": 4.090170593013811e-06, + "loss": 1.7716, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.2200379371643066, + "learning_rate": 4.088139723801787e-06, + "loss": 1.7266, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.465440034866333, + "learning_rate": 4.086108854589765e-06, + "loss": 1.741, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.39835262298584, + "learning_rate": 4.084077985377742e-06, + "loss": 1.757, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 1.9514222145080566, + "eval_runtime": 52.8068, + "eval_samples_per_second": 18.937, + "eval_steps_per_second": 0.795, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.526684522628784, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8195, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.341954469680786, + "learning_rate": 4.080016246953696e-06, + "loss": 1.7196, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.3755977153778076, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8115, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.3952431678771973, + "learning_rate": 4.075954508529651e-06, + "loss": 1.812, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.554089069366455, + "learning_rate": 4.073923639317629e-06, + "loss": 1.8184, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.2319302558898926, + "learning_rate": 4.071892770105605e-06, + "loss": 1.7448, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.284856081008911, + "learning_rate": 4.069861900893583e-06, + "loss": 1.7154, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.4081835746765137, + "learning_rate": 4.06783103168156e-06, + "loss": 1.7311, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.6115968227386475, + "learning_rate": 4.065800162469538e-06, + "loss": 1.7436, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.368004322052002, + "learning_rate": 4.063769293257514e-06, + "loss": 1.7811, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 1.9508955478668213, + "eval_runtime": 46.8711, + "eval_samples_per_second": 21.335, + "eval_steps_per_second": 0.896, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.4607419967651367, + "learning_rate": 4.061738424045492e-06, + "loss": 1.7563, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.2978763580322266, + "learning_rate": 4.059707554833469e-06, + "loss": 1.7755, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.3316237926483154, + "learning_rate": 4.057676685621446e-06, + "loss": 1.8226, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.3699169158935547, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.7755, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.195596218109131, + "learning_rate": 4.053614947197401e-06, + "loss": 1.6916, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.4524662494659424, + "learning_rate": 4.051584077985378e-06, + "loss": 1.774, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.573888063430786, + "learning_rate": 4.049553208773355e-06, + "loss": 1.7953, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.5157253742218018, + "learning_rate": 4.047522339561333e-06, + "loss": 1.7597, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.5588185787200928, + "learning_rate": 4.04549147034931e-06, + "loss": 1.7553, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.8950047492980957, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8647, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 1.945882797241211, + "eval_runtime": 45.481, + "eval_samples_per_second": 21.987, + "eval_steps_per_second": 0.923, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.217116117477417, + "learning_rate": 4.041429731925264e-06, + "loss": 1.7465, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.5601565837860107, + "learning_rate": 4.039398862713242e-06, + "loss": 1.7575, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.4540457725524902, + "learning_rate": 4.037367993501219e-06, + "loss": 1.8291, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.327181100845337, + "learning_rate": 4.035337124289196e-06, + "loss": 1.7409, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.485457181930542, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.7192, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.4613442420959473, + "learning_rate": 4.031275385865151e-06, + "loss": 1.7613, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.303217649459839, + "learning_rate": 4.029244516653128e-06, + "loss": 1.7799, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.3830621242523193, + "learning_rate": 4.027213647441105e-06, + "loss": 1.7045, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.342332601547241, + "learning_rate": 4.025182778229083e-06, + "loss": 1.7246, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.4014711380004883, + "learning_rate": 4.02315190901706e-06, + "loss": 1.7621, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 1.95199453830719, + "eval_runtime": 45.4754, + "eval_samples_per_second": 21.99, + "eval_steps_per_second": 0.924, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.6015400886535645, + "learning_rate": 4.021121039805037e-06, + "loss": 1.7536, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.866978406906128, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8405, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.389810085296631, + "learning_rate": 4.017059301380992e-06, + "loss": 1.7365, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.280442953109741, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8273, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.509330987930298, + "learning_rate": 4.012997562956945e-06, + "loss": 1.7127, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.452078104019165, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.7457, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.368337631225586, + "learning_rate": 4.0089358245329e-06, + "loss": 1.7412, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.4529192447662354, + "learning_rate": 4.006904955320878e-06, + "loss": 1.7434, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.647761821746826, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8013, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.387424945831299, + "learning_rate": 4.002843216896833e-06, + "loss": 1.7948, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 1.946029543876648, + "eval_runtime": 43.9666, + "eval_samples_per_second": 22.745, + "eval_steps_per_second": 0.955, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.9217846393585205, + "learning_rate": 4.000812347684809e-06, + "loss": 1.7384, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.400040864944458, + "learning_rate": 3.998781478472787e-06, + "loss": 1.7496, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.4050440788269043, + "learning_rate": 3.996750609260764e-06, + "loss": 1.7165, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 3.6001529693603516, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8679, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 2.4040098190307617, + "learning_rate": 3.992688870836718e-06, + "loss": 1.7671, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.5645527839660645, + "learning_rate": 3.990658001624696e-06, + "loss": 1.7374, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.799394130706787, + "learning_rate": 3.988627132412673e-06, + "loss": 1.796, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.5481607913970947, + "learning_rate": 3.98659626320065e-06, + "loss": 1.753, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.8851330280303955, + "learning_rate": 3.984565393988627e-06, + "loss": 1.7505, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.6668694019317627, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.7604, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 1.939759612083435, + "eval_runtime": 44.5019, + "eval_samples_per_second": 22.471, + "eval_steps_per_second": 0.944, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.320935010910034, + "learning_rate": 3.980503655564582e-06, + "loss": 1.7749, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.829897880554199, + "learning_rate": 3.978472786352559e-06, + "loss": 1.7763, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.8420908451080322, + "learning_rate": 3.976441917140536e-06, + "loss": 1.7598, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.8183, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.781, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 3.9271891117095947, + "learning_rate": 3.972380178716491e-06, + "loss": 1.7949, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.277897834777832, + "learning_rate": 3.970349309504468e-06, + "loss": 1.707, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 3.241063117980957, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.7175, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.514458417892456, + "learning_rate": 3.966287571080423e-06, + "loss": 1.7764, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.420212745666504, + "learning_rate": 3.9642567018684e-06, + "loss": 1.7618, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 1.9401386976242065, + "eval_runtime": 44.1672, + "eval_samples_per_second": 22.641, + "eval_steps_per_second": 0.951, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.3980538845062256, + "learning_rate": 3.962225832656377e-06, + "loss": 1.6922, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.892646312713623, + "learning_rate": 3.960194963444355e-06, + "loss": 1.7571, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.499260902404785, + "learning_rate": 3.958164094232332e-06, + "loss": 1.7501, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.409695625305176, + "learning_rate": 3.956133225020309e-06, + "loss": 1.7553, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.640069007873535, + "learning_rate": 3.954102355808286e-06, + "loss": 1.7526, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.9101760387420654, + "learning_rate": 3.952071486596264e-06, + "loss": 1.7483, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.2579474449157715, + "learning_rate": 3.950040617384241e-06, + "loss": 1.7379, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 4.039298057556152, + "learning_rate": 3.948009748172218e-06, + "loss": 1.7988, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.3232030868530273, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.7592, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.697164297103882, + "learning_rate": 3.943948009748173e-06, + "loss": 1.7622, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 1.9367746114730835, + "eval_runtime": 43.7013, + "eval_samples_per_second": 22.883, + "eval_steps_per_second": 0.961, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.480297803878784, + "learning_rate": 3.94191714053615e-06, + "loss": 1.8006, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.3624603748321533, + "learning_rate": 3.939886271324127e-06, + "loss": 1.7152, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.417187213897705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.7077, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.6575164794921875, + "learning_rate": 3.935824532900082e-06, + "loss": 1.7744, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.4896557331085205, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8052, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.475693941116333, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.7008, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.3210372924804688, + "learning_rate": 3.929731925264013e-06, + "loss": 1.7339, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.433785915374756, + "learning_rate": 3.927701056051991e-06, + "loss": 1.7544, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 3.181574583053589, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.7229, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.704399347305298, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.7086, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 1.9335075616836548, + "eval_runtime": 47.0248, + "eval_samples_per_second": 21.265, + "eval_steps_per_second": 0.893, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 3.5027623176574707, + "learning_rate": 3.921608448415922e-06, + "loss": 1.7622, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.450943946838379, + "learning_rate": 3.9195775792039e-06, + "loss": 1.7286, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.4603374004364014, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.7725, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.8217649459838867, + "learning_rate": 3.915515840779855e-06, + "loss": 1.7859, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.3308372497558594, + "learning_rate": 3.913484971567831e-06, + "loss": 1.6874, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.624776840209961, + "learning_rate": 3.911454102355808e-06, + "loss": 1.7632, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.4097838401794434, + "learning_rate": 3.909423233143786e-06, + "loss": 1.7576, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.487410545349121, + "learning_rate": 3.907392363931763e-06, + "loss": 1.831, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.647245168685913, + "learning_rate": 3.90536149471974e-06, + "loss": 1.8448, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.397664785385132, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.868, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 1.930379033088684, + "eval_runtime": 51.1662, + "eval_samples_per_second": 19.544, + "eval_steps_per_second": 0.821, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.277433395385742, + "learning_rate": 3.901299756295695e-06, + "loss": 1.7394, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 3.1312496662139893, + "learning_rate": 3.899268887083672e-06, + "loss": 1.7422, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.5011534690856934, + "learning_rate": 3.897238017871649e-06, + "loss": 1.7484, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.8716166019439697, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.7253, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.4346914291381836, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8197, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.348475933074951, + "learning_rate": 3.891145410235581e-06, + "loss": 1.7737, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.3996007442474365, + "learning_rate": 3.889114541023558e-06, + "loss": 1.7333, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.355262279510498, + "learning_rate": 3.887083671811536e-06, + "loss": 1.6902, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.335932731628418, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8206, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.2893190383911133, + "learning_rate": 3.88302193338749e-06, + "loss": 1.7533, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 1.9290764331817627, + "eval_runtime": 46.9373, + "eval_samples_per_second": 21.305, + "eval_steps_per_second": 0.895, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.739842176437378, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.7318, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.587916374206543, + "learning_rate": 3.878960194963445e-06, + "loss": 1.7979, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.406548500061035, + "learning_rate": 3.876929325751422e-06, + "loss": 1.6899, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.440321922302246, + "learning_rate": 3.874898456539399e-06, + "loss": 1.7546, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.320030450820923, + "learning_rate": 3.872867587327377e-06, + "loss": 1.7254, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.3795316219329834, + "learning_rate": 3.870836718115354e-06, + "loss": 1.728, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.299802541732788, + "learning_rate": 3.868805848903331e-06, + "loss": 1.7464, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.469719409942627, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.765, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.612109899520874, + "learning_rate": 3.864744110479286e-06, + "loss": 1.7827, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.623173952102661, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8019, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 1.9268505573272705, + "eval_runtime": 46.77, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.4146616458892822, + "learning_rate": 3.86068237205524e-06, + "loss": 1.7498, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.361994743347168, + "learning_rate": 3.858651502843217e-06, + "loss": 1.6926, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.493215560913086, + "learning_rate": 3.856620633631195e-06, + "loss": 1.736, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.4341812133789062, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8152, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.391331195831299, + "learning_rate": 3.852558895207149e-06, + "loss": 1.7717, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.419492244720459, + "learning_rate": 3.850528025995126e-06, + "loss": 1.7283, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.349090576171875, + "learning_rate": 3.848497156783104e-06, + "loss": 1.7317, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.298647880554199, + "learning_rate": 3.84646628757108e-06, + "loss": 1.7389, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.4612667560577393, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.7333, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.3139944076538086, + "learning_rate": 3.842404549147035e-06, + "loss": 1.6944, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 1.924438714981079, + "eval_runtime": 45.8276, + "eval_samples_per_second": 21.821, + "eval_steps_per_second": 0.916, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 2.17958664894104, + "learning_rate": 3.840373679935013e-06, + "loss": 1.7474, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.337859869003296, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.7345, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4777672290802, + "learning_rate": 3.836311941510967e-06, + "loss": 1.77, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.280158519744873, + "learning_rate": 3.834281072298944e-06, + "loss": 1.7069, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2387664318084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.7889, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.373771905899048, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8268, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.4950132369995117, + "learning_rate": 3.828188464662876e-06, + "loss": 1.7586, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.3296444416046143, + "learning_rate": 3.826157595450853e-06, + "loss": 1.6991, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.6653530597686768, + "learning_rate": 3.82412672623883e-06, + "loss": 1.7505, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.485128402709961, + "learning_rate": 3.822095857026808e-06, + "loss": 1.7196, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 1.9221889972686768, + "eval_runtime": 46.6581, + "eval_samples_per_second": 21.433, + "eval_steps_per_second": 0.9, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.3726181983947754, + "learning_rate": 3.820064987814785e-06, + "loss": 1.7296, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.233992099761963, + "learning_rate": 3.818034118602762e-06, + "loss": 1.7313, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.5388147830963135, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.7435, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.4787333011627197, + "learning_rate": 3.813972380178717e-06, + "loss": 1.7733, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.232564926147461, + "learning_rate": 3.811941510966694e-06, + "loss": 1.693, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.3882386684417725, + "learning_rate": 3.809910641754671e-06, + "loss": 1.6529, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.4785983562469482, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.7456, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.685098886489868, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.7162, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.353523015975952, + "learning_rate": 3.803818034118603e-06, + "loss": 1.7852, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.342163562774658, + "learning_rate": 3.80178716490658e-06, + "loss": 1.7767, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 1.9235820770263672, + "eval_runtime": 47.1327, + "eval_samples_per_second": 21.217, + "eval_steps_per_second": 0.891, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.550818920135498, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.7328, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.3204798698425293, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.7058, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.4752719402313232, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8051, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.5202794075012207, + "learning_rate": 3.793663688058489e-06, + "loss": 1.7966, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.26521635055542, + "learning_rate": 3.791632818846467e-06, + "loss": 1.7879, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.331515312194824, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.7814, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.37178635597229, + "learning_rate": 3.787571080422421e-06, + "loss": 1.7635, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.321417808532715, + "learning_rate": 3.785540211210398e-06, + "loss": 1.7513, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.485989809036255, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.7306, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.7497379779815674, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8409, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 1.9275448322296143, + "eval_runtime": 46.2088, + "eval_samples_per_second": 21.641, + "eval_steps_per_second": 0.909, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 4.9214019775390625, + "learning_rate": 3.77944760357433e-06, + "loss": 1.7834, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 4.505277156829834, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.7179, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.5772030353546143, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.7766, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.254185438156128, + "learning_rate": 3.773354995938262e-06, + "loss": 1.7697, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 3.632233142852783, + "learning_rate": 3.771324126726239e-06, + "loss": 1.7968, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.174461841583252, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.7346, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.3983266353607178, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.6534, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 3.393791913986206, + "learning_rate": 3.765231519090171e-06, + "loss": 1.7561, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.2853987216949463, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.6963, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.649789810180664, + "learning_rate": 3.761169780666125e-06, + "loss": 1.7395, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 1.9242008924484253, + "eval_runtime": 46.6017, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 3.417872190475464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.784, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.316666841506958, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.7154, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.486161231994629, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.7903, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 4.538443565368652, + "learning_rate": 3.753046303818034e-06, + "loss": 1.7711, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.9772889614105225, + "learning_rate": 3.751015434606012e-06, + "loss": 1.8046, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 7.2353715896606445, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.7189, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.736, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.8026, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 12.551762580871582, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.7788, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 3.0898046493530273, + "learning_rate": 3.742891957757921e-06, + "loss": 1.7878, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 1.9646916389465332, + "eval_runtime": 44.1583, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 8.120022773742676, + "learning_rate": 3.740861088545898e-06, + "loss": 1.768, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 6.485774040222168, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.746, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.85135555267334, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.6968, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 5.8551435470581055, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.6796, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 4.45880126953125, + "learning_rate": 3.732737611697807e-06, + "loss": 1.7219, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 3.0435831546783447, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.6959, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.8194847106933594, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.8152, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 3.1266117095947266, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.6937, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 3.642854690551758, + "learning_rate": 3.724614134849716e-06, + "loss": 1.7992, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 3.344496965408325, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.7583, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 1.9293720722198486, + "eval_runtime": 50.8602, + "eval_samples_per_second": 19.662, + "eval_steps_per_second": 0.826, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.7666447162628174, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7441, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.8819239139556885, + "learning_rate": 3.718521527213648e-06, + "loss": 1.6314, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.6030640602111816, + "learning_rate": 3.716490658001625e-06, + "loss": 1.7389, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.705246925354004, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.7408, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 3.181328535079956, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7092, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 3.0449445247650146, + "learning_rate": 3.710398050365557e-06, + "loss": 1.7399, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.650517225265503, + "learning_rate": 3.708367181153534e-06, + "loss": 1.6861, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.7593672275543213, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.6701, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 3.0486466884613037, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.7191, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.502850294113159, + "learning_rate": 3.702274573517466e-06, + "loss": 1.6982, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 1.9228183031082153, + "eval_runtime": 49.565, + "eval_samples_per_second": 20.176, + "eval_steps_per_second": 0.847, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.400770664215088, + "learning_rate": 3.700243704305443e-06, + "loss": 1.6666, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.5583295822143555, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.7971, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.519505739212036, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.7076, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 3.106309413909912, + "learning_rate": 3.694151096669375e-06, + "loss": 1.7279, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.516906499862671, + "learning_rate": 3.692120227457352e-06, + "loss": 1.6265, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.603166103363037, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.7122, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.842773199081421, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.7312, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.8860342502593994, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7229, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.517254114151001, + "learning_rate": 3.683996750609261e-06, + "loss": 1.7342, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.430025100708008, + "learning_rate": 3.681965881397238e-06, + "loss": 1.6817, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 1.9156773090362549, + "eval_runtime": 44.6903, + "eval_samples_per_second": 22.376, + "eval_steps_per_second": 0.94, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.340075969696045, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.7044, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.4758968353271484, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.6935, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.878903865814209, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.7021, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.388366222381592, + "learning_rate": 3.673842404549147e-06, + "loss": 1.7229, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.5789167881011963, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8112, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.743072509765625, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.7239, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.5485000610351562, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.7641, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.53190541267395, + "learning_rate": 3.665718927701056e-06, + "loss": 1.6996, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.3354969024658203, + "learning_rate": 3.663688058489034e-06, + "loss": 1.6623, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.37371826171875, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.7109, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 1.9122159481048584, + "eval_runtime": 43.8844, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.406137228012085, + "learning_rate": 3.659626320064988e-06, + "loss": 1.6958, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.4220657348632812, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7196, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.6203134059906006, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.6014, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.5388472080230713, + "learning_rate": 3.65353371242892e-06, + "loss": 1.7272, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.672658681869507, + "learning_rate": 3.651502843216897e-06, + "loss": 1.6786, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4366531372070312, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7198, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.4872887134552, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.7082, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.371206283569336, + "learning_rate": 3.645410235580829e-06, + "loss": 1.6964, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.496103286743164, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.716, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.658316135406494, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.7263, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 1.9115047454833984, + "eval_runtime": 43.8788, + "eval_samples_per_second": 22.79, + "eval_steps_per_second": 0.957, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.5931882858276367, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.7586, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.4697139263153076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.7232, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.7274153232574463, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.6917, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.683138132095337, + "learning_rate": 3.633225020308692e-06, + "loss": 1.7518, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.4510817527770996, + "learning_rate": 3.63119415109667e-06, + "loss": 1.6533, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.6939656734466553, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.6545, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.572864532470703, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.6965, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.4816951751708984, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.6851, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.640394926071167, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7336, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.6707329750061035, + "learning_rate": 3.621039805036556e-06, + "loss": 1.7182, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 1.91274094581604, + "eval_runtime": 44.35, + "eval_samples_per_second": 22.548, + "eval_steps_per_second": 0.947, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.6947290897369385, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.7255, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.5496068000793457, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.6209, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.6196374893188477, + "learning_rate": 3.614947197400488e-06, + "loss": 1.7277, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.4534807205200195, + "learning_rate": 3.612916328188465e-06, + "loss": 1.7404, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.5428082942962646, + "learning_rate": 3.610885458976442e-06, + "loss": 1.6618, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.5431482791900635, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.6743, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.707956552505493, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.7483, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.5643982887268066, + "learning_rate": 3.604792851340374e-06, + "loss": 1.7304, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.4612927436828613, + "learning_rate": 3.602761982128351e-06, + "loss": 1.6561, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.3140358924865723, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.6834, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 1.9124692678451538, + "eval_runtime": 44.0522, + "eval_samples_per_second": 22.7, + "eval_steps_per_second": 0.953, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4636526107788086, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.7395, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.541180372238159, + "learning_rate": 3.596669374492283e-06, + "loss": 1.6579, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.621335983276367, + "learning_rate": 3.59463850528026e-06, + "loss": 1.7332, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.363450050354004, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.7047, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.8609447479248047, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7416, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.457338571548462, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.7516, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.3943030834198, + "learning_rate": 3.586515028432169e-06, + "loss": 1.6436, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.4719340801239014, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.6988, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3442330360412598, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.7165, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.3391900062561035, + "learning_rate": 3.580422420796101e-06, + "loss": 1.6263, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 1.9100929498672485, + "eval_runtime": 44.2504, + "eval_samples_per_second": 22.599, + "eval_steps_per_second": 0.949, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.533566474914551, + "learning_rate": 3.578391551584078e-06, + "loss": 1.7141, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.464540481567383, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.7335, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.4789979457855225, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7074, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.8252034187316895, + "learning_rate": 3.57229894394801e-06, + "loss": 1.6808, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.5870537757873535, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.7171, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.349393129348755, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.6687, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.3343207836151123, + "learning_rate": 3.566206336311942e-06, + "loss": 1.6351, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.558645248413086, + "learning_rate": 3.564175467099919e-06, + "loss": 1.708, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.7598931789398193, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.6797, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.6048543453216553, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7117, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 1.906927466392517, + "eval_runtime": 44.5189, + "eval_samples_per_second": 22.462, + "eval_steps_per_second": 0.943, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.6124815940856934, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.7328, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.3410215377807617, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.6932, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.5247764587402344, + "learning_rate": 3.554021121039805e-06, + "loss": 1.6628, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.642958164215088, + "learning_rate": 3.551990251827783e-06, + "loss": 1.7035, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.3978517055511475, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.6796, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.3969287872314453, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.6626, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.5723509788513184, + "learning_rate": 3.545897644191714e-06, + "loss": 1.7135, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.545154571533203, + "learning_rate": 3.543866774979692e-06, + "loss": 1.6626, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.501058340072632, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.7112, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.695483684539795, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.6773, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 1.9063464403152466, + "eval_runtime": 44.2034, + "eval_samples_per_second": 22.623, + "eval_steps_per_second": 0.95, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 3.0584375858306885, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.7242, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.445220947265625, + "learning_rate": 3.535743298131601e-06, + "loss": 1.7349, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.427220106124878, + "learning_rate": 3.533712428919578e-06, + "loss": 1.677, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.5148212909698486, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.7007, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.644319534301758, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.7442, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.6118004322052, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.7509, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.5224156379699707, + "learning_rate": 3.525588952071487e-06, + "loss": 1.7279, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.4668405055999756, + "learning_rate": 3.523558082859464e-06, + "loss": 1.6754, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 3.8330910205841064, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.6815, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.569058895111084, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.6518, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 1.906802773475647, + "eval_runtime": 50.8521, + "eval_samples_per_second": 19.665, + "eval_steps_per_second": 0.826, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.705265522003174, + "learning_rate": 3.517465475223396e-06, + "loss": 1.7172, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 3.2667057514190674, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7641, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 3.5119211673736572, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.7178, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.7348034381866455, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.6815, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 3.019892454147339, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.6417, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 3.2486684322357178, + "learning_rate": 3.507311129163282e-06, + "loss": 1.6888, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.897071361541748, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7121, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.7342352867126465, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.6339, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.691178560256958, + "learning_rate": 3.501218521527214e-06, + "loss": 1.7069, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.548682689666748, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7218, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 1.9029825925827026, + "eval_runtime": 48.287, + "eval_samples_per_second": 20.71, + "eval_steps_per_second": 0.87, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.422056198120117, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.6486, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.4367074966430664, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.759, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.48002290725708, + "learning_rate": 3.493095044679123e-06, + "loss": 1.5902, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.5030295848846436, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.7108, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.5892179012298584, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.6154, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.4282920360565186, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.6807, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.4826700687408447, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7284, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.416597843170166, + "learning_rate": 3.482940698619009e-06, + "loss": 1.6972, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.4664933681488037, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.7119, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.541856288909912, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.6752, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 1.8999215364456177, + "eval_runtime": 46.3528, + "eval_samples_per_second": 21.574, + "eval_steps_per_second": 0.906, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4803109169006348, + "learning_rate": 3.476848090982941e-06, + "loss": 1.776, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.4363205432891846, + "learning_rate": 3.474817221770918e-06, + "loss": 1.7196, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.4377009868621826, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.6937, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.425791025161743, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.6547, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3216712474823, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.6789, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.65527081489563, + "learning_rate": 3.466693744922827e-06, + "loss": 1.7231, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.3415145874023438, + "learning_rate": 3.464662875710805e-06, + "loss": 1.6805, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.3864822387695312, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.7153, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.490896224975586, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.6866, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.339711904525757, + "learning_rate": 3.458570268074736e-06, + "loss": 1.6732, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 1.8985264301300049, + "eval_runtime": 46.7854, + "eval_samples_per_second": 21.374, + "eval_steps_per_second": 0.898, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.393709659576416, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.6549, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.564114809036255, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.7019, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3826403617858887, + "learning_rate": 3.452477660438668e-06, + "loss": 1.6427, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.3185465335845947, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.6916, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4503891468048096, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.6793, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.386446714401245, + "learning_rate": 3.4463850528026e-06, + "loss": 1.7186, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.6404714584350586, + "learning_rate": 3.444354183590577e-06, + "loss": 1.6605, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.342374324798584, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.6219, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.7503273487091064, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.693, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.5416243076324463, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7323, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 1.9016659259796143, + "eval_runtime": 48.5456, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.5787603855133057, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.7057, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.649528980255127, + "learning_rate": 3.434199837530463e-06, + "loss": 1.6348, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.504089832305908, + "learning_rate": 3.432168968318441e-06, + "loss": 1.7202, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.5624678134918213, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7351, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.641331911087036, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.7453, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.599500894546509, + "learning_rate": 3.426076360682372e-06, + "loss": 1.6942, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.668599843978882, + "learning_rate": 3.42404549147035e-06, + "loss": 1.6337, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.75667667388916, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.6988, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.641526937484741, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.7641, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.4972548484802246, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.7269, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 1.8979382514953613, + "eval_runtime": 45.6555, + "eval_samples_per_second": 21.903, + "eval_steps_per_second": 0.92, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.5454585552215576, + "learning_rate": 3.415922014622259e-06, + "loss": 1.7294, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.5921928882598877, + "learning_rate": 3.413891145410236e-06, + "loss": 1.7145, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.4028453826904297, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.6985, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.604522705078125, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.6586, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.528970241546631, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.734, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.565030574798584, + "learning_rate": 3.405767668562145e-06, + "loss": 1.6997, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.7735602855682373, + "learning_rate": 3.403736799350122e-06, + "loss": 1.6543, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.5116474628448486, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.7711, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.4827239513397217, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.6831, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.544309377670288, + "learning_rate": 3.397644191714054e-06, + "loss": 1.6905, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 1.8945367336273193, + "eval_runtime": 46.6027, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.456636905670166, + "learning_rate": 3.395613322502031e-06, + "loss": 1.646, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.5900352001190186, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.6967, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 3.3344573974609375, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.6947, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.5621678829193115, + "learning_rate": 3.389520714865963e-06, + "loss": 1.7084, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.5676684379577637, + "learning_rate": 3.38748984565394e-06, + "loss": 1.6788, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.600691318511963, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.6811, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.5324196815490723, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.661, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.7559397220611572, + "learning_rate": 3.381397238017872e-06, + "loss": 1.7473, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.526397466659546, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.3752379417419434, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.689, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 1.8995447158813477, + "eval_runtime": 48.0083, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.5902793407440186, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.6899, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 3.157573938369751, + "learning_rate": 3.373273761169781e-06, + "loss": 1.6792, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.628035545349121, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.7003, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.4810822010040283, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.6489, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.4859938621520996, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.6768, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.6773345470428467, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7046, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.9695260524749756, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.7341, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.8146679401397705, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.7472, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.534985065460205, + "learning_rate": 3.359057676685622e-06, + "loss": 1.6199, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.9351718425750732, + "learning_rate": 3.357026807473599e-06, + "loss": 1.6656, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 1.8962808847427368, + "eval_runtime": 47.0443, + "eval_samples_per_second": 21.257, + "eval_steps_per_second": 0.893, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.4404256343841553, + "learning_rate": 3.354995938261576e-06, + "loss": 1.6207, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.6691436767578125, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.6627, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.5436172485351562, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.6919, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.7167909145355225, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.6434, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.575446367263794, + "learning_rate": 3.346872461413485e-06, + "loss": 1.6554, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.3874504566192627, + "learning_rate": 3.344841592201463e-06, + "loss": 1.6691, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.66317081451416, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.6864, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.4396398067474365, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.7217, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.5002803802490234, + "learning_rate": 3.338748984565394e-06, + "loss": 1.7556, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.679577350616455, + "learning_rate": 3.336718115353372e-06, + "loss": 1.6822, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 1.8942477703094482, + "eval_runtime": 44.3804, + "eval_samples_per_second": 22.532, + "eval_steps_per_second": 0.946, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.6119370460510254, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.662, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.6024818420410156, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.687, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 4.128854751586914, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.6898, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3910412788391113, + "learning_rate": 3.32859463850528e-06, + "loss": 1.6274, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.8630096912384033, + "learning_rate": 3.326563769293258e-06, + "loss": 1.6554, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 4.201050281524658, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.6676, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.492168664932251, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.6319, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 3.6807360649108887, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.6776, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.7271926403045654, + "learning_rate": 3.318440292445167e-06, + "loss": 1.6383, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.5567307472229004, + "learning_rate": 3.316409423233144e-06, + "loss": 1.7325, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 1.892768383026123, + "eval_runtime": 50.909, + "eval_samples_per_second": 19.643, + "eval_steps_per_second": 0.825, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.5373101234436035, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.689, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.602923631668091, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.6861, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.507110118865967, + "learning_rate": 3.310316815597076e-06, + "loss": 1.7575, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.361786127090454, + "learning_rate": 3.308285946385053e-06, + "loss": 1.6267, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5360615253448486, + "learning_rate": 3.30625507717303e-06, + "loss": 1.6911, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.6238744258880615, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.7193, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.5358500480651855, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.6691, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.7285404205322266, + "learning_rate": 3.300162469536962e-06, + "loss": 1.7249, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.346160650253296, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.6958, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.5415399074554443, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.7544, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 1.8925012350082397, + "eval_runtime": 46.4957, + "eval_samples_per_second": 21.507, + "eval_steps_per_second": 0.903, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.5924015045166016, + "learning_rate": 3.294069861900894e-06, + "loss": 1.6901, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.4649693965911865, + "learning_rate": 3.292038992688871e-06, + "loss": 1.7234, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.3648035526275635, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.7324, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.5146145820617676, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.631, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.5919511318206787, + "learning_rate": 3.285946385052803e-06, + "loss": 1.706, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.299527168273926, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.608, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.480637788772583, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.7415, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.458183526992798, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.6806, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.4864983558654785, + "learning_rate": 3.277822908204712e-06, + "loss": 1.6373, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.3432068824768066, + "learning_rate": 3.275792038992689e-06, + "loss": 1.6832, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 1.8899296522140503, + "eval_runtime": 44.3309, + "eval_samples_per_second": 22.558, + "eval_steps_per_second": 0.947, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.365891218185425, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.6492, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.463261842727661, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.6847, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.3977205753326416, + "learning_rate": 3.269699431356621e-06, + "loss": 1.6789, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.4172558784484863, + "learning_rate": 3.267668562144598e-06, + "loss": 1.6495, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.663381814956665, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.7139, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.3702306747436523, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.6876, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 3.1759073734283447, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.7465, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.6344592571258545, + "learning_rate": 3.259545085296507e-06, + "loss": 1.6339, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.364490032196045, + "learning_rate": 3.257514216084484e-06, + "loss": 1.7394, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.5639748573303223, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.6928, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 1.8937300443649292, + "eval_runtime": 43.9356, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 3.152968645095825, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.6693, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.45776104927063, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.6999, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.5289807319641113, + "learning_rate": 3.249390739236393e-06, + "loss": 1.6523, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.4168646335601807, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.6823, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.5115208625793457, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.6816, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.685535192489624, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.6307, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.253084897994995, + "learning_rate": 3.241267262388302e-06, + "loss": 1.662, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.4825408458709717, + "learning_rate": 3.23923639317628e-06, + "loss": 1.6444, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.691168785095215, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.6912, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.978756904602051, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7536, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 1.8903727531433105, + "eval_runtime": 43.4099, + "eval_samples_per_second": 23.036, + "eval_steps_per_second": 0.968, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.317105531692505, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.6441, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 3.1030166149139404, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.7205, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.595792770385742, + "learning_rate": 3.229082047116166e-06, + "loss": 1.677, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.3315582275390625, + "learning_rate": 3.227051177904143e-06, + "loss": 1.6779, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.4262566566467285, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.7003, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.4438798427581787, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.6081, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.573531150817871, + "learning_rate": 3.220958570268075e-06, + "loss": 1.6489, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.477809429168701, + "learning_rate": 3.218927701056052e-06, + "loss": 1.6025, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.5499629974365234, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.7337, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.4699153900146484, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.6443, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 1.8867121934890747, + "eval_runtime": 42.9343, + "eval_samples_per_second": 23.291, + "eval_steps_per_second": 0.978, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.4200708866119385, + "learning_rate": 3.212835093419984e-06, + "loss": 1.6743, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.418069839477539, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.723, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.452284336090088, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.6358, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.4241347312927246, + "learning_rate": 3.206742485783916e-06, + "loss": 1.638, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.461860418319702, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.7017, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.591986894607544, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7065, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.478470802307129, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.6262, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.49831223487854, + "learning_rate": 3.198619008935825e-06, + "loss": 1.6871, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.485332727432251, + "learning_rate": 3.196588139723802e-06, + "loss": 1.6864, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.533710479736328, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.6794, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 1.88768470287323, + "eval_runtime": 43.5948, + "eval_samples_per_second": 22.938, + "eval_steps_per_second": 0.963, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.368609666824341, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.6877, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.4875223636627197, + "learning_rate": 3.190495532087734e-06, + "loss": 1.7406, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.5075156688690186, + "learning_rate": 3.188464662875711e-06, + "loss": 1.6545, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.428731918334961, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.6213, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.467618465423584, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.6263, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.561612129211426, + "learning_rate": 3.182372055239643e-06, + "loss": 1.6987, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.5035271644592285, + "learning_rate": 3.18034118602762e-06, + "loss": 1.6902, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.538311243057251, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.6523, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.486321210861206, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.6665, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.433173894882202, + "learning_rate": 3.174248578391552e-06, + "loss": 1.6605, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 1.8853652477264404, + "eval_runtime": 43.3417, + "eval_samples_per_second": 23.072, + "eval_steps_per_second": 0.969, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.4658470153808594, + "learning_rate": 3.172217709179529e-06, + "loss": 1.6615, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.452450752258301, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.6332, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.5310161113739014, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.6846, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.4504501819610596, + "learning_rate": 3.166125101543461e-06, + "loss": 1.6996, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.6004488468170166, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.6302, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.5000431537628174, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.6742, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.4869067668914795, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.6797, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.5018959045410156, + "learning_rate": 3.15800162469537e-06, + "loss": 1.673, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.6645195484161377, + "learning_rate": 3.155970755483347e-06, + "loss": 1.6908, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.3708560466766357, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.683, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 1.8829350471496582, + "eval_runtime": 43.5383, + "eval_samples_per_second": 22.968, + "eval_steps_per_second": 0.965, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.574108362197876, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.7155, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.4669814109802246, + "learning_rate": 3.149878147847279e-06, + "loss": 1.7529, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4713149070739746, + "learning_rate": 3.147847278635256e-06, + "loss": 1.6491, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.5128800868988037, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.6276, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.740370035171509, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.6778, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.6038050651550293, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.6527, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.782407283782959, + "learning_rate": 3.139723801787165e-06, + "loss": 1.6989, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.7673134803771973, + "learning_rate": 3.137692932575143e-06, + "loss": 1.6291, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.7472870349884033, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.6776, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.4887819290161133, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7009, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.8792160749435425, + "eval_runtime": 43.7334, + "eval_samples_per_second": 22.866, + "eval_steps_per_second": 0.96, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.4528491497039795, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.7245, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.5177690982818604, + "learning_rate": 3.129569455727051e-06, + "loss": 1.6852, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.9238510131835938, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.6994, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.446660041809082, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.6636, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.644211530685425, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.6983, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.697342872619629, + "learning_rate": 3.12144597887896e-06, + "loss": 1.7172, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.714172124862671, + "learning_rate": 3.119415109666938e-06, + "loss": 1.6702, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.471735715866089, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.6945, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.5089669227600098, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.7181, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.349229335784912, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.7297, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.880064845085144, + "eval_runtime": 47.0636, + "eval_samples_per_second": 21.248, + "eval_steps_per_second": 0.892, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.6843974590301514, + "learning_rate": 3.111291632818847e-06, + "loss": 1.6698, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.4325790405273438, + "learning_rate": 3.109260763606824e-06, + "loss": 1.6973, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.488459825515747, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.6552, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.4992082118988037, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.7114, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.4742603302001953, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.7073, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.556330442428589, + "learning_rate": 3.101137286758733e-06, + "loss": 1.7108, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.5711283683776855, + "learning_rate": 3.09910641754671e-06, + "loss": 1.6811, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.6467158794403076, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.7395, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.465963363647461, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.7006, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.622716188430786, + "learning_rate": 3.093013809910642e-06, + "loss": 1.7385, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.8808013200759888, + "eval_runtime": 45.5126, + "eval_samples_per_second": 21.972, + "eval_steps_per_second": 0.923, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.489701509475708, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.6949, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.4398739337921143, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.6802, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.3654873371124268, + "learning_rate": 3.086921202274574e-06, + "loss": 1.6452, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.518768548965454, + "learning_rate": 3.084890333062551e-06, + "loss": 1.6913, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.63173246383667, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.6611, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.531484603881836, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.611, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.4796738624572754, + "learning_rate": 3.078797725426483e-06, + "loss": 1.6764, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.39241099357605, + "learning_rate": 3.07676685621446e-06, + "loss": 1.6417, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3984735012054443, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.6546, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.5202760696411133, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.6879, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.879974365234375, + "eval_runtime": 44.395, + "eval_samples_per_second": 22.525, + "eval_steps_per_second": 0.946, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.5556983947753906, + "learning_rate": 3.070674248578392e-06, + "loss": 1.6733, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.3853955268859863, + "learning_rate": 3.068643379366369e-06, + "loss": 1.6274, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.4503982067108154, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.6896, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.5618021488189697, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.6592, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.6265292167663574, + "learning_rate": 3.062550771730301e-06, + "loss": 1.6634, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.4617390632629395, + "learning_rate": 3.060519902518278e-06, + "loss": 1.6032, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.4191603660583496, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.6871, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.5509119033813477, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.653, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.5399646759033203, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.7046, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.446967124938965, + "learning_rate": 3.052396425670187e-06, + "loss": 1.6547, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.8803874254226685, + "eval_runtime": 45.361, + "eval_samples_per_second": 22.045, + "eval_steps_per_second": 0.926, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.419206142425537, + "learning_rate": 3.050365556458164e-06, + "loss": 1.6517, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.4592065811157227, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.693, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.5429770946502686, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7124, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.5141642093658447, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.7103, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.362539291381836, + "learning_rate": 3.042242079610073e-06, + "loss": 1.6205, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.401175022125244, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.6479, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.5883190631866455, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.6732, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.5273513793945312, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.6673, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.6569600105285645, + "learning_rate": 3.034118602761982e-06, + "loss": 1.6396, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.8843557834625244, + "learning_rate": 3.03208773354996e-06, + "loss": 1.7136, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.8763896226882935, + "eval_runtime": 46.1886, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.6459810733795166, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.6939, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.4619321823120117, + "learning_rate": 3.028025995125914e-06, + "loss": 1.6197, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.390430450439453, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.6805, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.5494728088378906, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.6116, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.5397655963897705, + "learning_rate": 3.021933387489846e-06, + "loss": 1.6753, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.634716510772705, + "learning_rate": 3.019902518277823e-06, + "loss": 1.6892, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.5844202041625977, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.6401, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.577362060546875, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.6405, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.3554489612579346, + "learning_rate": 3.013809910641755e-06, + "loss": 1.6427, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.893918037414551, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.6739, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.8772739171981812, + "eval_runtime": 44.9972, + "eval_samples_per_second": 22.224, + "eval_steps_per_second": 0.933, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.4818382263183594, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.644, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.4990177154541016, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.6962, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.627502918243408, + "learning_rate": 3.005686433793664e-06, + "loss": 1.639, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.5949175357818604, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.7095, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.422765016555786, + "learning_rate": 3.001624695369618e-06, + "loss": 1.6572, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.708702564239502, + "learning_rate": 2.999593826157596e-06, + "loss": 1.6747, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.624866247177124, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.6643, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.5831358432769775, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.7036, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.5799436569213867, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.6628, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.7047722339630127, + "learning_rate": 2.991470349309505e-06, + "loss": 1.6591, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.8736577033996582, + "eval_runtime": 44.7095, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.411628246307373, + "learning_rate": 2.989439480097482e-06, + "loss": 1.6644, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.8774638175964355, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.7192, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.857028007507324, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7079, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.4619641304016113, + "learning_rate": 2.983346872461414e-06, + "loss": 1.6846, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.6566193103790283, + "learning_rate": 2.981316003249391e-06, + "loss": 1.6768, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.681217670440674, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.7002, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.39505672454834, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.65, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.6062734127044678, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.6659, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.6402504444122314, + "learning_rate": 2.9731925264013e-06, + "loss": 1.6491, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.497014284133911, + "learning_rate": 2.971161657189277e-06, + "loss": 1.6383, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.874670147895813, + "eval_runtime": 42.9696, + "eval_samples_per_second": 23.272, + "eval_steps_per_second": 0.977, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.9070584774017334, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.6388, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.580744981765747, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7078, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.4637644290924072, + "learning_rate": 2.965069049553209e-06, + "loss": 1.6266, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.4489941596984863, + "learning_rate": 2.963038180341186e-06, + "loss": 1.705, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.664654016494751, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.6964, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.7017202377319336, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.6356, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.3691024780273438, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.6566, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.540384292602539, + "learning_rate": 2.954914703493095e-06, + "loss": 1.6334, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.5240535736083984, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.6316, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 3.105684518814087, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.6769, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.8741132020950317, + "eval_runtime": 42.6795, + "eval_samples_per_second": 23.43, + "eval_steps_per_second": 0.984, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.494252920150757, + "learning_rate": 2.948822095857027e-06, + "loss": 1.6359, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 4.069019317626953, + "learning_rate": 2.946791226645004e-06, + "loss": 1.621, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.641447067260742, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.6502, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.6700453758239746, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.6593, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.686068296432495, + "learning_rate": 2.940698619008936e-06, + "loss": 1.67, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.409055471420288, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.6809, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.4707298278808594, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.69, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.4735982418060303, + "learning_rate": 2.934606011372868e-06, + "loss": 1.666, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.633183479309082, + "learning_rate": 2.932575142160845e-06, + "loss": 1.6597, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.4808642864227295, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.6667, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.8731157779693604, + "eval_runtime": 41.3423, + "eval_samples_per_second": 24.188, + "eval_steps_per_second": 1.016, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.4044833183288574, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.7028, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.942699909210205, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.6678, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.505645990371704, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.5972, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.700967788696289, + "learning_rate": 2.922420796100731e-06, + "loss": 1.7064, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.4805755615234375, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.6201, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.753181219100952, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.6704, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.488865613937378, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7057, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.5464885234832764, + "learning_rate": 2.91429731925264e-06, + "loss": 1.6678, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.4997060298919678, + "learning_rate": 2.912266450040618e-06, + "loss": 1.662, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.5162436962127686, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.6501, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.869848370552063, + "eval_runtime": 43.3583, + "eval_samples_per_second": 23.064, + "eval_steps_per_second": 0.969, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.616129159927368, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.6206, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.761080503463745, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7131, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.5994064807891846, + "learning_rate": 2.904142973192527e-06, + "loss": 1.6693, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 3.4927968978881836, + "learning_rate": 2.902112103980504e-06, + "loss": 1.6694, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.6050939559936523, + "learning_rate": 2.900081234768481e-06, + "loss": 1.6419, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.796086311340332, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.6354, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.5658411979675293, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.652, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.3802385330200195, + "learning_rate": 2.893988627132413e-06, + "loss": 1.5902, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.479189872741699, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.6617, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.4367568492889404, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.6691, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.868497371673584, + "eval_runtime": 45.172, + "eval_samples_per_second": 22.138, + "eval_steps_per_second": 0.93, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.51888370513916, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.6872, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.5192768573760986, + "learning_rate": 2.885865150284322e-06, + "loss": 1.6584, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.5898501873016357, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.6179, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.698101043701172, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.663, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.889942169189453, + "learning_rate": 2.879772542648254e-06, + "loss": 1.5954, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.7909929752349854, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.7264, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 3.1931703090667725, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7091, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.7934632301330566, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.7204, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.470545530319214, + "learning_rate": 2.871649065800163e-06, + "loss": 1.644, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 3.011662006378174, + "learning_rate": 2.86961819658814e-06, + "loss": 1.6743, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.8678737878799438, + "eval_runtime": 44.7875, + "eval_samples_per_second": 22.328, + "eval_steps_per_second": 0.938, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.60823655128479, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.6467, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.5347230434417725, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.7044, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.481715440750122, + "learning_rate": 2.863525588952072e-06, + "loss": 1.5969, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.4966487884521484, + "learning_rate": 2.861494719740049e-06, + "loss": 1.6231, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.8809802532196045, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.6562, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.570204734802246, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.6511, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.6073813438415527, + "learning_rate": 2.855402112103981e-06, + "loss": 1.6456, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.8125052452087402, + "learning_rate": 2.853371242891958e-06, + "loss": 1.6722, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 5.341272354125977, + "learning_rate": 2.851340373679935e-06, + "loss": 1.7025, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 3.138002395629883, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.6676, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.8733265399932861, + "eval_runtime": 45.4915, + "eval_samples_per_second": 21.982, + "eval_steps_per_second": 0.923, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 7.264789581298828, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.6335, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 3.8533623218536377, + "learning_rate": 2.845247766043867e-06, + "loss": 1.6029, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 4.1789164543151855, + "learning_rate": 2.843216896831844e-06, + "loss": 1.6368, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 3.117408037185669, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.6924, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.721306085586548, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.6914, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.7099668979644775, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.6508, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 3.7823967933654785, + "learning_rate": 2.835093419983753e-06, + "loss": 1.636, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.696370840072632, + "learning_rate": 2.833062550771731e-06, + "loss": 1.6175, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.7303690910339355, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.6654, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.867727041244507, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.6717, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.8720332384109497, + "eval_runtime": 44.164, + "eval_samples_per_second": 22.643, + "eval_steps_per_second": 0.951, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 3.1209630966186523, + "learning_rate": 2.826969943135662e-06, + "loss": 1.6635, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.5057783126831055, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.6309, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.6650168895721436, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.6365, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.371305465698242, + "learning_rate": 2.820877335499594e-06, + "loss": 1.603, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.387404203414917, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.6189, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.6265947818756104, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.6511, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.6108627319335938, + "learning_rate": 2.814784727863526e-06, + "loss": 1.6727, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.7021450996398926, + "learning_rate": 2.812753858651503e-06, + "loss": 1.6334, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.470118999481201, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.6155, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5722644329071045, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.6788, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.8727209568023682, + "eval_runtime": 44.6253, + "eval_samples_per_second": 22.409, + "eval_steps_per_second": 0.941, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.7143876552581787, + "learning_rate": 2.806661251015435e-06, + "loss": 1.6956, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.8522002696990967, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.7558, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.5352110862731934, + "learning_rate": 2.802599512591389e-06, + "loss": 1.6592, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.7065908908843994, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.6121, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.616450786590576, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.6643, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.528763771057129, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7104, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 3.725569009780884, + "learning_rate": 2.794476035743298e-06, + "loss": 1.6713, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.487497091293335, + "learning_rate": 2.792445166531276e-06, + "loss": 1.692, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.534766912460327, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.6257, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.8553240299224854, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.6147, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.8675988912582397, + "eval_runtime": 46.4445, + "eval_samples_per_second": 21.531, + "eval_steps_per_second": 0.904, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.5054166316986084, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.6082, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.4133870601654053, + "learning_rate": 2.784321689683185e-06, + "loss": 1.6148, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.5863375663757324, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7032, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.58491849899292, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.679, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.5225989818573, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.6688, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.366861343383789, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.7076, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.4159810543060303, + "learning_rate": 2.774167343623071e-06, + "loss": 1.6301, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.5394275188446045, + "learning_rate": 2.772136474411048e-06, + "loss": 1.5758, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.409821033477783, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.664, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.4548892974853516, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.643, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.8672850131988525, + "eval_runtime": 46.3149, + "eval_samples_per_second": 21.591, + "eval_steps_per_second": 0.907, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.7483904361724854, + "learning_rate": 2.76604386677498e-06, + "loss": 1.6898, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.538804769515991, + "learning_rate": 2.764012997562957e-06, + "loss": 1.6019, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.7153313159942627, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.691, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.6892242431640625, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.6386, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.550957679748535, + "learning_rate": 2.757920389926889e-06, + "loss": 1.6237, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.3658559322357178, + "learning_rate": 2.755889520714866e-06, + "loss": 1.6015, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.6524102687835693, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.6644, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.7067437171936035, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.687, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.4044861793518066, + "learning_rate": 2.749796913078798e-06, + "loss": 1.6535, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.4933829307556152, + "learning_rate": 2.747766043866775e-06, + "loss": 1.7017, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.8652688264846802, + "eval_runtime": 46.8502, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.7062129974365234, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7398, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.5704996585845947, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.6561, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.5875790119171143, + "learning_rate": 2.741673436230707e-06, + "loss": 1.643, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.7498223781585693, + "learning_rate": 2.739642567018684e-06, + "loss": 1.6652, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.7275257110595703, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.6559, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.6349947452545166, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.6489, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.7058634757995605, + "learning_rate": 2.733549959382616e-06, + "loss": 1.7176, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.481344223022461, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.685, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.468277931213379, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.6305, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.6183228492736816, + "learning_rate": 2.727457351746548e-06, + "loss": 1.6151, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.864329218864441, + "eval_runtime": 43.0443, + "eval_samples_per_second": 23.232, + "eval_steps_per_second": 0.976, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.50679349899292, + "learning_rate": 2.725426482534525e-06, + "loss": 1.6468, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.6565401554107666, + "learning_rate": 2.723395613322502e-06, + "loss": 1.6496, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.5332367420196533, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.6424, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.6645619869232178, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.8012, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.7359142303466797, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.6025, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.660728693008423, + "learning_rate": 2.715272136474411e-06, + "loss": 1.6601, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.5184221267700195, + "learning_rate": 2.713241267262389e-06, + "loss": 1.6383, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.643712043762207, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.6834, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.504753351211548, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.6292, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.3439948558807373, + "learning_rate": 2.70714865962632e-06, + "loss": 1.6558, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.861581563949585, + "eval_runtime": 43.9224, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4820125102996826, + "learning_rate": 2.705117790414298e-06, + "loss": 1.69, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.6156861782073975, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.6304, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.490123987197876, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.6573, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.4439024925231934, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.6368, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.891603708267212, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7111, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.481175661087036, + "learning_rate": 2.694963444354184e-06, + "loss": 1.6408, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.5378847122192383, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.6782, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.592477321624756, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.6085, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.9336471557617188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.6719, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.3115012645721436, + "learning_rate": 2.686839967506093e-06, + "loss": 1.5797, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.8649041652679443, + "eval_runtime": 43.711, + "eval_samples_per_second": 22.878, + "eval_steps_per_second": 0.961, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.627028465270996, + "learning_rate": 2.68480909829407e-06, + "loss": 1.5784, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.917670488357544, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7007, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.418285608291626, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.6613, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.607773780822754, + "learning_rate": 2.678716490658002e-06, + "loss": 1.6561, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.569714307785034, + "learning_rate": 2.676685621445979e-06, + "loss": 1.6541, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.359494209289551, + "learning_rate": 2.674654752233956e-06, + "loss": 1.6467, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.5155694484710693, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.7003, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.439147710800171, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.5679, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.5044116973876953, + "learning_rate": 2.668562144597888e-06, + "loss": 1.6626, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.4475765228271484, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.656, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.8636877536773682, + "eval_runtime": 43.5138, + "eval_samples_per_second": 22.981, + "eval_steps_per_second": 0.965, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4616832733154297, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.6101, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.4125123023986816, + "learning_rate": 2.66246953696182e-06, + "loss": 1.6021, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.571326494216919, + "learning_rate": 2.660438667749797e-06, + "loss": 1.6482, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.424332618713379, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.653, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.3034703731536865, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.6096, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.4758505821228027, + "learning_rate": 2.654346060113729e-06, + "loss": 1.6274, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 3.800152540206909, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.6217, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.479827880859375, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7202, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.604523181915283, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.6778, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 3.595973491668701, + "learning_rate": 2.646222583265638e-06, + "loss": 1.6678, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.8647090196609497, + "eval_runtime": 43.0415, + "eval_samples_per_second": 23.233, + "eval_steps_per_second": 0.976, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 3.297319173812866, + "learning_rate": 2.644191714053615e-06, + "loss": 1.637, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.423567056655884, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.6018, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 4.4547953605651855, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.6594, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 3.0775582790374756, + "learning_rate": 2.638099106417547e-06, + "loss": 1.6259, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.747938871383667, + "learning_rate": 2.636068237205524e-06, + "loss": 1.7208, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.5350255966186523, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.5705, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 3.0630743503570557, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.6507, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 3.5475759506225586, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7045, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.534303903579712, + "learning_rate": 2.627944760357433e-06, + "loss": 1.7173, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.6964452266693115, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.6823, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.865207552909851, + "eval_runtime": 42.9435, + "eval_samples_per_second": 23.286, + "eval_steps_per_second": 0.978, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.924957752227783, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.6036, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.7058141231536865, + "learning_rate": 2.621852152721365e-06, + "loss": 1.6701, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.735339641571045, + "learning_rate": 2.619821283509342e-06, + "loss": 1.607, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.573958396911621, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.667, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5776426792144775, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.6938, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 4.454444408416748, + "learning_rate": 2.613728675873274e-06, + "loss": 1.7139, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.4883337020874023, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.6524, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.672116994857788, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.6197, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 3.511932611465454, + "learning_rate": 2.607636068237206e-06, + "loss": 1.6739, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.5960047245025635, + "learning_rate": 2.605605199025183e-06, + "loss": 1.6516, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.8604435920715332, + "eval_runtime": 42.9247, + "eval_samples_per_second": 23.297, + "eval_steps_per_second": 0.978, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.565376043319702, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.6884, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.676114797592163, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.6693, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.582751750946045, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.6499, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.748908281326294, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.6837, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.5578067302703857, + "learning_rate": 2.595450852965069e-06, + "loss": 1.6959, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.5702805519104004, + "learning_rate": 2.593419983753047e-06, + "loss": 1.6429, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.4748787879943848, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.6878, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.6157331466674805, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.6246, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.8377959728240967, + "learning_rate": 2.587327376116978e-06, + "loss": 1.6451, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.485398292541504, + "learning_rate": 2.585296506904956e-06, + "loss": 1.6288, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.8593864440917969, + "eval_runtime": 44.6341, + "eval_samples_per_second": 22.404, + "eval_steps_per_second": 0.941, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.46850323677063, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.5766, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.673283576965332, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7383, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.935293197631836, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.5742, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.531733751296997, + "learning_rate": 2.577173030056865e-06, + "loss": 1.6714, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.528425455093384, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6802, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.5644803047180176, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.6201, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.382451295852661, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.5824, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.63320255279541, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.6685, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.488844394683838, + "learning_rate": 2.567018683996751e-06, + "loss": 1.6078, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.7845425605773926, + "learning_rate": 2.564987814784728e-06, + "loss": 1.638, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.8572624921798706, + "eval_runtime": 43.2181, + "eval_samples_per_second": 23.138, + "eval_steps_per_second": 0.972, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.709656000137329, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.6425, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.651277542114258, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7401, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.5424256324768066, + "learning_rate": 2.55889520714866e-06, + "loss": 1.6272, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.6508235931396484, + "learning_rate": 2.556864337936637e-06, + "loss": 1.6384, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.882636308670044, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.6965, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.518367290496826, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.6551, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.7977044582366943, + "learning_rate": 2.550771730300569e-06, + "loss": 1.679, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.353041172027588, + "learning_rate": 2.548740861088546e-06, + "loss": 1.5745, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.5718655586242676, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.5655, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.6955313682556152, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.6764, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.8569272756576538, + "eval_runtime": 45.3844, + "eval_samples_per_second": 22.034, + "eval_steps_per_second": 0.925, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4756696224212646, + "learning_rate": 2.542648253452478e-06, + "loss": 1.6522, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.573193311691284, + "learning_rate": 2.540617384240455e-06, + "loss": 1.6079, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.5967671871185303, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.6341, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.7896342277526855, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.6449, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.519728183746338, + "learning_rate": 2.534524776604387e-06, + "loss": 1.549, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.543139934539795, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.6335, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.5580177307128906, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6402, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.443547010421753, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.6421, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.7075212001800537, + "learning_rate": 2.526401299756296e-06, + "loss": 1.6283, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.6505324840545654, + "learning_rate": 2.524370430544273e-06, + "loss": 1.6028, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.8556184768676758, + "eval_runtime": 43.731, + "eval_samples_per_second": 22.867, + "eval_steps_per_second": 0.96, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.597172260284424, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.6631, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 3.1529037952423096, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.6219, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.5374975204467773, + "learning_rate": 2.518277822908205e-06, + "loss": 1.6476, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 3.0557401180267334, + "learning_rate": 2.516246953696182e-06, + "loss": 1.6627, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 6.759350776672363, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.6532, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.670427083969116, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.658, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 4.269686698913574, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.5873, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 13.37558650970459, + "learning_rate": 2.508123476848091e-06, + "loss": 1.6872, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 7.060877323150635, + "learning_rate": 2.506092607636069e-06, + "loss": 1.6406, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 11.259824752807617, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.6421, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.861360788345337, + "eval_runtime": 45.4438, + "eval_samples_per_second": 22.005, + "eval_steps_per_second": 0.924, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 5.861843585968018, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.701, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 3.9136719703674316, + "learning_rate": 2.5e-06, + "loss": 1.6289, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 12.720739364624023, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.6532, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 5.970798492431641, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7035, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 3.490950584411621, + "learning_rate": 2.493907392363932e-06, + "loss": 1.6565, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 5.826146602630615, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.655, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 9.610876083374023, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.7295, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 3.754833221435547, + "learning_rate": 2.487814784727864e-06, + "loss": 1.6406, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.7952232360839844, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.6463, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 6.3178558349609375, + "learning_rate": 2.483753046303818e-06, + "loss": 1.6474, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.8672106266021729, + "eval_runtime": 50.0163, + "eval_samples_per_second": 19.993, + "eval_steps_per_second": 0.84, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 7.4854960441589355, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.6798, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.8903026580810547, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.667, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.481947898864746, + "learning_rate": 2.47766043866775e-06, + "loss": 1.6677, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.8459818363189697, + "learning_rate": 2.475629569455727e-06, + "loss": 1.6129, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 5.00568962097168, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.6233, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 3.7508153915405273, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.6414, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.7895889282226562, + "learning_rate": 2.469536961819659e-06, + "loss": 1.65, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.532818555831909, + "learning_rate": 2.467506092607636e-06, + "loss": 1.6174, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.699639320373535, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7128, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.5636072158813477, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.6416, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.8608473539352417, + "eval_runtime": 43.8315, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.7466976642608643, + "learning_rate": 2.461413484971568e-06, + "loss": 1.7246, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.690990686416626, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.596, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.868340492248535, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.622, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.4456427097320557, + "learning_rate": 2.4553208773355e-06, + "loss": 1.6446, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.5499041080474854, + "learning_rate": 2.453290008123477e-06, + "loss": 1.6587, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.68770170211792, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.6612, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.589557647705078, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.6428, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.7014052867889404, + "learning_rate": 2.447197400487409e-06, + "loss": 1.6605, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.6242196559906006, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.6897, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.5285110473632812, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.6636, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.8544306755065918, + "eval_runtime": 45.6109, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.484952449798584, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.6636, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.5793070793151855, + "learning_rate": 2.439073923639318e-06, + "loss": 1.6044, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.5558605194091797, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.5885, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.6365678310394287, + "learning_rate": 2.435012185215272e-06, + "loss": 1.6236, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.428422451019287, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.6446, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.535776138305664, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.6252, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.743934154510498, + "learning_rate": 2.428919577579204e-06, + "loss": 1.729, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.4823739528656006, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.6251, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.4650003910064697, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.6191, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.4338109493255615, + "learning_rate": 2.422826969943136e-06, + "loss": 1.5919, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.8501240015029907, + "eval_runtime": 43.8558, + "eval_samples_per_second": 22.802, + "eval_steps_per_second": 0.958, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.6054813861846924, + "learning_rate": 2.420796100731113e-06, + "loss": 1.6485, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.390143394470215, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.661, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.467470645904541, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.6308, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.360908269882202, + "learning_rate": 2.414703493095045e-06, + "loss": 1.6394, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.5966176986694336, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.6138, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.528402328491211, + "learning_rate": 2.410641754670999e-06, + "loss": 1.6817, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.5319254398345947, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.6189, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.559260129928589, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.6476, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.5595016479492188, + "learning_rate": 2.404549147034931e-06, + "loss": 1.6878, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.488196849822998, + "learning_rate": 2.402518277822908e-06, + "loss": 1.6095, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.8484610319137573, + "eval_runtime": 43.4829, + "eval_samples_per_second": 22.998, + "eval_steps_per_second": 0.966, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6928117275238037, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.6533, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.485538959503174, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.6174, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.4725208282470703, + "learning_rate": 2.39642567018684e-06, + "loss": 1.6306, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.456028461456299, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.6586, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3731250762939453, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.6009, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.4567086696624756, + "learning_rate": 2.390333062550772e-06, + "loss": 1.5257, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.577141284942627, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7127, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.4612233638763428, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.638, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.463212728500366, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.5933, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.5121495723724365, + "learning_rate": 2.382209585702681e-06, + "loss": 1.68, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.8499811887741089, + "eval_runtime": 45.066, + "eval_samples_per_second": 22.19, + "eval_steps_per_second": 0.932, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.658245801925659, + "learning_rate": 2.380178716490658e-06, + "loss": 1.6354, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.6265509128570557, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.6755, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 3.016807794570923, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.6764, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.6552720069885254, + "learning_rate": 2.37408610885459e-06, + "loss": 1.6281, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.520078659057617, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.7077, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.528329849243164, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.6494, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.623776912689209, + "learning_rate": 2.367993501218522e-06, + "loss": 1.6436, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.56669020652771, + "learning_rate": 2.365962632006499e-06, + "loss": 1.6762, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.5415968894958496, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.6711, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.5164198875427246, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.6725, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.8490949869155884, + "eval_runtime": 43.612, + "eval_samples_per_second": 22.929, + "eval_steps_per_second": 0.963, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.626594066619873, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.6637, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3480069637298584, + "learning_rate": 2.357839155158408e-06, + "loss": 1.6561, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.644103527069092, + "learning_rate": 2.355808285946385e-06, + "loss": 1.6763, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.5539023876190186, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.653, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.5055317878723145, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.6643, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.4204092025756836, + "learning_rate": 2.349715678310317e-06, + "loss": 1.6133, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.5867843627929688, + "learning_rate": 2.347684809098294e-06, + "loss": 1.6804, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.5993685722351074, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.6101, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.565439462661743, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.6221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.5296084880828857, + "learning_rate": 2.341592201462226e-06, + "loss": 1.6234, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.8482695817947388, + "eval_runtime": 45.1972, + "eval_samples_per_second": 22.125, + "eval_steps_per_second": 0.929, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.6142499446868896, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.5886, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.486616849899292, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.6097, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.5778701305389404, + "learning_rate": 2.335499593826158e-06, + "loss": 1.6585, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.6165313720703125, + "learning_rate": 2.333468724614135e-06, + "loss": 1.663, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.4976565837860107, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.6083, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.485017776489258, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.6006, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.620819330215454, + "learning_rate": 2.327376116978067e-06, + "loss": 1.6641, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.528599262237549, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.6128, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.541442632675171, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.6545, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.4722139835357666, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.6146, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.848878264427185, + "eval_runtime": 43.7961, + "eval_samples_per_second": 22.833, + "eval_steps_per_second": 0.959, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.6106903553009033, + "learning_rate": 2.319252640129976e-06, + "loss": 1.7086, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.5255045890808105, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.6738, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.43829083442688, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.5861, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.527411460876465, + "learning_rate": 2.313160032493908e-06, + "loss": 1.6634, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.4998528957366943, + "learning_rate": 2.311129163281885e-06, + "loss": 1.6187, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.4838368892669678, + "learning_rate": 2.309098294069862e-06, + "loss": 1.5999, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.466900110244751, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.6446, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.583103895187378, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.6658, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.478386163711548, + "learning_rate": 2.303005686433794e-06, + "loss": 1.6261, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.5724942684173584, + "learning_rate": 2.300974817221771e-06, + "loss": 1.5955, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.8444501161575317, + "eval_runtime": 50.0263, + "eval_samples_per_second": 19.989, + "eval_steps_per_second": 0.84, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.601644277572632, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.6493, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.5638768672943115, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.6702, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.4802567958831787, + "learning_rate": 2.294882209585703e-06, + "loss": 1.6036, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.5114285945892334, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.642, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.487563133239746, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.673, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.6014463901519775, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.6261, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.53680157661438, + "learning_rate": 2.286758732737612e-06, + "loss": 1.6364, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.764408588409424, + "learning_rate": 2.284727863525589e-06, + "loss": 1.6791, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.8998825550079346, + "learning_rate": 2.282696994313566e-06, + "loss": 1.644, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.5797383785247803, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.6267, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.8467868566513062, + "eval_runtime": 49.4778, + "eval_samples_per_second": 20.211, + "eval_steps_per_second": 0.849, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.447178602218628, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.5955, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.6275408267974854, + "learning_rate": 2.276604386677498e-06, + "loss": 1.6117, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.6952037811279297, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.6894, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.5299718379974365, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.6361, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.837455987930298, + "learning_rate": 2.27051177904143e-06, + "loss": 1.6269, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.5406227111816406, + "learning_rate": 2.268480909829407e-06, + "loss": 1.6538, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.716038227081299, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.6691, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.883439064025879, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.6298, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.7631118297576904, + "learning_rate": 2.262388302193339e-06, + "loss": 1.6245, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.6777117252349854, + "learning_rate": 2.260357432981316e-06, + "loss": 1.6446, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.845224142074585, + "eval_runtime": 46.8476, + "eval_samples_per_second": 21.346, + "eval_steps_per_second": 0.897, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.6280243396759033, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.6398, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.471925973892212, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.6403, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.6293203830718994, + "learning_rate": 2.254264825345248e-06, + "loss": 1.642, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.651144027709961, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.6445, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.6489877700805664, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.6761, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.5869088172912598, + "learning_rate": 2.24817221770918e-06, + "loss": 1.6593, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.411802053451538, + "learning_rate": 2.246141348497157e-06, + "loss": 1.6232, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.522610902786255, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.6358, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.5430991649627686, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.6605, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.4440951347351074, + "learning_rate": 2.240048740861089e-06, + "loss": 1.5354, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.8440407514572144, + "eval_runtime": 48.0239, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.5492775440216064, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.668, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.7511909008026123, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6424, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.5176775455474854, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.6336, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.5408577919006348, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.5843, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.4791393280029297, + "learning_rate": 2.229894394800975e-06, + "loss": 1.555, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.383148670196533, + "learning_rate": 2.227863525588952e-06, + "loss": 1.598, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.429348945617676, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.5943, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.708796501159668, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.6382, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 3.0029661655426025, + "learning_rate": 2.221770917952884e-06, + "loss": 1.6699, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.5410728454589844, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.5965, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.8443207740783691, + "eval_runtime": 47.8671, + "eval_samples_per_second": 20.891, + "eval_steps_per_second": 0.877, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.5718178749084473, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.6214, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.474177837371826, + "learning_rate": 2.215678310316816e-06, + "loss": 1.6498, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.5960052013397217, + "learning_rate": 2.213647441104793e-06, + "loss": 1.6156, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.6217992305755615, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.6311, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.5601158142089844, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.6008, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.7786355018615723, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.62, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.548356771469116, + "learning_rate": 2.205523964256702e-06, + "loss": 1.6473, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.6265289783477783, + "learning_rate": 2.203493095044679e-06, + "loss": 1.6565, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.522379159927368, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.6764, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.5479297637939453, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.6194, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.8434110879898071, + "eval_runtime": 46.6043, + "eval_samples_per_second": 21.457, + "eval_steps_per_second": 0.901, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.5401806831359863, + "learning_rate": 2.197400487408611e-06, + "loss": 1.5911, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.466893434524536, + "learning_rate": 2.195369618196588e-06, + "loss": 1.6145, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.566653251647949, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.6558, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.664053201675415, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7089, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 3.6594645977020264, + "learning_rate": 2.18927701056052e-06, + "loss": 1.6549, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.68450927734375, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.6306, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.7094614505767822, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.6147, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.481332540512085, + "learning_rate": 2.183184402924452e-06, + "loss": 1.6612, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5651051998138428, + "learning_rate": 2.181153533712429e-06, + "loss": 1.6394, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.5570919513702393, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.6162, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.8433003425598145, + "eval_runtime": 46.8673, + "eval_samples_per_second": 21.337, + "eval_steps_per_second": 0.896, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.6206037998199463, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.6869, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.8962907791137695, + "learning_rate": 2.175060926076361e-06, + "loss": 1.7022, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.6211588382720947, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.6249, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.7770566940307617, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.5821, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.543658494949341, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.6587, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.508188486099243, + "learning_rate": 2.16693744922827e-06, + "loss": 1.6114, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.703335762023926, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.6152, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.469688892364502, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.6121, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.4836199283599854, + "learning_rate": 2.160844841592202e-06, + "loss": 1.5918, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.521851062774658, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.6228, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.845065712928772, + "eval_runtime": 45.3003, + "eval_samples_per_second": 22.075, + "eval_steps_per_second": 0.927, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.7808425426483154, + "learning_rate": 2.156783103168156e-06, + "loss": 1.5887, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.696708917617798, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.6517, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.7222893238067627, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.6817, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.504417896270752, + "learning_rate": 2.150690495532088e-06, + "loss": 1.647, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.574836015701294, + "learning_rate": 2.148659626320065e-06, + "loss": 1.6554, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.5004355907440186, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.6089, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.6979293823242188, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.6226, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.5239524841308594, + "learning_rate": 2.142567018683997e-06, + "loss": 1.6875, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.5218467712402344, + "learning_rate": 2.140536149471974e-06, + "loss": 1.6095, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.5778310298919678, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.6155, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.8443745374679565, + "eval_runtime": 46.6273, + "eval_samples_per_second": 21.447, + "eval_steps_per_second": 0.901, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.671973705291748, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.6696, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.5520615577697754, + "learning_rate": 2.134443541835906e-06, + "loss": 1.6291, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.39286208152771, + "learning_rate": 2.132412672623883e-06, + "loss": 1.6012, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.8277461528778076, + "learning_rate": 2.13038180341186e-06, + "loss": 1.6398, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.7147881984710693, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.6685, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.499044418334961, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.6128, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.514784574508667, + "learning_rate": 2.124289195775792e-06, + "loss": 1.6263, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.505342483520508, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.5532, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.608139753341675, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.5946, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.5578057765960693, + "learning_rate": 2.118196588139724e-06, + "loss": 1.6103, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.8425958156585693, + "eval_runtime": 44.206, + "eval_samples_per_second": 22.621, + "eval_steps_per_second": 0.95, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.4881014823913574, + "learning_rate": 2.116165718927701e-06, + "loss": 1.6364, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.4141550064086914, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.5943, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.649827718734741, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.5994, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.547238826751709, + "learning_rate": 2.110073111291633e-06, + "loss": 1.6247, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.6515395641326904, + "learning_rate": 2.10804224207961e-06, + "loss": 1.6543, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.5575809478759766, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.6618, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.466344118118286, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.5437, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4897878170013428, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6199, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.6578829288482666, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.5989, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.6173501014709473, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.6392, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.838721752166748, + "eval_runtime": 49.4357, + "eval_samples_per_second": 20.228, + "eval_steps_per_second": 0.85, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.5655064582824707, + "learning_rate": 2.095857026807474e-06, + "loss": 1.6276, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.685790777206421, + "learning_rate": 2.093826157595451e-06, + "loss": 1.6142, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.7278990745544434, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.6292, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.5769667625427246, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.6774, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.5120608806610107, + "learning_rate": 2.087733549959383e-06, + "loss": 1.5972, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.4658212661743164, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.5683, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.526524782180786, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.6687, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.3989250659942627, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.6184, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.589115858078003, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.6153, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.4985275268554688, + "learning_rate": 2.077579203899269e-06, + "loss": 1.5682, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.8374594449996948, + "eval_runtime": 47.0329, + "eval_samples_per_second": 21.262, + "eval_steps_per_second": 0.893, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5770835876464844, + "learning_rate": 2.075548334687246e-06, + "loss": 1.6437, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.5111560821533203, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.5711, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.6110589504241943, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.6125, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.6229705810546875, + "learning_rate": 2.069455727051178e-06, + "loss": 1.595, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.648716926574707, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.5908, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.673701763153076, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.6652, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.667478561401367, + "learning_rate": 2.06336311941511e-06, + "loss": 1.6088, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 3.5688586235046387, + "learning_rate": 2.061332250203087e-06, + "loss": 1.5852, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.512603282928467, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.6218, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.5296075344085693, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.6382, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.8406264781951904, + "eval_runtime": 43.8351, + "eval_samples_per_second": 22.813, + "eval_steps_per_second": 0.958, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.598994255065918, + "learning_rate": 2.055239642567019e-06, + "loss": 1.6263, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.93025279045105, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.6502, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 3.3655412197113037, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.5872, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.955120325088501, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.6404, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.5226638317108154, + "learning_rate": 2.047116165718928e-06, + "loss": 1.6102, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.5462093353271484, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.6757, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 4.066190242767334, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.6072, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.8548827171325684, + "learning_rate": 2.04102355808286e-06, + "loss": 1.5717, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.6206634044647217, + "learning_rate": 2.038992688870837e-06, + "loss": 1.5943, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.8793396949768066, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.6121, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.83925199508667, + "eval_runtime": 44.747, + "eval_samples_per_second": 22.348, + "eval_steps_per_second": 0.939, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.6215157508850098, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.6492, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.6103456020355225, + "learning_rate": 2.032900081234769e-06, + "loss": 1.6531, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.485775947570801, + "learning_rate": 2.030869212022746e-06, + "loss": 1.6086, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.399557113647461, + "learning_rate": 2.028838342810723e-06, + "loss": 1.6126, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.4711368083953857, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.5985, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.654580593109131, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.6557, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.5091657638549805, + "learning_rate": 2.022745735174655e-06, + "loss": 1.6181, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.5723490715026855, + "learning_rate": 2.020714865962632e-06, + "loss": 1.6421, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4211535453796387, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.6226, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.5550055503845215, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.5729, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.8397995233535767, + "eval_runtime": 44.4762, + "eval_samples_per_second": 22.484, + "eval_steps_per_second": 0.944, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.618758201599121, + "learning_rate": 2.014622258326564e-06, + "loss": 1.6356, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.6717913150787354, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.6246, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.5464835166931152, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.6027, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.557669162750244, + "learning_rate": 2.008529650690496e-06, + "loss": 1.6531, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.581970453262329, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.6076, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.609236478805542, + "learning_rate": 2.00446791226645e-06, + "loss": 1.5932, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.4249160289764404, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.6372, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.5093257427215576, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.6627, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.3989412784576416, + "learning_rate": 1.998375304630382e-06, + "loss": 1.6292, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4299519062042236, + "learning_rate": 1.996344435418359e-06, + "loss": 1.6385, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.8367400169372559, + "eval_runtime": 45.921, + "eval_samples_per_second": 21.777, + "eval_steps_per_second": 0.915, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.509530782699585, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.6441, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.591426134109497, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.6382, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.6569297313690186, + "learning_rate": 1.990251827782291e-06, + "loss": 1.6469, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.6453206539154053, + "learning_rate": 1.988220958570268e-06, + "loss": 1.6348, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.5522994995117188, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.6311, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.5981009006500244, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.585, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.617323637008667, + "learning_rate": 1.9821283509342e-06, + "loss": 1.584, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.6743686199188232, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.6334, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.418383836746216, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.5934, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.7952582836151123, + "learning_rate": 1.976035743298132e-06, + "loss": 1.6779, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.8379877805709839, + "eval_runtime": 46.545, + "eval_samples_per_second": 21.485, + "eval_steps_per_second": 0.902, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.5364768505096436, + "learning_rate": 1.974004874086109e-06, + "loss": 1.5802, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.446056365966797, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6141, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.527529239654541, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.6042, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.5953001976013184, + "learning_rate": 1.967912266450041e-06, + "loss": 1.5959, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.545994758605957, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.7091, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.3900482654571533, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.5946, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.51353120803833, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.6166, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.550389528274536, + "learning_rate": 1.95978878960195e-06, + "loss": 1.6817, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.6815762519836426, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.6594, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.591909885406494, + "learning_rate": 1.955727051177904e-06, + "loss": 1.5991, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.8390737771987915, + "eval_runtime": 46.8495, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.5111982822418213, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.5993, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.655891180038452, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.5915, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.471090316772461, + "learning_rate": 1.949634443541836e-06, + "loss": 1.5867, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.8599209785461426, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.6485, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4986093044281006, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.6084, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.6345789432525635, + "learning_rate": 1.943541835905768e-06, + "loss": 1.6719, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.5636799335479736, + "learning_rate": 1.941510966693745e-06, + "loss": 1.5817, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.5867199897766113, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.6498, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.6369481086730957, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.6791, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.4554147720336914, + "learning_rate": 1.935418359057677e-06, + "loss": 1.6319, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.8408559560775757, + "eval_runtime": 49.3515, + "eval_samples_per_second": 20.263, + "eval_steps_per_second": 0.851, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.5412509441375732, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.6673, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.5755672454833984, + "learning_rate": 1.931356620633631e-06, + "loss": 1.5636, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.602530002593994, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.6261, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.755247116088867, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.5817, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.528991460800171, + "learning_rate": 1.925264012997563e-06, + "loss": 1.6173, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.473057508468628, + "learning_rate": 1.92323314378554e-06, + "loss": 1.648, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.62630033493042, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.5864, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.7876687049865723, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.5804, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.491473436355591, + "learning_rate": 1.917140536149472e-06, + "loss": 1.5716, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.656989336013794, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.6595, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.839701771736145, + "eval_runtime": 44.5976, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.6800942420959473, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.6415, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.4725751876831055, + "learning_rate": 1.911047928513404e-06, + "loss": 1.5901, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.552624464035034, + "learning_rate": 1.909017059301381e-06, + "loss": 1.5956, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.6042184829711914, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.6328, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.5159223079681396, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.604, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.488041639328003, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.5539, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.659151077270508, + "learning_rate": 1.90089358245329e-06, + "loss": 1.5999, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.700378179550171, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.6485, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.3943607807159424, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.5736, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.8171095848083496, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6259, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.8379307985305786, + "eval_runtime": 48.6918, + "eval_samples_per_second": 20.537, + "eval_steps_per_second": 0.863, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.562655210494995, + "learning_rate": 1.892770105605199e-06, + "loss": 1.6439, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.5640337467193604, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.6113, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.5324809551239014, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.6331, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.393224000930786, + "learning_rate": 1.886677497969131e-06, + "loss": 1.5228, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 3.073945999145508, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.632, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.7066094875335693, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.5508, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.981933832168579, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6973, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.903444528579712, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.6036, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.8616178035736084, + "learning_rate": 1.876523151909017e-06, + "loss": 1.5723, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.8243541717529297, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.555, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.8437565565109253, + "eval_runtime": 48.5449, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.803142547607422, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.5861, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 3.146677017211914, + "learning_rate": 1.870430544272949e-06, + "loss": 1.6791, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 3.2554781436920166, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.5626, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.955395460128784, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.5893, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.941377878189087, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.5962, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.765620708465576, + "learning_rate": 1.862307067424858e-06, + "loss": 1.595, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.8529839515686035, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.6296, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.7770590782165527, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.5711, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.827406883239746, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.5966, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.927119016647339, + "learning_rate": 1.854183590576767e-06, + "loss": 1.6463, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.841707468032837, + "eval_runtime": 49.0965, + "eval_samples_per_second": 20.368, + "eval_steps_per_second": 0.855, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.8223392963409424, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.559, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.862473726272583, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.5189, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.9060091972351074, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.5631, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.8727498054504395, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6013, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.862298011779785, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.549, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.8221330642700195, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.6341, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.938336133956909, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.602, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.719489812850952, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.5626, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.6615874767303467, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.5535, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.835413932800293, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.5742, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.8375873565673828, + "eval_runtime": 45.731, + "eval_samples_per_second": 21.867, + "eval_steps_per_second": 0.918, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.773707389831543, + "learning_rate": 1.831844029244517e-06, + "loss": 1.5499, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 3.0959930419921875, + "learning_rate": 1.829813160032494e-06, + "loss": 1.5656, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.8521792888641357, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.5982, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.970618486404419, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.5731, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 3.0716512203216553, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.5824, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.8776299953460693, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.6283, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.72183895111084, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.5422, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.853234052658081, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.5533, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.7193424701690674, + "learning_rate": 1.815597075548335e-06, + "loss": 1.5923, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.9768450260162354, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.6277, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.8401803970336914, + "eval_runtime": 45.8062, + "eval_samples_per_second": 21.831, + "eval_steps_per_second": 0.917, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.752772569656372, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.5241, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.784980297088623, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.5657, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.7485857009887695, + "learning_rate": 1.807473598700244e-06, + "loss": 1.5501, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.7191033363342285, + "learning_rate": 1.805442729488221e-06, + "loss": 1.5718, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.697540044784546, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.5804, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.7175710201263428, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.5809, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.764824151992798, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.545, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.946669816970825, + "learning_rate": 1.79731925264013e-06, + "loss": 1.595, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.689326524734497, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.5676, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.759997606277466, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.6032, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.8379274606704712, + "eval_runtime": 45.5941, + "eval_samples_per_second": 21.933, + "eval_steps_per_second": 0.921, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.8656060695648193, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.6086, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.7155873775482178, + "learning_rate": 1.789195775792039e-06, + "loss": 1.5218, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.8487389087677, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6285, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.8070144653320312, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.5479, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.856282949447632, + "learning_rate": 1.783103168155971e-06, + "loss": 1.5689, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.7465922832489014, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.5733, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.824997901916504, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.5583, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.8030786514282227, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.5383, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.911377429962158, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.586, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.7452991008758545, + "learning_rate": 1.772948822095857e-06, + "loss": 1.6409, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.8399248123168945, + "eval_runtime": 46.1241, + "eval_samples_per_second": 21.681, + "eval_steps_per_second": 0.911, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.8255481719970703, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.6134, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.865100622177124, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.5791, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.740133047103882, + "learning_rate": 1.766856214459789e-06, + "loss": 1.5651, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.7219033241271973, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.5453, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.7266135215759277, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.6375, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.8476688861846924, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.5877, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.810105085372925, + "learning_rate": 1.758732737611698e-06, + "loss": 1.5796, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.78798246383667, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.5708, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.750525951385498, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.5567, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.818673849105835, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.5661, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.8374614715576172, + "eval_runtime": 43.5618, + "eval_samples_per_second": 22.956, + "eval_steps_per_second": 0.964, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.7967586517333984, + "learning_rate": 1.750609260763607e-06, + "loss": 1.5846, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.766272783279419, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.5721, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.8006162643432617, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.5261, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 3.714587688446045, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.6599, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 3.143052339553833, + "learning_rate": 1.742485783915516e-06, + "loss": 1.659, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.8256232738494873, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.5076, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.8843302726745605, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.5862, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 6.692182540893555, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.6144, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.928053617477417, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.571, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.8071024417877197, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.5861, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.837381362915039, + "eval_runtime": 44.7089, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 3.070399522781372, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.5636, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 3.298100709915161, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.5717, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 5.891775131225586, + "learning_rate": 1.726238830219334e-06, + "loss": 1.5757, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.838676691055298, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.6369, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.8816466331481934, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.5885, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 3.1703031063079834, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.5966, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 3.504777669906616, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.4932, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 3.1451492309570312, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.5071, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.7158446311950684, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.5725, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.8391170501708984, + "learning_rate": 1.712022745735175e-06, + "loss": 1.5742, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.8371202945709229, + "eval_runtime": 43.2464, + "eval_samples_per_second": 23.123, + "eval_steps_per_second": 0.971, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.8346052169799805, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.5808, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.767813205718994, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.5209, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.7955846786499023, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.581, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.8616271018981934, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.5758, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.7975544929504395, + "learning_rate": 1.701868399675061e-06, + "loss": 1.5912, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 3.095912456512451, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.6522, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 3.102691888809204, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.6004, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.8405165672302246, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.5649, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.8038063049316406, + "learning_rate": 1.69374492282697e-06, + "loss": 1.545, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.8661468029022217, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6297, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.8386919498443604, + "eval_runtime": 46.0334, + "eval_samples_per_second": 21.723, + "eval_steps_per_second": 0.912, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.8217391967773438, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.5694, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 3.099724769592285, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.5986, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.9951272010803223, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.5429, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.9366989135742188, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.5666, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 3.0171356201171875, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.6212, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.952259063720703, + "learning_rate": 1.679528838342811e-06, + "loss": 1.5611, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.667905330657959, + "learning_rate": 1.677497969130788e-06, + "loss": 1.5556, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.8001160621643066, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.5854, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.6954760551452637, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.5278, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.8026583194732666, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.5103, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.8384881019592285, + "eval_runtime": 46.1899, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.898869752883911, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6188, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 3.0432069301605225, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.5945, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.7744672298431396, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.5642, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.8844289779663086, + "learning_rate": 1.663281884646629e-06, + "loss": 1.5657, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 3.0022218227386475, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.6297, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.918320417404175, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.5928, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.913848876953125, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.6219, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.954702615737915, + "learning_rate": 1.655158407798538e-06, + "loss": 1.5812, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.7966420650482178, + "learning_rate": 1.653127538586515e-06, + "loss": 1.6034, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 3.1328368186950684, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.6168, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.836020827293396, + "eval_runtime": 46.1979, + "eval_samples_per_second": 21.646, + "eval_steps_per_second": 0.909, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.8109259605407715, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.6091, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.823570728302002, + "learning_rate": 1.647034930950447e-06, + "loss": 1.5504, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.885016918182373, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.5784, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 3.612942934036255, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.6014, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.8981637954711914, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.5552, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.6703410148620605, + "learning_rate": 1.638911454102356e-06, + "loss": 1.5567, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.7087886333465576, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.613, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.7711687088012695, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.5586, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.8473854064941406, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.565, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 3.783824920654297, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.5909, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.8384798765182495, + "eval_runtime": 46.2339, + "eval_samples_per_second": 21.629, + "eval_steps_per_second": 0.908, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.9971532821655273, + "learning_rate": 1.628757108042242e-06, + "loss": 1.6251, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.8581976890563965, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.5632, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.9119839668273926, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.5806, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.840101718902588, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.6706, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.8574070930480957, + "learning_rate": 1.620633631194151e-06, + "loss": 1.6138, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.820692539215088, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.5534, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7690017223358154, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.5922, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 3.062229871749878, + "learning_rate": 1.614541023558083e-06, + "loss": 1.5893, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.727062225341797, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.5851, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.7693302631378174, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.6031, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.8332058191299438, + "eval_runtime": 45.0301, + "eval_samples_per_second": 22.207, + "eval_steps_per_second": 0.933, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.803758382797241, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.5504, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.775770902633667, + "learning_rate": 1.606417546709992e-06, + "loss": 1.5707, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.7698636054992676, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.5763, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.739940881729126, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.5386, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.7594099044799805, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.5805, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.853273391723633, + "learning_rate": 1.598294069861901e-06, + "loss": 1.5548, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.9859204292297363, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6278, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.742898941040039, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.555, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.769071340560913, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.5504, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.7191832065582275, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6014, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.8336049318313599, + "eval_runtime": 47.0078, + "eval_samples_per_second": 21.273, + "eval_steps_per_second": 0.893, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.842529773712158, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.5723, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.768066883087158, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.5272, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7481656074523926, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.5843, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.806687593460083, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.5598, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.8214614391326904, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.5972, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.8541247844696045, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.5988, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 3.000246286392212, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.583, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.9164586067199707, + "learning_rate": 1.573923639317628e-06, + "loss": 1.5949, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.8099935054779053, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.5959, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.88322377204895, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.6062, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.832599401473999, + "eval_runtime": 45.9576, + "eval_samples_per_second": 21.759, + "eval_steps_per_second": 0.914, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.8019371032714844, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.5445, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.8585665225982666, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6079, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.7423956394195557, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.5537, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.7420766353607178, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.5846, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.8864388465881348, + "learning_rate": 1.559707554833469e-06, + "loss": 1.6213, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.7483463287353516, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.5012, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.7340080738067627, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.555, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.8000667095184326, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.5259, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.9176089763641357, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.5726, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.9147658348083496, + "learning_rate": 1.549553208773355e-06, + "loss": 1.5685, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.834657907485962, + "eval_runtime": 45.8709, + "eval_samples_per_second": 21.8, + "eval_steps_per_second": 0.916, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.9109392166137695, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.5533, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 3.0263078212738037, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.6277, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.9347317218780518, + "learning_rate": 1.543460601137287e-06, + "loss": 1.5738, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.8229548931121826, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.5969, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.906954288482666, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.5772, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.9214975833892822, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.5147, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.844081163406372, + "learning_rate": 1.535337124289196e-06, + "loss": 1.5635, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 3.0214555263519287, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.6466, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.9254050254821777, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.5837, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.8986024856567383, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.5997, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.8318188190460205, + "eval_runtime": 46.7695, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.974231719970703, + "learning_rate": 1.5272136474411047e-06, + "loss": 1.6143, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.7890946865081787, + "learning_rate": 1.525182778229082e-06, + "loss": 1.5956, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.820800304412842, + "learning_rate": 1.5231519090170593e-06, + "loss": 1.5626, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.84753680229187, + "learning_rate": 1.5211210398050365e-06, + "loss": 1.5473, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.829115152359009, + "learning_rate": 1.5190901705930138e-06, + "loss": 1.591, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.786787509918213, + "learning_rate": 1.517059301380991e-06, + "loss": 1.565, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.7778282165527344, + "learning_rate": 1.5150284321689684e-06, + "loss": 1.5666, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.9727330207824707, + "learning_rate": 1.5129975629569456e-06, + "loss": 1.577, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.876600980758667, + "learning_rate": 1.510966693744923e-06, + "loss": 1.5742, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.7339680194854736, + "learning_rate": 1.5089358245329002e-06, + "loss": 1.5378, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 1.834323763847351, + "eval_runtime": 44.2017, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.9201254844665527, + "learning_rate": 1.5069049553208775e-06, + "loss": 1.5962, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.8632285594940186, + "learning_rate": 1.5048740861088547e-06, + "loss": 1.589, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.9003729820251465, + "learning_rate": 1.502843216896832e-06, + "loss": 1.5125, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.81120228767395, + "learning_rate": 1.500812347684809e-06, + "loss": 1.5809, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.747528553009033, + "learning_rate": 1.4987814784727864e-06, + "loss": 1.5611, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.6891186237335205, + "learning_rate": 1.4967506092607636e-06, + "loss": 1.5387, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.9105594158172607, + "learning_rate": 1.494719740048741e-06, + "loss": 1.5921, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.8186421394348145, + "learning_rate": 1.4926888708367182e-06, + "loss": 1.6075, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 3.0198328495025635, + "learning_rate": 1.4906580016246955e-06, + "loss": 1.611, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 3.0839414596557617, + "learning_rate": 1.4886271324126727e-06, + "loss": 1.6298, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 1.8325375318527222, + "eval_runtime": 53.1022, + "eval_samples_per_second": 18.832, + "eval_steps_per_second": 0.791, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.8375303745269775, + "learning_rate": 1.48659626320065e-06, + "loss": 1.4942, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.857344627380371, + "learning_rate": 1.4845653939886273e-06, + "loss": 1.5182, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.7296979427337646, + "learning_rate": 1.4825345247766046e-06, + "loss": 1.5081, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.9989287853240967, + "learning_rate": 1.4805036555645818e-06, + "loss": 1.5577, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.9123456478118896, + "learning_rate": 1.4784727863525591e-06, + "loss": 1.5339, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.845898151397705, + "learning_rate": 1.4764419171405362e-06, + "loss": 1.6084, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 3.089202404022217, + "learning_rate": 1.4744110479285135e-06, + "loss": 1.6133, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.9908018112182617, + "learning_rate": 1.4723801787164907e-06, + "loss": 1.5984, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.838611125946045, + "learning_rate": 1.470349309504468e-06, + "loss": 1.5971, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.902926445007324, + "learning_rate": 1.4683184402924453e-06, + "loss": 1.6029, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 1.8312546014785767, + "eval_runtime": 46.3457, + "eval_samples_per_second": 21.577, + "eval_steps_per_second": 0.906, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.8279829025268555, + "learning_rate": 1.4662875710804226e-06, + "loss": 1.65, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.7380146980285645, + "learning_rate": 1.4642567018683998e-06, + "loss": 1.5673, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.723890781402588, + "learning_rate": 1.4622258326563771e-06, + "loss": 1.5573, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.8826022148132324, + "learning_rate": 1.4601949634443544e-06, + "loss": 1.5648, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.948598861694336, + "learning_rate": 1.4581640942323317e-06, + "loss": 1.6603, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.8282432556152344, + "learning_rate": 1.456133225020309e-06, + "loss": 1.5875, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.874748468399048, + "learning_rate": 1.4541023558082862e-06, + "loss": 1.5713, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.894249677658081, + "learning_rate": 1.4520714865962635e-06, + "loss": 1.6063, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.907095193862915, + "learning_rate": 1.4500406173842406e-06, + "loss": 1.5744, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.892090082168579, + "learning_rate": 1.4480097481722178e-06, + "loss": 1.5877, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 1.8305796384811401, + "eval_runtime": 48.3583, + "eval_samples_per_second": 20.679, + "eval_steps_per_second": 0.869, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.7991275787353516, + "learning_rate": 1.4459788789601951e-06, + "loss": 1.5475, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.9188716411590576, + "learning_rate": 1.4439480097481724e-06, + "loss": 1.5385, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.8575944900512695, + "learning_rate": 1.4419171405361497e-06, + "loss": 1.5855, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.814990520477295, + "learning_rate": 1.439886271324127e-06, + "loss": 1.5923, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.7953567504882812, + "learning_rate": 1.4378554021121042e-06, + "loss": 1.5724, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.882002115249634, + "learning_rate": 1.4358245329000815e-06, + "loss": 1.6086, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.910799503326416, + "learning_rate": 1.4337936636880588e-06, + "loss": 1.5933, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.911015748977661, + "learning_rate": 1.431762794476036e-06, + "loss": 1.5356, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.9788129329681396, + "learning_rate": 1.4297319252640133e-06, + "loss": 1.5546, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.8637843132019043, + "learning_rate": 1.4277010560519906e-06, + "loss": 1.5423, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 1.8307991027832031, + "eval_runtime": 45.4486, + "eval_samples_per_second": 22.003, + "eval_steps_per_second": 0.924, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 3.007390260696411, + "learning_rate": 1.4256701868399674e-06, + "loss": 1.6244, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.908496856689453, + "learning_rate": 1.4236393176279447e-06, + "loss": 1.5767, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.9650135040283203, + "learning_rate": 1.421608448415922e-06, + "loss": 1.6639, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.8085806369781494, + "learning_rate": 1.4195775792038993e-06, + "loss": 1.5486, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.7759034633636475, + "learning_rate": 1.4175467099918765e-06, + "loss": 1.5847, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.9309468269348145, + "learning_rate": 1.4155158407798538e-06, + "loss": 1.6014, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.8542673587799072, + "learning_rate": 1.413484971567831e-06, + "loss": 1.5183, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 3.0022106170654297, + "learning_rate": 1.4114541023558084e-06, + "loss": 1.5511, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.826852321624756, + "learning_rate": 1.4094232331437856e-06, + "loss": 1.6079, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.8682327270507812, + "learning_rate": 1.407392363931763e-06, + "loss": 1.6579, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 1.8299224376678467, + "eval_runtime": 44.6401, + "eval_samples_per_second": 22.401, + "eval_steps_per_second": 0.941, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.8816471099853516, + "learning_rate": 1.4053614947197402e-06, + "loss": 1.5825, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.696047306060791, + "learning_rate": 1.4033306255077175e-06, + "loss": 1.5068, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.9243481159210205, + "learning_rate": 1.4012997562956945e-06, + "loss": 1.6172, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.810744285583496, + "learning_rate": 1.3992688870836718e-06, + "loss": 1.5976, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.7787344455718994, + "learning_rate": 1.397238017871649e-06, + "loss": 1.5829, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.817336082458496, + "learning_rate": 1.3952071486596264e-06, + "loss": 1.5482, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.7691237926483154, + "learning_rate": 1.3931762794476036e-06, + "loss": 1.6098, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.7974281311035156, + "learning_rate": 1.391145410235581e-06, + "loss": 1.5591, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.8233275413513184, + "learning_rate": 1.3891145410235582e-06, + "loss": 1.4975, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.7465834617614746, + "learning_rate": 1.3870836718115355e-06, + "loss": 1.5682, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 1.8317815065383911, + "eval_runtime": 44.8326, + "eval_samples_per_second": 22.305, + "eval_steps_per_second": 0.937, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.9384610652923584, + "learning_rate": 1.3850528025995127e-06, + "loss": 1.5627, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.6946020126342773, + "learning_rate": 1.38302193338749e-06, + "loss": 1.4851, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.875818967819214, + "learning_rate": 1.3809910641754673e-06, + "loss": 1.595, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.871393918991089, + "learning_rate": 1.3789601949634446e-06, + "loss": 1.5297, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.8451645374298096, + "learning_rate": 1.3769293257514218e-06, + "loss": 1.5912, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.808422803878784, + "learning_rate": 1.374898456539399e-06, + "loss": 1.5316, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.866929292678833, + "learning_rate": 1.3728675873273762e-06, + "loss": 1.6223, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.969078302383423, + "learning_rate": 1.3708367181153535e-06, + "loss": 1.5855, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.739272117614746, + "learning_rate": 1.3688058489033307e-06, + "loss": 1.5915, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.9254791736602783, + "learning_rate": 1.366774979691308e-06, + "loss": 1.5989, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 1.8284611701965332, + "eval_runtime": 45.0516, + "eval_samples_per_second": 22.197, + "eval_steps_per_second": 0.932, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.7639904022216797, + "learning_rate": 1.3647441104792853e-06, + "loss": 1.5735, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.782290458679199, + "learning_rate": 1.3627132412672626e-06, + "loss": 1.5365, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.8718478679656982, + "learning_rate": 1.3606823720552398e-06, + "loss": 1.5776, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.770900249481201, + "learning_rate": 1.3586515028432171e-06, + "loss": 1.5242, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.7527379989624023, + "learning_rate": 1.3566206336311944e-06, + "loss": 1.5783, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.724205732345581, + "learning_rate": 1.3545897644191717e-06, + "loss": 1.5556, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.7595510482788086, + "learning_rate": 1.352558895207149e-06, + "loss": 1.6084, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.832383394241333, + "learning_rate": 1.3505280259951258e-06, + "loss": 1.5755, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.8154635429382324, + "learning_rate": 1.348497156783103e-06, + "loss": 1.6091, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.808480739593506, + "learning_rate": 1.3464662875710804e-06, + "loss": 1.5403, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 1.8285717964172363, + "eval_runtime": 44.0301, + "eval_samples_per_second": 22.712, + "eval_steps_per_second": 0.954, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.7877376079559326, + "learning_rate": 1.3444354183590576e-06, + "loss": 1.5985, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.854957103729248, + "learning_rate": 1.342404549147035e-06, + "loss": 1.5576, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.8522772789001465, + "learning_rate": 1.3403736799350122e-06, + "loss": 1.5914, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.866199493408203, + "learning_rate": 1.3383428107229895e-06, + "loss": 1.6094, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.9277379512786865, + "learning_rate": 1.3363119415109667e-06, + "loss": 1.6065, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.8854689598083496, + "learning_rate": 1.334281072298944e-06, + "loss": 1.5892, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.7135169506073, + "learning_rate": 1.3322502030869213e-06, + "loss": 1.5733, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.8702995777130127, + "learning_rate": 1.3302193338748986e-06, + "loss": 1.6054, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.8330469131469727, + "learning_rate": 1.3281884646628758e-06, + "loss": 1.5902, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 3.016582727432251, + "learning_rate": 1.3261575954508531e-06, + "loss": 1.5931, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 1.82985520362854, + "eval_runtime": 45.8601, + "eval_samples_per_second": 21.805, + "eval_steps_per_second": 0.916, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.8396756649017334, + "learning_rate": 1.3241267262388302e-06, + "loss": 1.5882, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.7104387283325195, + "learning_rate": 1.3220958570268074e-06, + "loss": 1.5223, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.787862777709961, + "learning_rate": 1.3200649878147847e-06, + "loss": 1.5549, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.9957401752471924, + "learning_rate": 1.318034118602762e-06, + "loss": 1.6036, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.916990280151367, + "learning_rate": 1.3160032493907393e-06, + "loss": 1.5947, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.972477912902832, + "learning_rate": 1.3139723801787166e-06, + "loss": 1.624, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.84188175201416, + "learning_rate": 1.3119415109666938e-06, + "loss": 1.5708, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.908295154571533, + "learning_rate": 1.309910641754671e-06, + "loss": 1.5412, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.8915605545043945, + "learning_rate": 1.3078797725426484e-06, + "loss": 1.5463, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.7805707454681396, + "learning_rate": 1.3058489033306257e-06, + "loss": 1.5555, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 1.8281073570251465, + "eval_runtime": 44.7855, + "eval_samples_per_second": 22.329, + "eval_steps_per_second": 0.938, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.933260679244995, + "learning_rate": 1.303818034118603e-06, + "loss": 1.578, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.970050096511841, + "learning_rate": 1.3017871649065802e-06, + "loss": 1.5606, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.717027187347412, + "learning_rate": 1.2997562956945573e-06, + "loss": 1.5446, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.966905355453491, + "learning_rate": 1.2977254264825345e-06, + "loss": 1.549, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.847358465194702, + "learning_rate": 1.2956945572705118e-06, + "loss": 1.581, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.814897060394287, + "learning_rate": 1.293663688058489e-06, + "loss": 1.5416, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.788320302963257, + "learning_rate": 1.2916328188464664e-06, + "loss": 1.5018, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.944441556930542, + "learning_rate": 1.2896019496344436e-06, + "loss": 1.5318, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.8705525398254395, + "learning_rate": 1.287571080422421e-06, + "loss": 1.5953, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.7703897953033447, + "learning_rate": 1.2855402112103982e-06, + "loss": 1.5287, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 1.8304927349090576, + "eval_runtime": 47.1736, + "eval_samples_per_second": 21.198, + "eval_steps_per_second": 0.89, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.862663745880127, + "learning_rate": 1.2835093419983755e-06, + "loss": 1.541, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.8288817405700684, + "learning_rate": 1.2814784727863528e-06, + "loss": 1.5466, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 3.0720043182373047, + "learning_rate": 1.27944760357433e-06, + "loss": 1.6106, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.9464635848999023, + "learning_rate": 1.2774167343623073e-06, + "loss": 1.6051, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.9039576053619385, + "learning_rate": 1.2753858651502846e-06, + "loss": 1.5525, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.8839735984802246, + "learning_rate": 1.2733549959382616e-06, + "loss": 1.5551, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.915053129196167, + "learning_rate": 1.271324126726239e-06, + "loss": 1.5853, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.862844467163086, + "learning_rate": 1.2692932575142162e-06, + "loss": 1.5233, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.717557430267334, + "learning_rate": 1.2672623883021935e-06, + "loss": 1.5138, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 3.1381704807281494, + "learning_rate": 1.2652315190901707e-06, + "loss": 1.6073, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 1.82700514793396, + "eval_runtime": 51.3349, + "eval_samples_per_second": 19.48, + "eval_steps_per_second": 0.818, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.8603286743164062, + "learning_rate": 1.263200649878148e-06, + "loss": 1.5713, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.9590084552764893, + "learning_rate": 1.2611697806661253e-06, + "loss": 1.5952, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.8419439792633057, + "learning_rate": 1.2591389114541026e-06, + "loss": 1.5729, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.7028636932373047, + "learning_rate": 1.2571080422420798e-06, + "loss": 1.5427, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.819112539291382, + "learning_rate": 1.2550771730300571e-06, + "loss": 1.6221, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.821469783782959, + "learning_rate": 1.2530463038180344e-06, + "loss": 1.5756, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.979747772216797, + "learning_rate": 1.2510154346060117e-06, + "loss": 1.5662, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 3.176698684692383, + "learning_rate": 1.2489845653939887e-06, + "loss": 1.591, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.8962442874908447, + "learning_rate": 1.246953696181966e-06, + "loss": 1.6549, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.840876817703247, + "learning_rate": 1.2449228269699433e-06, + "loss": 1.5636, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 1.8279746770858765, + "eval_runtime": 45.3966, + "eval_samples_per_second": 22.028, + "eval_steps_per_second": 0.925, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.8487303256988525, + "learning_rate": 1.2428919577579204e-06, + "loss": 1.5345, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.8009743690490723, + "learning_rate": 1.2408610885458976e-06, + "loss": 1.5673, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.7863199710845947, + "learning_rate": 1.238830219333875e-06, + "loss": 1.5351, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 3.4834156036376953, + "learning_rate": 1.2367993501218522e-06, + "loss": 1.6171, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.9101381301879883, + "learning_rate": 1.2347684809098295e-06, + "loss": 1.6026, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.750434160232544, + "learning_rate": 1.2327376116978067e-06, + "loss": 1.5199, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.8388402462005615, + "learning_rate": 1.230706742485784e-06, + "loss": 1.5253, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 3.000380754470825, + "learning_rate": 1.2286758732737613e-06, + "loss": 1.6463, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 3.0284345149993896, + "learning_rate": 1.2266450040617386e-06, + "loss": 1.637, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.9016034603118896, + "learning_rate": 1.2246141348497158e-06, + "loss": 1.5794, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 1.8288570642471313, + "eval_runtime": 47.0365, + "eval_samples_per_second": 21.26, + "eval_steps_per_second": 0.893, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 3.27822208404541, + "learning_rate": 1.2225832656376931e-06, + "loss": 1.5734, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.8195855617523193, + "learning_rate": 1.2205523964256704e-06, + "loss": 1.5512, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.7829315662384033, + "learning_rate": 1.2185215272136477e-06, + "loss": 1.5804, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.8760147094726562, + "learning_rate": 1.2164906580016247e-06, + "loss": 1.5858, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.8194808959960938, + "learning_rate": 1.214459788789602e-06, + "loss": 1.485, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.901355266571045, + "learning_rate": 1.2124289195775793e-06, + "loss": 1.5287, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 3.0580921173095703, + "learning_rate": 1.2103980503655566e-06, + "loss": 1.5596, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 3.0313403606414795, + "learning_rate": 1.2083671811535338e-06, + "loss": 1.5536, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.7663581371307373, + "learning_rate": 1.2063363119415111e-06, + "loss": 1.5479, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.7729651927948, + "learning_rate": 1.2043054427294882e-06, + "loss": 1.56, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 1.8264360427856445, + "eval_runtime": 45.8995, + "eval_samples_per_second": 21.787, + "eval_steps_per_second": 0.915, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 3.0862157344818115, + "learning_rate": 1.2022745735174654e-06, + "loss": 1.5869, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.9891650676727295, + "learning_rate": 1.2002437043054427e-06, + "loss": 1.6249, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.8348777294158936, + "learning_rate": 1.19821283509342e-06, + "loss": 1.6116, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.915382146835327, + "learning_rate": 1.1961819658813973e-06, + "loss": 1.5783, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 3.086989402770996, + "learning_rate": 1.1941510966693746e-06, + "loss": 1.5473, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.8165829181671143, + "learning_rate": 1.1921202274573518e-06, + "loss": 1.5617, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 3.121267795562744, + "learning_rate": 1.190089358245329e-06, + "loss": 1.6125, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.889802932739258, + "learning_rate": 1.1880584890333064e-06, + "loss": 1.5637, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.8893845081329346, + "learning_rate": 1.1860276198212837e-06, + "loss": 1.5929, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.900737762451172, + "learning_rate": 1.183996750609261e-06, + "loss": 1.5798, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 1.826514720916748, + "eval_runtime": 46.0672, + "eval_samples_per_second": 21.707, + "eval_steps_per_second": 0.912, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.8097946643829346, + "learning_rate": 1.1819658813972382e-06, + "loss": 1.5653, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.82616925239563, + "learning_rate": 1.1799350121852153e-06, + "loss": 1.612, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 3.3667945861816406, + "learning_rate": 1.1779041429731925e-06, + "loss": 1.6107, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.9167160987854004, + "learning_rate": 1.1758732737611698e-06, + "loss": 1.5424, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.9599924087524414, + "learning_rate": 1.173842404549147e-06, + "loss": 1.6195, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.8982062339782715, + "learning_rate": 1.1718115353371244e-06, + "loss": 1.5579, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.915224075317383, + "learning_rate": 1.1697806661251016e-06, + "loss": 1.6204, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.9213180541992188, + "learning_rate": 1.167749796913079e-06, + "loss": 1.5988, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.9090816974639893, + "learning_rate": 1.1657189277010562e-06, + "loss": 1.5773, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.8555307388305664, + "learning_rate": 1.1636880584890335e-06, + "loss": 1.5687, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 1.8246150016784668, + "eval_runtime": 45.9157, + "eval_samples_per_second": 21.779, + "eval_steps_per_second": 0.915, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.8956985473632812, + "learning_rate": 1.1616571892770108e-06, + "loss": 1.5878, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 4.177687644958496, + "learning_rate": 1.159626320064988e-06, + "loss": 1.579, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.9070162773132324, + "learning_rate": 1.1575954508529653e-06, + "loss": 1.5926, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.7478721141815186, + "learning_rate": 1.1555645816409426e-06, + "loss": 1.5898, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.6728436946868896, + "learning_rate": 1.1535337124289196e-06, + "loss": 1.5345, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.809030532836914, + "learning_rate": 1.151502843216897e-06, + "loss": 1.5174, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 3.078012466430664, + "learning_rate": 1.1494719740048742e-06, + "loss": 1.5914, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 4.21104621887207, + "learning_rate": 1.1474411047928515e-06, + "loss": 1.5694, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 3.0975046157836914, + "learning_rate": 1.1454102355808287e-06, + "loss": 1.545, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.925477981567383, + "learning_rate": 1.143379366368806e-06, + "loss": 1.5396, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 1.8294752836227417, + "eval_runtime": 47.0872, + "eval_samples_per_second": 21.237, + "eval_steps_per_second": 0.892, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 3.391608238220215, + "learning_rate": 1.141348497156783e-06, + "loss": 1.6501, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 3.053389549255371, + "learning_rate": 1.1393176279447604e-06, + "loss": 1.6376, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.8790252208709717, + "learning_rate": 1.1372867587327376e-06, + "loss": 1.5002, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 3.0458412170410156, + "learning_rate": 1.135255889520715e-06, + "loss": 1.5955, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.8105313777923584, + "learning_rate": 1.1332250203086922e-06, + "loss": 1.5849, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.893143892288208, + "learning_rate": 1.1311941510966695e-06, + "loss": 1.5554, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.86181640625, + "learning_rate": 1.1291632818846467e-06, + "loss": 1.5409, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.749263286590576, + "learning_rate": 1.127132412672624e-06, + "loss": 1.544, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.759432554244995, + "learning_rate": 1.1251015434606013e-06, + "loss": 1.5787, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.8302371501922607, + "learning_rate": 1.1230706742485786e-06, + "loss": 1.6204, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 1.8242546319961548, + "eval_runtime": 45.7883, + "eval_samples_per_second": 21.84, + "eval_steps_per_second": 0.917, + "step": 2200 + }, + { + "epoch": 2.412054794520548, + "grad_norm": 2.8081696033477783, + "learning_rate": 1.1210398050365558e-06, + "loss": 1.5497, + "step": 2201 + }, + { + "epoch": 2.413150684931507, + "grad_norm": 2.777566432952881, + "learning_rate": 1.1190089358245331e-06, + "loss": 1.5249, + "step": 2202 + }, + { + "epoch": 2.4142465753424656, + "grad_norm": 2.851879835128784, + "learning_rate": 1.1169780666125102e-06, + "loss": 1.5885, + "step": 2203 + }, + { + "epoch": 2.4153424657534246, + "grad_norm": 2.859072685241699, + "learning_rate": 1.1149471974004875e-06, + "loss": 1.5438, + "step": 2204 + }, + { + "epoch": 2.4164383561643836, + "grad_norm": 2.908538818359375, + "learning_rate": 1.1129163281884647e-06, + "loss": 1.6276, + "step": 2205 + }, + { + "epoch": 2.417534246575342, + "grad_norm": 2.7648227214813232, + "learning_rate": 1.110885458976442e-06, + "loss": 1.5367, + "step": 2206 + }, + { + "epoch": 2.4186301369863012, + "grad_norm": 2.9282727241516113, + "learning_rate": 1.1088545897644193e-06, + "loss": 1.6055, + "step": 2207 + }, + { + "epoch": 2.4197260273972603, + "grad_norm": 2.7551443576812744, + "learning_rate": 1.1068237205523966e-06, + "loss": 1.5984, + "step": 2208 + }, + { + "epoch": 2.4208219178082193, + "grad_norm": 2.88720440864563, + "learning_rate": 1.1047928513403736e-06, + "loss": 1.5988, + "step": 2209 + }, + { + "epoch": 2.421917808219178, + "grad_norm": 2.75961971282959, + "learning_rate": 1.102761982128351e-06, + "loss": 1.5291, + "step": 2210 + }, + { + "epoch": 2.421917808219178, + "eval_loss": 1.8230210542678833, + "eval_runtime": 45.9261, + "eval_samples_per_second": 21.774, + "eval_steps_per_second": 0.915, + "step": 2210 + }, + { + "epoch": 2.423013698630137, + "grad_norm": 2.7809903621673584, + "learning_rate": 1.1007311129163282e-06, + "loss": 1.5563, + "step": 2211 + }, + { + "epoch": 2.424109589041096, + "grad_norm": 2.7741317749023438, + "learning_rate": 1.0987002437043055e-06, + "loss": 1.5594, + "step": 2212 + }, + { + "epoch": 2.425205479452055, + "grad_norm": 2.840890884399414, + "learning_rate": 1.0966693744922827e-06, + "loss": 1.52, + "step": 2213 + }, + { + "epoch": 2.4263013698630136, + "grad_norm": 2.9647538661956787, + "learning_rate": 1.09463850528026e-06, + "loss": 1.5447, + "step": 2214 + }, + { + "epoch": 2.4273972602739726, + "grad_norm": 2.9884531497955322, + "learning_rate": 1.0926076360682373e-06, + "loss": 1.5426, + "step": 2215 + }, + { + "epoch": 2.4284931506849317, + "grad_norm": 2.8884270191192627, + "learning_rate": 1.0905767668562146e-06, + "loss": 1.5546, + "step": 2216 + }, + { + "epoch": 2.4295890410958902, + "grad_norm": 2.97591495513916, + "learning_rate": 1.0885458976441918e-06, + "loss": 1.632, + "step": 2217 + }, + { + "epoch": 2.4306849315068493, + "grad_norm": 2.944126605987549, + "learning_rate": 1.0865150284321691e-06, + "loss": 1.5845, + "step": 2218 + }, + { + "epoch": 2.4317808219178083, + "grad_norm": 2.9660258293151855, + "learning_rate": 1.0844841592201464e-06, + "loss": 1.5939, + "step": 2219 + }, + { + "epoch": 2.432876712328767, + "grad_norm": 2.8848304748535156, + "learning_rate": 1.0824532900081237e-06, + "loss": 1.5848, + "step": 2220 + }, + { + "epoch": 2.432876712328767, + "eval_loss": 1.8266901969909668, + "eval_runtime": 51.9703, + "eval_samples_per_second": 19.242, + "eval_steps_per_second": 0.808, + "step": 2220 + }, + { + "epoch": 2.433972602739726, + "grad_norm": 2.9656991958618164, + "learning_rate": 1.080422420796101e-06, + "loss": 1.568, + "step": 2221 + }, + { + "epoch": 2.435068493150685, + "grad_norm": 2.9461898803710938, + "learning_rate": 1.078391551584078e-06, + "loss": 1.5922, + "step": 2222 + }, + { + "epoch": 2.436164383561644, + "grad_norm": 2.937289237976074, + "learning_rate": 1.0763606823720553e-06, + "loss": 1.6414, + "step": 2223 + }, + { + "epoch": 2.4372602739726026, + "grad_norm": 2.8320538997650146, + "learning_rate": 1.0743298131600326e-06, + "loss": 1.5995, + "step": 2224 + }, + { + "epoch": 2.4383561643835616, + "grad_norm": 2.8221585750579834, + "learning_rate": 1.0722989439480098e-06, + "loss": 1.549, + "step": 2225 + }, + { + "epoch": 2.4394520547945207, + "grad_norm": 2.854454517364502, + "learning_rate": 1.070268074735987e-06, + "loss": 1.5474, + "step": 2226 + }, + { + "epoch": 2.4405479452054797, + "grad_norm": 2.7813878059387207, + "learning_rate": 1.0682372055239644e-06, + "loss": 1.5638, + "step": 2227 + }, + { + "epoch": 2.4416438356164383, + "grad_norm": 2.985661029815674, + "learning_rate": 1.0662063363119414e-06, + "loss": 1.5457, + "step": 2228 + }, + { + "epoch": 2.4427397260273973, + "grad_norm": 2.863631010055542, + "learning_rate": 1.0641754670999187e-06, + "loss": 1.5824, + "step": 2229 + }, + { + "epoch": 2.4438356164383563, + "grad_norm": 2.8328969478607178, + "learning_rate": 1.062144597887896e-06, + "loss": 1.5708, + "step": 2230 + }, + { + "epoch": 2.4438356164383563, + "eval_loss": 1.8261159658432007, + "eval_runtime": 50.0095, + "eval_samples_per_second": 19.996, + "eval_steps_per_second": 0.84, + "step": 2230 + }, + { + "epoch": 2.444931506849315, + "grad_norm": 2.9791696071624756, + "learning_rate": 1.0601137286758733e-06, + "loss": 1.6242, + "step": 2231 + }, + { + "epoch": 2.446027397260274, + "grad_norm": 2.7748308181762695, + "learning_rate": 1.0580828594638505e-06, + "loss": 1.499, + "step": 2232 + }, + { + "epoch": 2.447123287671233, + "grad_norm": 2.8511860370635986, + "learning_rate": 1.0560519902518278e-06, + "loss": 1.5787, + "step": 2233 + }, + { + "epoch": 2.4482191780821916, + "grad_norm": 2.802671194076538, + "learning_rate": 1.054021121039805e-06, + "loss": 1.4995, + "step": 2234 + }, + { + "epoch": 2.4493150684931506, + "grad_norm": 2.8005664348602295, + "learning_rate": 1.0519902518277824e-06, + "loss": 1.5083, + "step": 2235 + }, + { + "epoch": 2.4504109589041096, + "grad_norm": 2.92136287689209, + "learning_rate": 1.0499593826157596e-06, + "loss": 1.5807, + "step": 2236 + }, + { + "epoch": 2.4515068493150687, + "grad_norm": 2.850313186645508, + "learning_rate": 1.047928513403737e-06, + "loss": 1.5592, + "step": 2237 + }, + { + "epoch": 2.4526027397260273, + "grad_norm": 2.745370864868164, + "learning_rate": 1.0458976441917142e-06, + "loss": 1.495, + "step": 2238 + }, + { + "epoch": 2.4536986301369863, + "grad_norm": 2.794156789779663, + "learning_rate": 1.0438667749796915e-06, + "loss": 1.5024, + "step": 2239 + }, + { + "epoch": 2.4547945205479453, + "grad_norm": 2.8183324337005615, + "learning_rate": 1.0418359057676688e-06, + "loss": 1.5461, + "step": 2240 + }, + { + "epoch": 2.4547945205479453, + "eval_loss": 1.8265395164489746, + "eval_runtime": 47.1962, + "eval_samples_per_second": 21.188, + "eval_steps_per_second": 0.89, + "step": 2240 + }, + { + "epoch": 2.455890410958904, + "grad_norm": 2.863630771636963, + "learning_rate": 1.0398050365556458e-06, + "loss": 1.5977, + "step": 2241 + }, + { + "epoch": 2.456986301369863, + "grad_norm": 2.7688169479370117, + "learning_rate": 1.037774167343623e-06, + "loss": 1.5399, + "step": 2242 + }, + { + "epoch": 2.458082191780822, + "grad_norm": 2.8812880516052246, + "learning_rate": 1.0357432981316004e-06, + "loss": 1.5722, + "step": 2243 + }, + { + "epoch": 2.459178082191781, + "grad_norm": 3.2114691734313965, + "learning_rate": 1.0337124289195776e-06, + "loss": 1.6012, + "step": 2244 + }, + { + "epoch": 2.4602739726027396, + "grad_norm": 2.961271047592163, + "learning_rate": 1.031681559707555e-06, + "loss": 1.606, + "step": 2245 + }, + { + "epoch": 2.4613698630136986, + "grad_norm": 2.835495710372925, + "learning_rate": 1.0296506904955322e-06, + "loss": 1.5518, + "step": 2246 + }, + { + "epoch": 2.4624657534246577, + "grad_norm": 2.8796403408050537, + "learning_rate": 1.0276198212835095e-06, + "loss": 1.573, + "step": 2247 + }, + { + "epoch": 2.4635616438356163, + "grad_norm": 2.774409770965576, + "learning_rate": 1.0255889520714867e-06, + "loss": 1.5485, + "step": 2248 + }, + { + "epoch": 2.4646575342465753, + "grad_norm": 2.8847763538360596, + "learning_rate": 1.023558082859464e-06, + "loss": 1.5168, + "step": 2249 + }, + { + "epoch": 2.4657534246575343, + "grad_norm": 2.9533803462982178, + "learning_rate": 1.0215272136474413e-06, + "loss": 1.6179, + "step": 2250 + }, + { + "epoch": 2.4657534246575343, + "eval_loss": 1.8261420726776123, + "eval_runtime": 46.1356, + "eval_samples_per_second": 21.675, + "eval_steps_per_second": 0.91, + "step": 2250 + }, + { + "epoch": 2.466849315068493, + "grad_norm": 2.8762729167938232, + "learning_rate": 1.0194963444354186e-06, + "loss": 1.5109, + "step": 2251 + }, + { + "epoch": 2.467945205479452, + "grad_norm": 2.9228789806365967, + "learning_rate": 1.0174654752233958e-06, + "loss": 1.5861, + "step": 2252 + }, + { + "epoch": 2.469041095890411, + "grad_norm": 2.9467368125915527, + "learning_rate": 1.015434606011373e-06, + "loss": 1.5533, + "step": 2253 + }, + { + "epoch": 2.47013698630137, + "grad_norm": 2.9693222045898438, + "learning_rate": 1.0134037367993502e-06, + "loss": 1.5741, + "step": 2254 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 2.933368682861328, + "learning_rate": 1.0113728675873275e-06, + "loss": 1.5555, + "step": 2255 + }, + { + "epoch": 2.4723287671232876, + "grad_norm": 2.9675047397613525, + "learning_rate": 1.0093419983753047e-06, + "loss": 1.5803, + "step": 2256 + }, + { + "epoch": 2.4734246575342467, + "grad_norm": 2.931260108947754, + "learning_rate": 1.007311129163282e-06, + "loss": 1.5263, + "step": 2257 + }, + { + "epoch": 2.4745205479452057, + "grad_norm": 2.8160653114318848, + "learning_rate": 1.0052802599512593e-06, + "loss": 1.5281, + "step": 2258 + }, + { + "epoch": 2.4756164383561643, + "grad_norm": 3.0168352127075195, + "learning_rate": 1.0032493907392364e-06, + "loss": 1.5746, + "step": 2259 + }, + { + "epoch": 2.4767123287671233, + "grad_norm": 2.7623541355133057, + "learning_rate": 1.0012185215272136e-06, + "loss": 1.5181, + "step": 2260 + }, + { + "epoch": 2.4767123287671233, + "eval_loss": 1.8249253034591675, + "eval_runtime": 47.2027, + "eval_samples_per_second": 21.185, + "eval_steps_per_second": 0.89, + "step": 2260 + }, + { + "epoch": 2.4778082191780824, + "grad_norm": 2.8098902702331543, + "learning_rate": 9.99187652315191e-07, + "loss": 1.6099, + "step": 2261 + }, + { + "epoch": 2.478904109589041, + "grad_norm": 2.8562352657318115, + "learning_rate": 9.971567831031682e-07, + "loss": 1.5795, + "step": 2262 + }, + { + "epoch": 2.48, + "grad_norm": 2.7867422103881836, + "learning_rate": 9.951259138911455e-07, + "loss": 1.5104, + "step": 2263 + }, + { + "epoch": 2.481095890410959, + "grad_norm": 2.9658195972442627, + "learning_rate": 9.930950446791227e-07, + "loss": 1.5891, + "step": 2264 + }, + { + "epoch": 2.4821917808219176, + "grad_norm": 2.824355363845825, + "learning_rate": 9.910641754671e-07, + "loss": 1.6385, + "step": 2265 + }, + { + "epoch": 2.4832876712328766, + "grad_norm": 2.833772897720337, + "learning_rate": 9.890333062550773e-07, + "loss": 1.5759, + "step": 2266 + }, + { + "epoch": 2.4843835616438357, + "grad_norm": 2.9420840740203857, + "learning_rate": 9.870024370430546e-07, + "loss": 1.573, + "step": 2267 + }, + { + "epoch": 2.4854794520547947, + "grad_norm": 2.84629225730896, + "learning_rate": 9.849715678310318e-07, + "loss": 1.5606, + "step": 2268 + }, + { + "epoch": 2.4865753424657533, + "grad_norm": 2.7983481884002686, + "learning_rate": 9.829406986190091e-07, + "loss": 1.5718, + "step": 2269 + }, + { + "epoch": 2.4876712328767123, + "grad_norm": 2.8418283462524414, + "learning_rate": 9.809098294069864e-07, + "loss": 1.5052, + "step": 2270 + }, + { + "epoch": 2.4876712328767123, + "eval_loss": 1.8237050771713257, + "eval_runtime": 53.2157, + "eval_samples_per_second": 18.791, + "eval_steps_per_second": 0.789, + "step": 2270 + }, + { + "epoch": 2.4887671232876714, + "grad_norm": 2.8386685848236084, + "learning_rate": 9.788789601949637e-07, + "loss": 1.5638, + "step": 2271 + }, + { + "epoch": 2.48986301369863, + "grad_norm": 2.8796494007110596, + "learning_rate": 9.768480909829407e-07, + "loss": 1.6258, + "step": 2272 + }, + { + "epoch": 2.490958904109589, + "grad_norm": 2.8450164794921875, + "learning_rate": 9.74817221770918e-07, + "loss": 1.5567, + "step": 2273 + }, + { + "epoch": 2.492054794520548, + "grad_norm": 2.8501436710357666, + "learning_rate": 9.727863525588953e-07, + "loss": 1.5731, + "step": 2274 + }, + { + "epoch": 2.493150684931507, + "grad_norm": 2.8499205112457275, + "learning_rate": 9.707554833468726e-07, + "loss": 1.5383, + "step": 2275 + }, + { + "epoch": 2.4942465753424656, + "grad_norm": 2.904446601867676, + "learning_rate": 9.687246141348498e-07, + "loss": 1.5423, + "step": 2276 + }, + { + "epoch": 2.4953424657534247, + "grad_norm": 2.7885360717773438, + "learning_rate": 9.666937449228271e-07, + "loss": 1.5583, + "step": 2277 + }, + { + "epoch": 2.4964383561643837, + "grad_norm": 2.8230857849121094, + "learning_rate": 9.646628757108042e-07, + "loss": 1.531, + "step": 2278 + }, + { + "epoch": 2.4975342465753423, + "grad_norm": 2.9093189239501953, + "learning_rate": 9.626320064987814e-07, + "loss": 1.5901, + "step": 2279 + }, + { + "epoch": 2.4986301369863013, + "grad_norm": 2.853430986404419, + "learning_rate": 9.606011372867587e-07, + "loss": 1.5425, + "step": 2280 + }, + { + "epoch": 2.4986301369863013, + "eval_loss": 1.824331283569336, + "eval_runtime": 51.6196, + "eval_samples_per_second": 19.372, + "eval_steps_per_second": 0.814, + "step": 2280 + }, + { + "epoch": 2.4997260273972604, + "grad_norm": 2.857558488845825, + "learning_rate": 9.58570268074736e-07, + "loss": 1.555, + "step": 2281 + }, + { + "epoch": 2.500821917808219, + "grad_norm": 2.758384943008423, + "learning_rate": 9.565393988627133e-07, + "loss": 1.5546, + "step": 2282 + }, + { + "epoch": 2.501917808219178, + "grad_norm": 2.8295326232910156, + "learning_rate": 9.545085296506906e-07, + "loss": 1.5842, + "step": 2283 + }, + { + "epoch": 2.503013698630137, + "grad_norm": 2.855534315109253, + "learning_rate": 9.524776604386677e-07, + "loss": 1.5636, + "step": 2284 + }, + { + "epoch": 2.504109589041096, + "grad_norm": 2.7737905979156494, + "learning_rate": 9.50446791226645e-07, + "loss": 1.5101, + "step": 2285 + }, + { + "epoch": 2.5052054794520546, + "grad_norm": 2.7556989192962646, + "learning_rate": 9.484159220146223e-07, + "loss": 1.6216, + "step": 2286 + }, + { + "epoch": 2.5063013698630137, + "grad_norm": 2.8598365783691406, + "learning_rate": 9.463850528025995e-07, + "loss": 1.5818, + "step": 2287 + }, + { + "epoch": 2.5073972602739727, + "grad_norm": 3.0465996265411377, + "learning_rate": 9.443541835905768e-07, + "loss": 1.6237, + "step": 2288 + }, + { + "epoch": 2.5084931506849317, + "grad_norm": 3.067974328994751, + "learning_rate": 9.423233143785541e-07, + "loss": 1.6346, + "step": 2289 + }, + { + "epoch": 2.5095890410958903, + "grad_norm": 2.9208505153656006, + "learning_rate": 9.402924451665313e-07, + "loss": 1.543, + "step": 2290 + }, + { + "epoch": 2.5095890410958903, + "eval_loss": 1.8239408731460571, + "eval_runtime": 52.1634, + "eval_samples_per_second": 19.171, + "eval_steps_per_second": 0.805, + "step": 2290 + }, + { + "epoch": 2.5106849315068493, + "grad_norm": 3.0587332248687744, + "learning_rate": 9.382615759545085e-07, + "loss": 1.5814, + "step": 2291 + }, + { + "epoch": 2.5117808219178084, + "grad_norm": 2.9887471199035645, + "learning_rate": 9.362307067424858e-07, + "loss": 1.5481, + "step": 2292 + }, + { + "epoch": 2.512876712328767, + "grad_norm": 3.127073049545288, + "learning_rate": 9.341998375304631e-07, + "loss": 1.6121, + "step": 2293 + }, + { + "epoch": 2.513972602739726, + "grad_norm": 2.7613704204559326, + "learning_rate": 9.321689683184404e-07, + "loss": 1.5337, + "step": 2294 + }, + { + "epoch": 2.515068493150685, + "grad_norm": 2.9512479305267334, + "learning_rate": 9.301380991064176e-07, + "loss": 1.5161, + "step": 2295 + }, + { + "epoch": 2.5161643835616436, + "grad_norm": 2.9375245571136475, + "learning_rate": 9.281072298943949e-07, + "loss": 1.5538, + "step": 2296 + }, + { + "epoch": 2.5172602739726027, + "grad_norm": 3.8778791427612305, + "learning_rate": 9.260763606823721e-07, + "loss": 1.5736, + "step": 2297 + }, + { + "epoch": 2.5183561643835617, + "grad_norm": 3.176246404647827, + "learning_rate": 9.240454914703494e-07, + "loss": 1.5855, + "step": 2298 + }, + { + "epoch": 2.5194520547945203, + "grad_norm": 2.826871395111084, + "learning_rate": 9.220146222583266e-07, + "loss": 1.526, + "step": 2299 + }, + { + "epoch": 2.5205479452054793, + "grad_norm": 2.7983620166778564, + "learning_rate": 9.199837530463039e-07, + "loss": 1.5656, + "step": 2300 + }, + { + "epoch": 2.5205479452054793, + "eval_loss": 1.8233634233474731, + "eval_runtime": 51.1613, + "eval_samples_per_second": 19.546, + "eval_steps_per_second": 0.821, + "step": 2300 + }, + { + "epoch": 2.5216438356164383, + "grad_norm": 2.8449342250823975, + "learning_rate": 9.179528838342812e-07, + "loss": 1.5574, + "step": 2301 + }, + { + "epoch": 2.5227397260273974, + "grad_norm": 2.8809049129486084, + "learning_rate": 9.159220146222585e-07, + "loss": 1.5828, + "step": 2302 + }, + { + "epoch": 2.5238356164383564, + "grad_norm": 2.7337660789489746, + "learning_rate": 9.138911454102356e-07, + "loss": 1.5006, + "step": 2303 + }, + { + "epoch": 2.524931506849315, + "grad_norm": 2.7337660789489746, + "learning_rate": 9.138911454102356e-07, + "loss": 1.5592, + "step": 2304 + }, + { + "epoch": 2.526027397260274, + "grad_norm": 3.1104655265808105, + "learning_rate": 9.118602761982129e-07, + "loss": 1.5642, + "step": 2305 + }, + { + "epoch": 2.527123287671233, + "grad_norm": 4.62161922454834, + "learning_rate": 9.098294069861902e-07, + "loss": 1.5333, + "step": 2306 + }, + { + "epoch": 2.5282191780821917, + "grad_norm": 2.879892110824585, + "learning_rate": 9.077985377741675e-07, + "loss": 1.504, + "step": 2307 + }, + { + "epoch": 2.5293150684931507, + "grad_norm": 2.9683990478515625, + "learning_rate": 9.057676685621447e-07, + "loss": 1.6184, + "step": 2308 + }, + { + "epoch": 2.5304109589041097, + "grad_norm": 2.919980049133301, + "learning_rate": 9.03736799350122e-07, + "loss": 1.5908, + "step": 2309 + }, + { + "epoch": 2.5315068493150683, + "grad_norm": 3.0819506645202637, + "learning_rate": 9.017059301380991e-07, + "loss": 1.6252, + "step": 2310 + }, + { + "epoch": 2.5315068493150683, + "eval_loss": 1.8249709606170654, + "eval_runtime": 44.4295, + "eval_samples_per_second": 22.508, + "eval_steps_per_second": 0.945, + "step": 2310 + }, + { + "epoch": 2.5326027397260273, + "grad_norm": 3.0078251361846924, + "learning_rate": 8.996750609260764e-07, + "loss": 1.5653, + "step": 2311 + }, + { + "epoch": 2.5336986301369864, + "grad_norm": 3.0950827598571777, + "learning_rate": 8.976441917140536e-07, + "loss": 1.6341, + "step": 2312 + }, + { + "epoch": 2.534794520547945, + "grad_norm": 2.9090089797973633, + "learning_rate": 8.956133225020309e-07, + "loss": 1.5441, + "step": 2313 + }, + { + "epoch": 2.535890410958904, + "grad_norm": 2.8889381885528564, + "learning_rate": 8.935824532900082e-07, + "loss": 1.5371, + "step": 2314 + }, + { + "epoch": 2.536986301369863, + "grad_norm": 3.3039140701293945, + "learning_rate": 8.915515840779855e-07, + "loss": 1.5936, + "step": 2315 + }, + { + "epoch": 2.538082191780822, + "grad_norm": 2.9188146591186523, + "learning_rate": 8.895207148659626e-07, + "loss": 1.5787, + "step": 2316 + }, + { + "epoch": 2.539178082191781, + "grad_norm": 2.8546876907348633, + "learning_rate": 8.874898456539399e-07, + "loss": 1.58, + "step": 2317 + }, + { + "epoch": 2.5402739726027397, + "grad_norm": 2.8007283210754395, + "learning_rate": 8.854589764419172e-07, + "loss": 1.518, + "step": 2318 + }, + { + "epoch": 2.5413698630136987, + "grad_norm": 2.8198978900909424, + "learning_rate": 8.834281072298945e-07, + "loss": 1.5695, + "step": 2319 + }, + { + "epoch": 2.5424657534246577, + "grad_norm": 3.3483028411865234, + "learning_rate": 8.813972380178717e-07, + "loss": 1.5814, + "step": 2320 + }, + { + "epoch": 2.5424657534246577, + "eval_loss": 1.8233474493026733, + "eval_runtime": 49.2421, + "eval_samples_per_second": 20.308, + "eval_steps_per_second": 0.853, + "step": 2320 + }, + { + "epoch": 2.5435616438356163, + "grad_norm": 2.838481903076172, + "learning_rate": 8.79366368805849e-07, + "loss": 1.5311, + "step": 2321 + }, + { + "epoch": 2.5446575342465754, + "grad_norm": 2.8281280994415283, + "learning_rate": 8.773354995938262e-07, + "loss": 1.5235, + "step": 2322 + }, + { + "epoch": 2.5457534246575344, + "grad_norm": 2.8211021423339844, + "learning_rate": 8.753046303818035e-07, + "loss": 1.5484, + "step": 2323 + }, + { + "epoch": 2.546849315068493, + "grad_norm": 2.865307331085205, + "learning_rate": 8.732737611697807e-07, + "loss": 1.5892, + "step": 2324 + }, + { + "epoch": 2.547945205479452, + "grad_norm": 2.87202787399292, + "learning_rate": 8.71242891957758e-07, + "loss": 1.5961, + "step": 2325 + }, + { + "epoch": 2.549041095890411, + "grad_norm": 2.9286715984344482, + "learning_rate": 8.692120227457353e-07, + "loss": 1.5597, + "step": 2326 + }, + { + "epoch": 2.5501369863013696, + "grad_norm": 2.965538740158081, + "learning_rate": 8.671811535337126e-07, + "loss": 1.6278, + "step": 2327 + }, + { + "epoch": 2.5512328767123287, + "grad_norm": 2.991199254989624, + "learning_rate": 8.651502843216898e-07, + "loss": 1.5904, + "step": 2328 + }, + { + "epoch": 2.5523287671232877, + "grad_norm": 2.902418613433838, + "learning_rate": 8.63119415109667e-07, + "loss": 1.584, + "step": 2329 + }, + { + "epoch": 2.5534246575342463, + "grad_norm": 2.8904943466186523, + "learning_rate": 8.610885458976443e-07, + "loss": 1.5178, + "step": 2330 + }, + { + "epoch": 2.5534246575342463, + "eval_loss": 1.8210104703903198, + "eval_runtime": 49.456, + "eval_samples_per_second": 20.22, + "eval_steps_per_second": 0.849, + "step": 2330 + }, + { + "epoch": 2.5545205479452053, + "grad_norm": 2.778411865234375, + "learning_rate": 8.590576766856216e-07, + "loss": 1.5318, + "step": 2331 + }, + { + "epoch": 2.5556164383561644, + "grad_norm": 2.8281290531158447, + "learning_rate": 8.570268074735988e-07, + "loss": 1.5399, + "step": 2332 + }, + { + "epoch": 2.5567123287671234, + "grad_norm": 2.8783960342407227, + "learning_rate": 8.549959382615761e-07, + "loss": 1.5653, + "step": 2333 + }, + { + "epoch": 2.5578082191780824, + "grad_norm": 2.7188220024108887, + "learning_rate": 8.529650690495534e-07, + "loss": 1.5113, + "step": 2334 + }, + { + "epoch": 2.558904109589041, + "grad_norm": 2.705050468444824, + "learning_rate": 8.509341998375305e-07, + "loss": 1.4955, + "step": 2335 + }, + { + "epoch": 2.56, + "grad_norm": 2.798614740371704, + "learning_rate": 8.489033306255077e-07, + "loss": 1.529, + "step": 2336 + }, + { + "epoch": 2.561095890410959, + "grad_norm": 2.795304298400879, + "learning_rate": 8.46872461413485e-07, + "loss": 1.5431, + "step": 2337 + }, + { + "epoch": 2.5621917808219177, + "grad_norm": 2.838188648223877, + "learning_rate": 8.448415922014623e-07, + "loss": 1.5772, + "step": 2338 + }, + { + "epoch": 2.5632876712328767, + "grad_norm": 2.9334065914154053, + "learning_rate": 8.428107229894396e-07, + "loss": 1.5621, + "step": 2339 + }, + { + "epoch": 2.5643835616438357, + "grad_norm": 2.9324002265930176, + "learning_rate": 8.407798537774168e-07, + "loss": 1.5422, + "step": 2340 + }, + { + "epoch": 2.5643835616438357, + "eval_loss": 1.8223708868026733, + "eval_runtime": 50.7517, + "eval_samples_per_second": 19.704, + "eval_steps_per_second": 0.828, + "step": 2340 + }, + { + "epoch": 2.5654794520547943, + "grad_norm": 2.7721550464630127, + "learning_rate": 8.38748984565394e-07, + "loss": 1.5507, + "step": 2341 + }, + { + "epoch": 2.5665753424657534, + "grad_norm": 2.918529510498047, + "learning_rate": 8.367181153533713e-07, + "loss": 1.5703, + "step": 2342 + }, + { + "epoch": 2.5676712328767124, + "grad_norm": 2.926311492919922, + "learning_rate": 8.346872461413486e-07, + "loss": 1.5611, + "step": 2343 + }, + { + "epoch": 2.568767123287671, + "grad_norm": 2.9318113327026367, + "learning_rate": 8.326563769293258e-07, + "loss": 1.6348, + "step": 2344 + }, + { + "epoch": 2.56986301369863, + "grad_norm": 2.8003222942352295, + "learning_rate": 8.306255077173031e-07, + "loss": 1.5064, + "step": 2345 + }, + { + "epoch": 2.570958904109589, + "grad_norm": 2.797013998031616, + "learning_rate": 8.285946385052804e-07, + "loss": 1.5907, + "step": 2346 + }, + { + "epoch": 2.572054794520548, + "grad_norm": 2.7971713542938232, + "learning_rate": 8.265637692932575e-07, + "loss": 1.5844, + "step": 2347 + }, + { + "epoch": 2.573150684931507, + "grad_norm": 2.8959572315216064, + "learning_rate": 8.245329000812348e-07, + "loss": 1.5173, + "step": 2348 + }, + { + "epoch": 2.5742465753424657, + "grad_norm": 2.8146274089813232, + "learning_rate": 8.225020308692121e-07, + "loss": 1.5615, + "step": 2349 + }, + { + "epoch": 2.5753424657534247, + "grad_norm": 2.817338705062866, + "learning_rate": 8.204711616571894e-07, + "loss": 1.5076, + "step": 2350 + }, + { + "epoch": 2.5753424657534247, + "eval_loss": 1.8196357488632202, + "eval_runtime": 48.5561, + "eval_samples_per_second": 20.595, + "eval_steps_per_second": 0.865, + "step": 2350 + }, + { + "epoch": 2.5764383561643838, + "grad_norm": 2.9276442527770996, + "learning_rate": 8.184402924451667e-07, + "loss": 1.5776, + "step": 2351 + }, + { + "epoch": 2.5775342465753424, + "grad_norm": 2.860095739364624, + "learning_rate": 8.164094232331439e-07, + "loss": 1.5677, + "step": 2352 + }, + { + "epoch": 2.5786301369863014, + "grad_norm": 2.917368173599243, + "learning_rate": 8.14378554021121e-07, + "loss": 1.5489, + "step": 2353 + }, + { + "epoch": 2.5797260273972604, + "grad_norm": 2.8872427940368652, + "learning_rate": 8.123476848090983e-07, + "loss": 1.5458, + "step": 2354 + }, + { + "epoch": 2.580821917808219, + "grad_norm": 2.8507397174835205, + "learning_rate": 8.103168155970755e-07, + "loss": 1.5662, + "step": 2355 + }, + { + "epoch": 2.581917808219178, + "grad_norm": 2.8826792240142822, + "learning_rate": 8.082859463850528e-07, + "loss": 1.5574, + "step": 2356 + }, + { + "epoch": 2.583013698630137, + "grad_norm": 2.8728246688842773, + "learning_rate": 8.062550771730301e-07, + "loss": 1.5516, + "step": 2357 + }, + { + "epoch": 2.5841095890410957, + "grad_norm": 2.9101784229278564, + "learning_rate": 8.042242079610074e-07, + "loss": 1.5472, + "step": 2358 + }, + { + "epoch": 2.5852054794520547, + "grad_norm": 2.791224718093872, + "learning_rate": 8.021933387489846e-07, + "loss": 1.5663, + "step": 2359 + }, + { + "epoch": 2.5863013698630137, + "grad_norm": 2.983250856399536, + "learning_rate": 8.001624695369618e-07, + "loss": 1.5908, + "step": 2360 + }, + { + "epoch": 2.5863013698630137, + "eval_loss": 1.8226059675216675, + "eval_runtime": 48.2046, + "eval_samples_per_second": 20.745, + "eval_steps_per_second": 0.871, + "step": 2360 + }, + { + "epoch": 2.5873972602739728, + "grad_norm": 2.920762538909912, + "learning_rate": 7.981316003249391e-07, + "loss": 1.5111, + "step": 2361 + }, + { + "epoch": 2.5884931506849314, + "grad_norm": 2.983764171600342, + "learning_rate": 7.961007311129164e-07, + "loss": 1.566, + "step": 2362 + }, + { + "epoch": 2.5895890410958904, + "grad_norm": 2.976851463317871, + "learning_rate": 7.940698619008936e-07, + "loss": 1.6571, + "step": 2363 + }, + { + "epoch": 2.5906849315068494, + "grad_norm": 2.856072187423706, + "learning_rate": 7.920389926888709e-07, + "loss": 1.5176, + "step": 2364 + }, + { + "epoch": 2.5917808219178085, + "grad_norm": 2.8299734592437744, + "learning_rate": 7.900081234768482e-07, + "loss": 1.587, + "step": 2365 + }, + { + "epoch": 2.592876712328767, + "grad_norm": 2.937560558319092, + "learning_rate": 7.879772542648254e-07, + "loss": 1.5769, + "step": 2366 + }, + { + "epoch": 2.593972602739726, + "grad_norm": 2.8437840938568115, + "learning_rate": 7.859463850528026e-07, + "loss": 1.5665, + "step": 2367 + }, + { + "epoch": 2.595068493150685, + "grad_norm": 3.0254647731781006, + "learning_rate": 7.839155158407799e-07, + "loss": 1.5783, + "step": 2368 + }, + { + "epoch": 2.5961643835616437, + "grad_norm": 2.895967960357666, + "learning_rate": 7.818846466287572e-07, + "loss": 1.5783, + "step": 2369 + }, + { + "epoch": 2.5972602739726027, + "grad_norm": 2.8868582248687744, + "learning_rate": 7.798537774167345e-07, + "loss": 1.527, + "step": 2370 + }, + { + "epoch": 2.5972602739726027, + "eval_loss": 1.8213152885437012, + "eval_runtime": 51.0508, + "eval_samples_per_second": 19.588, + "eval_steps_per_second": 0.823, + "step": 2370 + }, + { + "epoch": 2.5983561643835618, + "grad_norm": 2.906033992767334, + "learning_rate": 7.778229082047117e-07, + "loss": 1.5688, + "step": 2371 + }, + { + "epoch": 2.5994520547945204, + "grad_norm": 2.846966505050659, + "learning_rate": 7.757920389926889e-07, + "loss": 1.5184, + "step": 2372 + }, + { + "epoch": 2.6005479452054794, + "grad_norm": 3.124784231185913, + "learning_rate": 7.737611697806662e-07, + "loss": 1.5953, + "step": 2373 + }, + { + "epoch": 2.6016438356164384, + "grad_norm": 3.014098644256592, + "learning_rate": 7.717303005686435e-07, + "loss": 1.6318, + "step": 2374 + }, + { + "epoch": 2.602739726027397, + "grad_norm": 2.82450532913208, + "learning_rate": 7.696994313566207e-07, + "loss": 1.5106, + "step": 2375 + }, + { + "epoch": 2.603835616438356, + "grad_norm": 2.8513712882995605, + "learning_rate": 7.67668562144598e-07, + "loss": 1.5053, + "step": 2376 + }, + { + "epoch": 2.604931506849315, + "grad_norm": 2.99210786819458, + "learning_rate": 7.656376929325753e-07, + "loss": 1.5844, + "step": 2377 + }, + { + "epoch": 2.606027397260274, + "grad_norm": 2.858902931213379, + "learning_rate": 7.636068237205524e-07, + "loss": 1.5156, + "step": 2378 + }, + { + "epoch": 2.607123287671233, + "grad_norm": 2.9171338081359863, + "learning_rate": 7.615759545085296e-07, + "loss": 1.5474, + "step": 2379 + }, + { + "epoch": 2.6082191780821917, + "grad_norm": 2.868882417678833, + "learning_rate": 7.595450852965069e-07, + "loss": 1.5826, + "step": 2380 + }, + { + "epoch": 2.6082191780821917, + "eval_loss": 1.8230814933776855, + "eval_runtime": 46.2884, + "eval_samples_per_second": 21.604, + "eval_steps_per_second": 0.907, + "step": 2380 + }, + { + "epoch": 2.6093150684931508, + "grad_norm": 2.8466811180114746, + "learning_rate": 7.575142160844842e-07, + "loss": 1.5272, + "step": 2381 + }, + { + "epoch": 2.61041095890411, + "grad_norm": 2.8146281242370605, + "learning_rate": 7.554833468724615e-07, + "loss": 1.5303, + "step": 2382 + }, + { + "epoch": 2.6115068493150684, + "grad_norm": 2.8843190670013428, + "learning_rate": 7.534524776604387e-07, + "loss": 1.5767, + "step": 2383 + }, + { + "epoch": 2.6126027397260274, + "grad_norm": 2.988900899887085, + "learning_rate": 7.51421608448416e-07, + "loss": 1.5335, + "step": 2384 + }, + { + "epoch": 2.6136986301369864, + "grad_norm": 3.4128847122192383, + "learning_rate": 7.493907392363932e-07, + "loss": 1.485, + "step": 2385 + }, + { + "epoch": 2.614794520547945, + "grad_norm": 2.8596994876861572, + "learning_rate": 7.473598700243705e-07, + "loss": 1.5597, + "step": 2386 + }, + { + "epoch": 2.615890410958904, + "grad_norm": 2.923269748687744, + "learning_rate": 7.453290008123477e-07, + "loss": 1.5736, + "step": 2387 + }, + { + "epoch": 2.616986301369863, + "grad_norm": 2.8989574909210205, + "learning_rate": 7.43298131600325e-07, + "loss": 1.5458, + "step": 2388 + }, + { + "epoch": 2.6180821917808217, + "grad_norm": 2.935607671737671, + "learning_rate": 7.412672623883023e-07, + "loss": 1.4983, + "step": 2389 + }, + { + "epoch": 2.6191780821917807, + "grad_norm": 3.215050220489502, + "learning_rate": 7.392363931762796e-07, + "loss": 1.5571, + "step": 2390 + }, + { + "epoch": 2.6191780821917807, + "eval_loss": 1.8221906423568726, + "eval_runtime": 47.9911, + "eval_samples_per_second": 20.837, + "eval_steps_per_second": 0.875, + "step": 2390 + }, + { + "epoch": 2.6202739726027398, + "grad_norm": 3.0844898223876953, + "learning_rate": 7.372055239642567e-07, + "loss": 1.5767, + "step": 2391 + }, + { + "epoch": 2.621369863013699, + "grad_norm": 3.113297700881958, + "learning_rate": 7.35174654752234e-07, + "loss": 1.5895, + "step": 2392 + }, + { + "epoch": 2.6224657534246574, + "grad_norm": 2.904573917388916, + "learning_rate": 7.331437855402113e-07, + "loss": 1.5162, + "step": 2393 + }, + { + "epoch": 2.6235616438356164, + "grad_norm": 2.8379499912261963, + "learning_rate": 7.311129163281886e-07, + "loss": 1.5901, + "step": 2394 + }, + { + "epoch": 2.6246575342465754, + "grad_norm": 2.9121007919311523, + "learning_rate": 7.290820471161658e-07, + "loss": 1.5908, + "step": 2395 + }, + { + "epoch": 2.6257534246575345, + "grad_norm": 2.8151888847351074, + "learning_rate": 7.270511779041431e-07, + "loss": 1.5163, + "step": 2396 + }, + { + "epoch": 2.626849315068493, + "grad_norm": 2.9227914810180664, + "learning_rate": 7.250203086921203e-07, + "loss": 1.5466, + "step": 2397 + }, + { + "epoch": 2.627945205479452, + "grad_norm": 2.834669589996338, + "learning_rate": 7.229894394800976e-07, + "loss": 1.5724, + "step": 2398 + }, + { + "epoch": 2.629041095890411, + "grad_norm": 2.7271687984466553, + "learning_rate": 7.209585702680748e-07, + "loss": 1.5099, + "step": 2399 + }, + { + "epoch": 2.6301369863013697, + "grad_norm": 2.7215335369110107, + "learning_rate": 7.189277010560521e-07, + "loss": 1.5419, + "step": 2400 + }, + { + "epoch": 2.6301369863013697, + "eval_loss": 1.8201849460601807, + "eval_runtime": 50.7567, + "eval_samples_per_second": 19.702, + "eval_steps_per_second": 0.827, + "step": 2400 + }, + { + "epoch": 2.6312328767123287, + "grad_norm": 2.840486764907837, + "learning_rate": 7.168968318440294e-07, + "loss": 1.5203, + "step": 2401 + }, + { + "epoch": 2.632328767123288, + "grad_norm": 2.885180711746216, + "learning_rate": 7.148659626320067e-07, + "loss": 1.6037, + "step": 2402 + }, + { + "epoch": 2.6334246575342464, + "grad_norm": 2.9077260494232178, + "learning_rate": 7.128350934199837e-07, + "loss": 1.587, + "step": 2403 + }, + { + "epoch": 2.6345205479452054, + "grad_norm": 2.781118154525757, + "learning_rate": 7.10804224207961e-07, + "loss": 1.545, + "step": 2404 + }, + { + "epoch": 2.6356164383561644, + "grad_norm": 2.815735101699829, + "learning_rate": 7.087733549959383e-07, + "loss": 1.5327, + "step": 2405 + }, + { + "epoch": 2.636712328767123, + "grad_norm": 2.9639029502868652, + "learning_rate": 7.067424857839155e-07, + "loss": 1.5591, + "step": 2406 + }, + { + "epoch": 2.637808219178082, + "grad_norm": 2.9713387489318848, + "learning_rate": 7.047116165718928e-07, + "loss": 1.5416, + "step": 2407 + }, + { + "epoch": 2.638904109589041, + "grad_norm": 3.1526248455047607, + "learning_rate": 7.026807473598701e-07, + "loss": 1.607, + "step": 2408 + }, + { + "epoch": 2.64, + "grad_norm": 3.009585380554199, + "learning_rate": 7.006498781478473e-07, + "loss": 1.5223, + "step": 2409 + }, + { + "epoch": 2.641095890410959, + "grad_norm": 2.7750561237335205, + "learning_rate": 6.986190089358245e-07, + "loss": 1.5663, + "step": 2410 + }, + { + "epoch": 2.641095890410959, + "eval_loss": 1.8215185403823853, + "eval_runtime": 44.6832, + "eval_samples_per_second": 22.38, + "eval_steps_per_second": 0.94, + "step": 2410 + }, + { + "epoch": 2.6421917808219177, + "grad_norm": 3.1443002223968506, + "learning_rate": 6.965881397238018e-07, + "loss": 1.5878, + "step": 2411 + }, + { + "epoch": 2.643287671232877, + "grad_norm": 2.8203115463256836, + "learning_rate": 6.945572705117791e-07, + "loss": 1.5749, + "step": 2412 + }, + { + "epoch": 2.644383561643836, + "grad_norm": 2.878829002380371, + "learning_rate": 6.925264012997564e-07, + "loss": 1.5788, + "step": 2413 + }, + { + "epoch": 2.6454794520547944, + "grad_norm": 2.9213924407958984, + "learning_rate": 6.904955320877336e-07, + "loss": 1.57, + "step": 2414 + }, + { + "epoch": 2.6465753424657534, + "grad_norm": 3.0402517318725586, + "learning_rate": 6.884646628757109e-07, + "loss": 1.6245, + "step": 2415 + }, + { + "epoch": 2.6476712328767125, + "grad_norm": 2.874117612838745, + "learning_rate": 6.864337936636881e-07, + "loss": 1.6213, + "step": 2416 + }, + { + "epoch": 2.648767123287671, + "grad_norm": 2.841010570526123, + "learning_rate": 6.844029244516654e-07, + "loss": 1.5244, + "step": 2417 + }, + { + "epoch": 2.64986301369863, + "grad_norm": 2.8916752338409424, + "learning_rate": 6.823720552396426e-07, + "loss": 1.5824, + "step": 2418 + }, + { + "epoch": 2.650958904109589, + "grad_norm": 2.8366785049438477, + "learning_rate": 6.803411860276199e-07, + "loss": 1.5663, + "step": 2419 + }, + { + "epoch": 2.6520547945205477, + "grad_norm": 2.881715774536133, + "learning_rate": 6.783103168155972e-07, + "loss": 1.5599, + "step": 2420 + }, + { + "epoch": 2.6520547945205477, + "eval_loss": 1.8196642398834229, + "eval_runtime": 50.2214, + "eval_samples_per_second": 19.912, + "eval_steps_per_second": 0.836, + "step": 2420 + }, + { + "epoch": 2.6531506849315067, + "grad_norm": 2.83880615234375, + "learning_rate": 6.762794476035745e-07, + "loss": 1.5839, + "step": 2421 + }, + { + "epoch": 2.6542465753424658, + "grad_norm": 2.8980135917663574, + "learning_rate": 6.742485783915515e-07, + "loss": 1.5898, + "step": 2422 + }, + { + "epoch": 2.655342465753425, + "grad_norm": 2.82328724861145, + "learning_rate": 6.722177091795288e-07, + "loss": 1.5778, + "step": 2423 + }, + { + "epoch": 2.6564383561643834, + "grad_norm": 2.7781100273132324, + "learning_rate": 6.701868399675061e-07, + "loss": 1.5381, + "step": 2424 + }, + { + "epoch": 2.6575342465753424, + "grad_norm": 2.94514536857605, + "learning_rate": 6.681559707554834e-07, + "loss": 1.5305, + "step": 2425 + }, + { + "epoch": 2.6586301369863015, + "grad_norm": 2.9485371112823486, + "learning_rate": 6.661251015434606e-07, + "loss": 1.5348, + "step": 2426 + }, + { + "epoch": 2.6597260273972605, + "grad_norm": 2.8444228172302246, + "learning_rate": 6.640942323314379e-07, + "loss": 1.5651, + "step": 2427 + }, + { + "epoch": 2.660821917808219, + "grad_norm": 2.923813819885254, + "learning_rate": 6.620633631194151e-07, + "loss": 1.5827, + "step": 2428 + }, + { + "epoch": 2.661917808219178, + "grad_norm": 2.8742222785949707, + "learning_rate": 6.600324939073924e-07, + "loss": 1.5464, + "step": 2429 + }, + { + "epoch": 2.663013698630137, + "grad_norm": 2.963705539703369, + "learning_rate": 6.580016246953696e-07, + "loss": 1.615, + "step": 2430 + }, + { + "epoch": 2.663013698630137, + "eval_loss": 1.8202346563339233, + "eval_runtime": 53.0891, + "eval_samples_per_second": 18.836, + "eval_steps_per_second": 0.791, + "step": 2430 + }, + { + "epoch": 2.6641095890410957, + "grad_norm": 2.9104220867156982, + "learning_rate": 6.559707554833469e-07, + "loss": 1.5432, + "step": 2431 + }, + { + "epoch": 2.6652054794520548, + "grad_norm": 2.9200713634490967, + "learning_rate": 6.539398862713242e-07, + "loss": 1.5988, + "step": 2432 + }, + { + "epoch": 2.666301369863014, + "grad_norm": 2.861328363418579, + "learning_rate": 6.519090170593015e-07, + "loss": 1.5721, + "step": 2433 + }, + { + "epoch": 2.6673972602739724, + "grad_norm": 2.9656898975372314, + "learning_rate": 6.498781478472786e-07, + "loss": 1.598, + "step": 2434 + }, + { + "epoch": 2.6684931506849314, + "grad_norm": 2.9453625679016113, + "learning_rate": 6.478472786352559e-07, + "loss": 1.564, + "step": 2435 + }, + { + "epoch": 2.6695890410958905, + "grad_norm": 2.8961451053619385, + "learning_rate": 6.458164094232332e-07, + "loss": 1.5021, + "step": 2436 + }, + { + "epoch": 2.670684931506849, + "grad_norm": 2.8768153190612793, + "learning_rate": 6.437855402112105e-07, + "loss": 1.5226, + "step": 2437 + }, + { + "epoch": 2.671780821917808, + "grad_norm": 2.942457914352417, + "learning_rate": 6.417546709991877e-07, + "loss": 1.5903, + "step": 2438 + }, + { + "epoch": 2.672876712328767, + "grad_norm": 2.82218074798584, + "learning_rate": 6.39723801787165e-07, + "loss": 1.5596, + "step": 2439 + }, + { + "epoch": 2.673972602739726, + "grad_norm": 2.828130006790161, + "learning_rate": 6.376929325751423e-07, + "loss": 1.5491, + "step": 2440 + }, + { + "epoch": 2.673972602739726, + "eval_loss": 1.820644736289978, + "eval_runtime": 48.3133, + "eval_samples_per_second": 20.698, + "eval_steps_per_second": 0.869, + "step": 2440 + }, + { + "epoch": 2.675068493150685, + "grad_norm": 2.8218741416931152, + "learning_rate": 6.356620633631195e-07, + "loss": 1.5274, + "step": 2441 + }, + { + "epoch": 2.6761643835616438, + "grad_norm": 3.038788080215454, + "learning_rate": 6.336311941510967e-07, + "loss": 1.5708, + "step": 2442 + }, + { + "epoch": 2.677260273972603, + "grad_norm": 3.0761072635650635, + "learning_rate": 6.31600324939074e-07, + "loss": 1.6051, + "step": 2443 + }, + { + "epoch": 2.678356164383562, + "grad_norm": 2.881915807723999, + "learning_rate": 6.295694557270513e-07, + "loss": 1.5661, + "step": 2444 + }, + { + "epoch": 2.6794520547945204, + "grad_norm": 2.8945391178131104, + "learning_rate": 6.275385865150286e-07, + "loss": 1.5515, + "step": 2445 + }, + { + "epoch": 2.6805479452054795, + "grad_norm": 2.8491926193237305, + "learning_rate": 6.255077173030058e-07, + "loss": 1.5675, + "step": 2446 + }, + { + "epoch": 2.6816438356164385, + "grad_norm": 2.860481023788452, + "learning_rate": 6.23476848090983e-07, + "loss": 1.5485, + "step": 2447 + }, + { + "epoch": 2.682739726027397, + "grad_norm": 2.8958401679992676, + "learning_rate": 6.214459788789602e-07, + "loss": 1.5927, + "step": 2448 + }, + { + "epoch": 2.683835616438356, + "grad_norm": 2.8343098163604736, + "learning_rate": 6.194151096669375e-07, + "loss": 1.5443, + "step": 2449 + }, + { + "epoch": 2.684931506849315, + "grad_norm": 2.8697097301483154, + "learning_rate": 6.173842404549147e-07, + "loss": 1.5521, + "step": 2450 + }, + { + "epoch": 2.684931506849315, + "eval_loss": 1.8198002576828003, + "eval_runtime": 50.7198, + "eval_samples_per_second": 19.716, + "eval_steps_per_second": 0.828, + "step": 2450 + }, + { + "epoch": 2.6860273972602737, + "grad_norm": 2.8190839290618896, + "learning_rate": 6.15353371242892e-07, + "loss": 1.4845, + "step": 2451 + }, + { + "epoch": 2.6871232876712328, + "grad_norm": 2.7570931911468506, + "learning_rate": 6.133225020308693e-07, + "loss": 1.5452, + "step": 2452 + }, + { + "epoch": 2.688219178082192, + "grad_norm": 2.960784673690796, + "learning_rate": 6.112916328188466e-07, + "loss": 1.6249, + "step": 2453 + }, + { + "epoch": 2.689315068493151, + "grad_norm": 2.7679786682128906, + "learning_rate": 6.092607636068238e-07, + "loss": 1.5327, + "step": 2454 + }, + { + "epoch": 2.69041095890411, + "grad_norm": 2.8044891357421875, + "learning_rate": 6.07229894394801e-07, + "loss": 1.5435, + "step": 2455 + }, + { + "epoch": 2.6915068493150685, + "grad_norm": 2.8200135231018066, + "learning_rate": 6.051990251827783e-07, + "loss": 1.5759, + "step": 2456 + }, + { + "epoch": 2.6926027397260275, + "grad_norm": 2.904179096221924, + "learning_rate": 6.031681559707556e-07, + "loss": 1.5597, + "step": 2457 + }, + { + "epoch": 2.6936986301369865, + "grad_norm": 2.8314926624298096, + "learning_rate": 6.011372867587327e-07, + "loss": 1.5394, + "step": 2458 + }, + { + "epoch": 2.694794520547945, + "grad_norm": 2.8240511417388916, + "learning_rate": 5.9910641754671e-07, + "loss": 1.5517, + "step": 2459 + }, + { + "epoch": 2.695890410958904, + "grad_norm": 2.844630718231201, + "learning_rate": 5.970755483346873e-07, + "loss": 1.5426, + "step": 2460 + }, + { + "epoch": 2.695890410958904, + "eval_loss": 1.8208848237991333, + "eval_runtime": 52.7964, + "eval_samples_per_second": 18.941, + "eval_steps_per_second": 0.796, + "step": 2460 + }, + { + "epoch": 2.696986301369863, + "grad_norm": 2.926901340484619, + "learning_rate": 5.950446791226646e-07, + "loss": 1.5457, + "step": 2461 + }, + { + "epoch": 2.6980821917808218, + "grad_norm": 2.845421314239502, + "learning_rate": 5.930138099106418e-07, + "loss": 1.5397, + "step": 2462 + }, + { + "epoch": 2.699178082191781, + "grad_norm": 2.984661817550659, + "learning_rate": 5.909829406986191e-07, + "loss": 1.5649, + "step": 2463 + }, + { + "epoch": 2.70027397260274, + "grad_norm": 2.8373823165893555, + "learning_rate": 5.889520714865963e-07, + "loss": 1.5255, + "step": 2464 + }, + { + "epoch": 2.7013698630136984, + "grad_norm": 2.9435815811157227, + "learning_rate": 5.869212022745735e-07, + "loss": 1.5705, + "step": 2465 + }, + { + "epoch": 2.7024657534246574, + "grad_norm": 2.8383870124816895, + "learning_rate": 5.848903330625508e-07, + "loss": 1.5457, + "step": 2466 + }, + { + "epoch": 2.7035616438356165, + "grad_norm": 2.899071216583252, + "learning_rate": 5.828594638505281e-07, + "loss": 1.5474, + "step": 2467 + }, + { + "epoch": 2.704657534246575, + "grad_norm": 2.9021337032318115, + "learning_rate": 5.808285946385054e-07, + "loss": 1.5549, + "step": 2468 + }, + { + "epoch": 2.705753424657534, + "grad_norm": 2.9504072666168213, + "learning_rate": 5.787977254264827e-07, + "loss": 1.6014, + "step": 2469 + }, + { + "epoch": 2.706849315068493, + "grad_norm": 2.9223573207855225, + "learning_rate": 5.767668562144598e-07, + "loss": 1.5574, + "step": 2470 + }, + { + "epoch": 2.706849315068493, + "eval_loss": 1.8203133344650269, + "eval_runtime": 49.5424, + "eval_samples_per_second": 20.185, + "eval_steps_per_second": 0.848, + "step": 2470 + }, + { + "epoch": 2.707945205479452, + "grad_norm": 2.8095791339874268, + "learning_rate": 5.747359870024371e-07, + "loss": 1.5462, + "step": 2471 + }, + { + "epoch": 2.709041095890411, + "grad_norm": 2.8600597381591797, + "learning_rate": 5.727051177904144e-07, + "loss": 1.5872, + "step": 2472 + }, + { + "epoch": 2.71013698630137, + "grad_norm": 2.8269565105438232, + "learning_rate": 5.706742485783915e-07, + "loss": 1.5218, + "step": 2473 + }, + { + "epoch": 2.711232876712329, + "grad_norm": 2.7480826377868652, + "learning_rate": 5.686433793663688e-07, + "loss": 1.5049, + "step": 2474 + }, + { + "epoch": 2.712328767123288, + "grad_norm": 2.9086508750915527, + "learning_rate": 5.666125101543461e-07, + "loss": 1.604, + "step": 2475 + }, + { + "epoch": 2.7134246575342464, + "grad_norm": 2.791057825088501, + "learning_rate": 5.645816409423234e-07, + "loss": 1.5647, + "step": 2476 + }, + { + "epoch": 2.7145205479452055, + "grad_norm": 2.8584253787994385, + "learning_rate": 5.625507717303006e-07, + "loss": 1.5477, + "step": 2477 + }, + { + "epoch": 2.7156164383561645, + "grad_norm": 2.8320658206939697, + "learning_rate": 5.605199025182779e-07, + "loss": 1.5245, + "step": 2478 + }, + { + "epoch": 2.716712328767123, + "grad_norm": 2.867785930633545, + "learning_rate": 5.584890333062551e-07, + "loss": 1.6039, + "step": 2479 + }, + { + "epoch": 2.717808219178082, + "grad_norm": 2.942857027053833, + "learning_rate": 5.564581640942324e-07, + "loss": 1.5313, + "step": 2480 + }, + { + "epoch": 2.717808219178082, + "eval_loss": 1.821783423423767, + "eval_runtime": 51.124, + "eval_samples_per_second": 19.56, + "eval_steps_per_second": 0.822, + "step": 2480 + }, + { + "epoch": 2.718904109589041, + "grad_norm": 2.8085427284240723, + "learning_rate": 5.544272948822096e-07, + "loss": 1.5575, + "step": 2481 + }, + { + "epoch": 2.7199999999999998, + "grad_norm": 2.8479437828063965, + "learning_rate": 5.523964256701868e-07, + "loss": 1.5816, + "step": 2482 + }, + { + "epoch": 2.721095890410959, + "grad_norm": 2.8424737453460693, + "learning_rate": 5.503655564581641e-07, + "loss": 1.5434, + "step": 2483 + }, + { + "epoch": 2.722191780821918, + "grad_norm": 2.8704795837402344, + "learning_rate": 5.483346872461414e-07, + "loss": 1.5565, + "step": 2484 + }, + { + "epoch": 2.723287671232877, + "grad_norm": 2.868283987045288, + "learning_rate": 5.463038180341186e-07, + "loss": 1.5173, + "step": 2485 + }, + { + "epoch": 2.724383561643836, + "grad_norm": 2.8845019340515137, + "learning_rate": 5.442729488220959e-07, + "loss": 1.5581, + "step": 2486 + }, + { + "epoch": 2.7254794520547945, + "grad_norm": 2.83331036567688, + "learning_rate": 5.422420796100732e-07, + "loss": 1.5502, + "step": 2487 + }, + { + "epoch": 2.7265753424657535, + "grad_norm": 2.8796298503875732, + "learning_rate": 5.402112103980505e-07, + "loss": 1.5737, + "step": 2488 + }, + { + "epoch": 2.7276712328767125, + "grad_norm": 2.939396381378174, + "learning_rate": 5.381803411860276e-07, + "loss": 1.6211, + "step": 2489 + }, + { + "epoch": 2.728767123287671, + "grad_norm": 2.8413333892822266, + "learning_rate": 5.361494719740049e-07, + "loss": 1.5263, + "step": 2490 + }, + { + "epoch": 2.728767123287671, + "eval_loss": 1.819288969039917, + "eval_runtime": 50.1547, + "eval_samples_per_second": 19.938, + "eval_steps_per_second": 0.837, + "step": 2490 + }, + { + "epoch": 2.72986301369863, + "grad_norm": 2.9400246143341064, + "learning_rate": 5.341186027619822e-07, + "loss": 1.5439, + "step": 2491 + }, + { + "epoch": 2.730958904109589, + "grad_norm": 2.8259570598602295, + "learning_rate": 5.320877335499594e-07, + "loss": 1.6046, + "step": 2492 + }, + { + "epoch": 2.732054794520548, + "grad_norm": 2.9316084384918213, + "learning_rate": 5.300568643379366e-07, + "loss": 1.569, + "step": 2493 + }, + { + "epoch": 2.733150684931507, + "grad_norm": 2.8864996433258057, + "learning_rate": 5.280259951259139e-07, + "loss": 1.5384, + "step": 2494 + }, + { + "epoch": 2.734246575342466, + "grad_norm": 3.0581302642822266, + "learning_rate": 5.259951259138912e-07, + "loss": 1.5318, + "step": 2495 + }, + { + "epoch": 2.7353424657534244, + "grad_norm": 2.8676114082336426, + "learning_rate": 5.239642567018685e-07, + "loss": 1.543, + "step": 2496 + }, + { + "epoch": 2.7364383561643835, + "grad_norm": 2.935861110687256, + "learning_rate": 5.219333874898457e-07, + "loss": 1.5603, + "step": 2497 + }, + { + "epoch": 2.7375342465753425, + "grad_norm": 2.963350296020508, + "learning_rate": 5.199025182778229e-07, + "loss": 1.5479, + "step": 2498 + }, + { + "epoch": 2.7386301369863015, + "grad_norm": 2.9598515033721924, + "learning_rate": 5.178716490658002e-07, + "loss": 1.5116, + "step": 2499 + }, + { + "epoch": 2.73972602739726, + "grad_norm": 2.7877373695373535, + "learning_rate": 5.158407798537775e-07, + "loss": 1.4805, + "step": 2500 + }, + { + "epoch": 2.73972602739726, + "eval_loss": 1.8200916051864624, + "eval_runtime": 50.4013, + "eval_samples_per_second": 19.841, + "eval_steps_per_second": 0.833, + "step": 2500 + }, + { + "epoch": 2.740821917808219, + "grad_norm": 2.9831666946411133, + "learning_rate": 5.138099106417547e-07, + "loss": 1.5455, + "step": 2501 + }, + { + "epoch": 2.741917808219178, + "grad_norm": 2.9333431720733643, + "learning_rate": 5.11779041429732e-07, + "loss": 1.5281, + "step": 2502 + }, + { + "epoch": 2.743013698630137, + "grad_norm": 2.907478094100952, + "learning_rate": 5.097481722177093e-07, + "loss": 1.6127, + "step": 2503 + }, + { + "epoch": 2.744109589041096, + "grad_norm": 2.958024024963379, + "learning_rate": 5.077173030056865e-07, + "loss": 1.601, + "step": 2504 + }, + { + "epoch": 2.745205479452055, + "grad_norm": 3.0101583003997803, + "learning_rate": 5.056864337936637e-07, + "loss": 1.5734, + "step": 2505 + }, + { + "epoch": 2.746301369863014, + "grad_norm": 2.9242193698883057, + "learning_rate": 5.03655564581641e-07, + "loss": 1.6734, + "step": 2506 + }, + { + "epoch": 2.7473972602739725, + "grad_norm": 3.003774404525757, + "learning_rate": 5.016246953696182e-07, + "loss": 1.5234, + "step": 2507 + }, + { + "epoch": 2.7484931506849315, + "grad_norm": 2.88041353225708, + "learning_rate": 4.995938261575955e-07, + "loss": 1.5705, + "step": 2508 + }, + { + "epoch": 2.7495890410958905, + "grad_norm": 2.9543464183807373, + "learning_rate": 4.975629569455727e-07, + "loss": 1.6155, + "step": 2509 + }, + { + "epoch": 2.750684931506849, + "grad_norm": 2.866711139678955, + "learning_rate": 4.9553208773355e-07, + "loss": 1.5469, + "step": 2510 + }, + { + "epoch": 2.750684931506849, + "eval_loss": 1.817280888557434, + "eval_runtime": 45.7924, + "eval_samples_per_second": 21.838, + "eval_steps_per_second": 0.917, + "step": 2510 + }, + { + "epoch": 2.751780821917808, + "grad_norm": 2.94659161567688, + "learning_rate": 4.935012185215273e-07, + "loss": 1.5198, + "step": 2511 + }, + { + "epoch": 2.752876712328767, + "grad_norm": 2.782384157180786, + "learning_rate": 4.914703493095046e-07, + "loss": 1.5286, + "step": 2512 + }, + { + "epoch": 2.7539726027397258, + "grad_norm": 2.86202335357666, + "learning_rate": 4.894394800974818e-07, + "loss": 1.6437, + "step": 2513 + }, + { + "epoch": 2.755068493150685, + "grad_norm": 2.9339122772216797, + "learning_rate": 4.87408610885459e-07, + "loss": 1.5821, + "step": 2514 + }, + { + "epoch": 2.756164383561644, + "grad_norm": 2.883612871170044, + "learning_rate": 4.853777416734363e-07, + "loss": 1.5605, + "step": 2515 + }, + { + "epoch": 2.757260273972603, + "grad_norm": 2.861711025238037, + "learning_rate": 4.833468724614136e-07, + "loss": 1.5416, + "step": 2516 + }, + { + "epoch": 2.758356164383562, + "grad_norm": 2.764655590057373, + "learning_rate": 4.813160032493907e-07, + "loss": 1.5159, + "step": 2517 + }, + { + "epoch": 2.7594520547945205, + "grad_norm": 2.880544900894165, + "learning_rate": 4.79285134037368e-07, + "loss": 1.5336, + "step": 2518 + }, + { + "epoch": 2.7605479452054795, + "grad_norm": 2.851407766342163, + "learning_rate": 4.772542648253453e-07, + "loss": 1.5445, + "step": 2519 + }, + { + "epoch": 2.7616438356164386, + "grad_norm": 2.9547126293182373, + "learning_rate": 4.752233956133225e-07, + "loss": 1.6028, + "step": 2520 + }, + { + "epoch": 2.7616438356164386, + "eval_loss": 1.8189756870269775, + "eval_runtime": 50.0014, + "eval_samples_per_second": 19.999, + "eval_steps_per_second": 0.84, + "step": 2520 + }, + { + "epoch": 2.762739726027397, + "grad_norm": 2.963991641998291, + "learning_rate": 4.731925264012998e-07, + "loss": 1.565, + "step": 2521 + }, + { + "epoch": 2.763835616438356, + "grad_norm": 2.953455686569214, + "learning_rate": 4.7116165718927705e-07, + "loss": 1.5705, + "step": 2522 + }, + { + "epoch": 2.764931506849315, + "grad_norm": 2.846900701522827, + "learning_rate": 4.6913078797725427e-07, + "loss": 1.5404, + "step": 2523 + }, + { + "epoch": 2.766027397260274, + "grad_norm": 3.0525879859924316, + "learning_rate": 4.6709991876523155e-07, + "loss": 1.622, + "step": 2524 + }, + { + "epoch": 2.767123287671233, + "grad_norm": 2.948841094970703, + "learning_rate": 4.650690495532088e-07, + "loss": 1.5752, + "step": 2525 + }, + { + "epoch": 2.768219178082192, + "grad_norm": 2.8494491577148438, + "learning_rate": 4.6303818034118605e-07, + "loss": 1.503, + "step": 2526 + }, + { + "epoch": 2.7693150684931505, + "grad_norm": 2.8788719177246094, + "learning_rate": 4.610073111291633e-07, + "loss": 1.5483, + "step": 2527 + }, + { + "epoch": 2.7704109589041095, + "grad_norm": 2.86061954498291, + "learning_rate": 4.589764419171406e-07, + "loss": 1.5142, + "step": 2528 + }, + { + "epoch": 2.7715068493150685, + "grad_norm": 2.995512008666992, + "learning_rate": 4.569455727051178e-07, + "loss": 1.598, + "step": 2529 + }, + { + "epoch": 2.7726027397260276, + "grad_norm": 2.810875177383423, + "learning_rate": 4.549147034930951e-07, + "loss": 1.5676, + "step": 2530 + }, + { + "epoch": 2.7726027397260276, + "eval_loss": 1.8178433179855347, + "eval_runtime": 50.3071, + "eval_samples_per_second": 19.878, + "eval_steps_per_second": 0.835, + "step": 2530 + }, + { + "epoch": 2.773698630136986, + "grad_norm": 2.827171564102173, + "learning_rate": 4.5288383428107237e-07, + "loss": 1.5384, + "step": 2531 + }, + { + "epoch": 2.774794520547945, + "grad_norm": 2.802733898162842, + "learning_rate": 4.5085296506904954e-07, + "loss": 1.5294, + "step": 2532 + }, + { + "epoch": 2.775890410958904, + "grad_norm": 2.9027161598205566, + "learning_rate": 4.488220958570268e-07, + "loss": 1.5219, + "step": 2533 + }, + { + "epoch": 2.7769863013698632, + "grad_norm": 2.9201040267944336, + "learning_rate": 4.467912266450041e-07, + "loss": 1.5847, + "step": 2534 + }, + { + "epoch": 2.778082191780822, + "grad_norm": 2.793375253677368, + "learning_rate": 4.447603574329813e-07, + "loss": 1.5232, + "step": 2535 + }, + { + "epoch": 2.779178082191781, + "grad_norm": 2.9842793941497803, + "learning_rate": 4.427294882209586e-07, + "loss": 1.5679, + "step": 2536 + }, + { + "epoch": 2.78027397260274, + "grad_norm": 2.942744255065918, + "learning_rate": 4.4069861900893587e-07, + "loss": 1.6284, + "step": 2537 + }, + { + "epoch": 2.7813698630136985, + "grad_norm": 2.8464622497558594, + "learning_rate": 4.386677497969131e-07, + "loss": 1.5694, + "step": 2538 + }, + { + "epoch": 2.7824657534246575, + "grad_norm": 2.9041919708251953, + "learning_rate": 4.3663688058489037e-07, + "loss": 1.5758, + "step": 2539 + }, + { + "epoch": 2.7835616438356166, + "grad_norm": 2.962247133255005, + "learning_rate": 4.3460601137286764e-07, + "loss": 1.5625, + "step": 2540 + }, + { + "epoch": 2.7835616438356166, + "eval_loss": 1.8173555135726929, + "eval_runtime": 49.7157, + "eval_samples_per_second": 20.114, + "eval_steps_per_second": 0.845, + "step": 2540 + }, + { + "epoch": 2.784657534246575, + "grad_norm": 2.886990547180176, + "learning_rate": 4.325751421608449e-07, + "loss": 1.5927, + "step": 2541 + }, + { + "epoch": 2.785753424657534, + "grad_norm": 2.888997793197632, + "learning_rate": 4.3054427294882214e-07, + "loss": 1.5434, + "step": 2542 + }, + { + "epoch": 2.786849315068493, + "grad_norm": 2.7936785221099854, + "learning_rate": 4.285134037367994e-07, + "loss": 1.4914, + "step": 2543 + }, + { + "epoch": 2.787945205479452, + "grad_norm": 2.8689370155334473, + "learning_rate": 4.264825345247767e-07, + "loss": 1.5206, + "step": 2544 + }, + { + "epoch": 2.789041095890411, + "grad_norm": 2.666140079498291, + "learning_rate": 4.2445166531275386e-07, + "loss": 1.5108, + "step": 2545 + }, + { + "epoch": 2.79013698630137, + "grad_norm": 2.999361515045166, + "learning_rate": 4.2242079610073114e-07, + "loss": 1.6066, + "step": 2546 + }, + { + "epoch": 2.791232876712329, + "grad_norm": 2.919015407562256, + "learning_rate": 4.203899268887084e-07, + "loss": 1.6208, + "step": 2547 + }, + { + "epoch": 2.792328767123288, + "grad_norm": 2.774463415145874, + "learning_rate": 4.1835905767668564e-07, + "loss": 1.5628, + "step": 2548 + }, + { + "epoch": 2.7934246575342465, + "grad_norm": 2.7784156799316406, + "learning_rate": 4.163281884646629e-07, + "loss": 1.5711, + "step": 2549 + }, + { + "epoch": 2.7945205479452055, + "grad_norm": 2.8377902507781982, + "learning_rate": 4.142973192526402e-07, + "loss": 1.5766, + "step": 2550 + }, + { + "epoch": 2.7945205479452055, + "eval_loss": 1.8163820505142212, + "eval_runtime": 51.6657, + "eval_samples_per_second": 19.355, + "eval_steps_per_second": 0.813, + "step": 2550 + }, + { + "epoch": 2.7956164383561646, + "grad_norm": 3.265052318572998, + "learning_rate": 4.122664500406174e-07, + "loss": 1.5373, + "step": 2551 + }, + { + "epoch": 2.796712328767123, + "grad_norm": 2.8727524280548096, + "learning_rate": 4.102355808285947e-07, + "loss": 1.569, + "step": 2552 + }, + { + "epoch": 2.797808219178082, + "grad_norm": 2.904752016067505, + "learning_rate": 4.0820471161657196e-07, + "loss": 1.5814, + "step": 2553 + }, + { + "epoch": 2.7989041095890412, + "grad_norm": 2.7889554500579834, + "learning_rate": 4.0617384240454913e-07, + "loss": 1.4965, + "step": 2554 + }, + { + "epoch": 2.8, + "grad_norm": 2.8267621994018555, + "learning_rate": 4.041429731925264e-07, + "loss": 1.5322, + "step": 2555 + }, + { + "epoch": 2.801095890410959, + "grad_norm": 2.907101631164551, + "learning_rate": 4.021121039805037e-07, + "loss": 1.5455, + "step": 2556 + }, + { + "epoch": 2.802191780821918, + "grad_norm": 2.9485068321228027, + "learning_rate": 4.000812347684809e-07, + "loss": 1.5586, + "step": 2557 + }, + { + "epoch": 2.8032876712328765, + "grad_norm": 2.8086092472076416, + "learning_rate": 3.980503655564582e-07, + "loss": 1.5677, + "step": 2558 + }, + { + "epoch": 2.8043835616438355, + "grad_norm": 2.834923505783081, + "learning_rate": 3.9601949634443546e-07, + "loss": 1.5367, + "step": 2559 + }, + { + "epoch": 2.8054794520547945, + "grad_norm": 2.8668291568756104, + "learning_rate": 3.939886271324127e-07, + "loss": 1.5903, + "step": 2560 + }, + { + "epoch": 2.8054794520547945, + "eval_loss": 1.8173569440841675, + "eval_runtime": 48.5732, + "eval_samples_per_second": 20.588, + "eval_steps_per_second": 0.865, + "step": 2560 + }, + { + "epoch": 2.8065753424657536, + "grad_norm": 3.003795862197876, + "learning_rate": 3.9195775792038996e-07, + "loss": 1.5653, + "step": 2561 + }, + { + "epoch": 2.807671232876712, + "grad_norm": 2.8089170455932617, + "learning_rate": 3.8992688870836723e-07, + "loss": 1.5565, + "step": 2562 + }, + { + "epoch": 2.808767123287671, + "grad_norm": 2.8321988582611084, + "learning_rate": 3.8789601949634446e-07, + "loss": 1.4748, + "step": 2563 + }, + { + "epoch": 2.8098630136986302, + "grad_norm": 2.8655641078948975, + "learning_rate": 3.8586515028432173e-07, + "loss": 1.5502, + "step": 2564 + }, + { + "epoch": 2.8109589041095893, + "grad_norm": 2.7576475143432617, + "learning_rate": 3.83834281072299e-07, + "loss": 1.5138, + "step": 2565 + }, + { + "epoch": 2.812054794520548, + "grad_norm": 2.89864182472229, + "learning_rate": 3.818034118602762e-07, + "loss": 1.5553, + "step": 2566 + }, + { + "epoch": 2.813150684931507, + "grad_norm": 3.032454490661621, + "learning_rate": 3.7977254264825345e-07, + "loss": 1.5683, + "step": 2567 + }, + { + "epoch": 2.814246575342466, + "grad_norm": 2.8610739707946777, + "learning_rate": 3.7774167343623073e-07, + "loss": 1.5897, + "step": 2568 + }, + { + "epoch": 2.8153424657534245, + "grad_norm": 2.927236795425415, + "learning_rate": 3.75710804224208e-07, + "loss": 1.5342, + "step": 2569 + }, + { + "epoch": 2.8164383561643835, + "grad_norm": 2.75993013381958, + "learning_rate": 3.7367993501218523e-07, + "loss": 1.5331, + "step": 2570 + }, + { + "epoch": 2.8164383561643835, + "eval_loss": 1.8160063028335571, + "eval_runtime": 50.568, + "eval_samples_per_second": 19.775, + "eval_steps_per_second": 0.831, + "step": 2570 + }, + { + "epoch": 2.8175342465753426, + "grad_norm": 2.781172275543213, + "learning_rate": 3.716490658001625e-07, + "loss": 1.5071, + "step": 2571 + }, + { + "epoch": 2.818630136986301, + "grad_norm": 2.876645565032959, + "learning_rate": 3.696181965881398e-07, + "loss": 1.5586, + "step": 2572 + }, + { + "epoch": 2.81972602739726, + "grad_norm": 2.871903896331787, + "learning_rate": 3.67587327376117e-07, + "loss": 1.5492, + "step": 2573 + }, + { + "epoch": 2.8208219178082192, + "grad_norm": 2.888033390045166, + "learning_rate": 3.655564581640943e-07, + "loss": 1.5123, + "step": 2574 + }, + { + "epoch": 2.821917808219178, + "grad_norm": 2.7957611083984375, + "learning_rate": 3.6352558895207155e-07, + "loss": 1.5675, + "step": 2575 + }, + { + "epoch": 2.823013698630137, + "grad_norm": 2.8597512245178223, + "learning_rate": 3.614947197400488e-07, + "loss": 1.5578, + "step": 2576 + }, + { + "epoch": 2.824109589041096, + "grad_norm": 2.9219770431518555, + "learning_rate": 3.5946385052802605e-07, + "loss": 1.5407, + "step": 2577 + }, + { + "epoch": 2.825205479452055, + "grad_norm": 2.880618095397949, + "learning_rate": 3.5743298131600333e-07, + "loss": 1.6036, + "step": 2578 + }, + { + "epoch": 2.826301369863014, + "grad_norm": 2.857084274291992, + "learning_rate": 3.554021121039805e-07, + "loss": 1.5496, + "step": 2579 + }, + { + "epoch": 2.8273972602739725, + "grad_norm": 2.7948296070098877, + "learning_rate": 3.533712428919578e-07, + "loss": 1.5631, + "step": 2580 + }, + { + "epoch": 2.8273972602739725, + "eval_loss": 1.8157116174697876, + "eval_runtime": 51.0427, + "eval_samples_per_second": 19.591, + "eval_steps_per_second": 0.823, + "step": 2580 + }, + { + "epoch": 2.8284931506849316, + "grad_norm": 2.9175124168395996, + "learning_rate": 3.5134037367993505e-07, + "loss": 1.5187, + "step": 2581 + }, + { + "epoch": 2.8295890410958906, + "grad_norm": 2.923126220703125, + "learning_rate": 3.4930950446791227e-07, + "loss": 1.5466, + "step": 2582 + }, + { + "epoch": 2.830684931506849, + "grad_norm": 2.853585958480835, + "learning_rate": 3.4727863525588955e-07, + "loss": 1.5941, + "step": 2583 + }, + { + "epoch": 2.831780821917808, + "grad_norm": 2.9729127883911133, + "learning_rate": 3.452477660438668e-07, + "loss": 1.5626, + "step": 2584 + }, + { + "epoch": 2.8328767123287673, + "grad_norm": 2.900005578994751, + "learning_rate": 3.4321689683184405e-07, + "loss": 1.5117, + "step": 2585 + }, + { + "epoch": 2.833972602739726, + "grad_norm": 2.8693535327911377, + "learning_rate": 3.411860276198213e-07, + "loss": 1.5607, + "step": 2586 + }, + { + "epoch": 2.835068493150685, + "grad_norm": 2.9128522872924805, + "learning_rate": 3.391551584077986e-07, + "loss": 1.5625, + "step": 2587 + }, + { + "epoch": 2.836164383561644, + "grad_norm": 2.783365249633789, + "learning_rate": 3.3712428919577577e-07, + "loss": 1.5202, + "step": 2588 + }, + { + "epoch": 2.8372602739726025, + "grad_norm": 2.846964120864868, + "learning_rate": 3.3509341998375304e-07, + "loss": 1.5489, + "step": 2589 + }, + { + "epoch": 2.8383561643835615, + "grad_norm": 2.9657390117645264, + "learning_rate": 3.330625507717303e-07, + "loss": 1.6139, + "step": 2590 + }, + { + "epoch": 2.8383561643835615, + "eval_loss": 1.8159494400024414, + "eval_runtime": 47.7292, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 2590 + }, + { + "epoch": 2.8394520547945206, + "grad_norm": 3.2273998260498047, + "learning_rate": 3.3103168155970754e-07, + "loss": 1.6173, + "step": 2591 + }, + { + "epoch": 2.8405479452054796, + "grad_norm": 2.946399211883545, + "learning_rate": 3.290008123476848e-07, + "loss": 1.5546, + "step": 2592 + }, + { + "epoch": 2.8416438356164386, + "grad_norm": 2.792555809020996, + "learning_rate": 3.269699431356621e-07, + "loss": 1.4948, + "step": 2593 + }, + { + "epoch": 2.842739726027397, + "grad_norm": 2.91678524017334, + "learning_rate": 3.249390739236393e-07, + "loss": 1.6064, + "step": 2594 + }, + { + "epoch": 2.8438356164383563, + "grad_norm": 2.9972245693206787, + "learning_rate": 3.229082047116166e-07, + "loss": 1.5914, + "step": 2595 + }, + { + "epoch": 2.8449315068493153, + "grad_norm": 2.853240728378296, + "learning_rate": 3.2087733549959387e-07, + "loss": 1.575, + "step": 2596 + }, + { + "epoch": 2.846027397260274, + "grad_norm": 2.9793026447296143, + "learning_rate": 3.1884646628757114e-07, + "loss": 1.5828, + "step": 2597 + }, + { + "epoch": 2.847123287671233, + "grad_norm": 2.803582191467285, + "learning_rate": 3.1681559707554837e-07, + "loss": 1.5277, + "step": 2598 + }, + { + "epoch": 2.848219178082192, + "grad_norm": 2.8925442695617676, + "learning_rate": 3.1478472786352564e-07, + "loss": 1.5616, + "step": 2599 + }, + { + "epoch": 2.8493150684931505, + "grad_norm": 2.86979603767395, + "learning_rate": 3.127538586515029e-07, + "loss": 1.5452, + "step": 2600 + }, + { + "epoch": 2.8493150684931505, + "eval_loss": 1.815985918045044, + "eval_runtime": 50.7957, + "eval_samples_per_second": 19.687, + "eval_steps_per_second": 0.827, + "step": 2600 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 8.105004464340992e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2700/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2700/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..cf2513c6be393ffd7f18b9966efe1ea94961ce76 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2700/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1e03f5804d1958af28b457d1c95eadeccb6d55a564daf236124a0bd65854c268 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/generation_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/latest b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/latest new file mode 100644 index 0000000000000000000000000000000000000000..bb5cfcd38ec9615f610a0b181977e932b7a8cdd6 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/latest @@ -0,0 +1 @@ +global_step2736 \ No newline at end of file diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/rng_state_0.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..ecf4e7e459680f3d62f3b8bb2b4f24f7dda0e86e --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:189c243889f5a95c2cd9f32498254c93e6dfd12bda2206ef6c80d95c0e8a4540 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/rng_state_1.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..e01453da9fa144b92db94705950b723477238929 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c91dd138e49d38ee81ba13a91f33285d682f7412fc98d49f06ab9966e027cde0 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..4f8e48186d2a88c25078e975e6d5b5212398d032 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:31508ab4a29ee0b942e59def28e1a7381e3d4ecb77be81d2393ceb3508dda8aa +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/scheduler.pt b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..07b01fa3461395c0264e60be969a821f164b348d --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5614e392df14a20160941ec26942ad76d033e087bdfc5fea74771afaf2949a31 +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/special_tokens_map.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..64864ca9e155c855c228c33597e11d92c7d85d6c --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/trainer_state.json @@ -0,0 +1,21369 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.9983561643835617, + "eval_steps": 10, + "global_step": 2736, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.5818793773651123, + "learning_rate": 4.569455727051178e-06, + "loss": 1.85, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 3.0734691619873047, + "learning_rate": 4.567424857839156e-06, + "loss": 1.8682, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.496253252029419, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9007, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.360891342163086, + "learning_rate": 4.5633631194151105e-06, + "loss": 1.8262, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.5731117725372314, + "learning_rate": 4.561332250203087e-06, + "loss": 1.8369, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.700558662414551, + "learning_rate": 4.559301380991065e-06, + "loss": 1.7825, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.561306953430176, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.8174, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.618889570236206, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.8812, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.4671308994293213, + "learning_rate": 4.553208773354996e-06, + "loss": 1.7959, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.4571878910064697, + "learning_rate": 4.551177904142974e-06, + "loss": 1.8943, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.010777711868286, + "eval_runtime": 43.7424, + "eval_samples_per_second": 22.861, + "eval_steps_per_second": 0.96, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.4786667823791504, + "learning_rate": 4.5491470349309505e-06, + "loss": 1.8147, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.2955000400543213, + "learning_rate": 4.547116165718929e-06, + "loss": 1.8547, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.3873913288116455, + "learning_rate": 4.545085296506905e-06, + "loss": 1.8156, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.2449734210968018, + "learning_rate": 4.543054427294882e-06, + "loss": 1.8053, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.3969857692718506, + "learning_rate": 4.54102355808286e-06, + "loss": 1.8085, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.3446357250213623, + "learning_rate": 4.538992688870837e-06, + "loss": 1.8355, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.408762216567993, + "learning_rate": 4.536961819658814e-06, + "loss": 1.8319, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.362962245941162, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.8422, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.402984857559204, + "learning_rate": 4.532900081234769e-06, + "loss": 1.8125, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.3964452743530273, + "learning_rate": 4.530869212022746e-06, + "loss": 1.8417, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.006908416748047, + "eval_runtime": 44.2701, + "eval_samples_per_second": 22.589, + "eval_steps_per_second": 0.949, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.595233201980591, + "learning_rate": 4.528838342810723e-06, + "loss": 1.8275, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.326117753982544, + "learning_rate": 4.526807473598701e-06, + "loss": 1.8405, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.4904465675354004, + "learning_rate": 4.524776604386678e-06, + "loss": 1.8172, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.4591610431671143, + "learning_rate": 4.522745735174655e-06, + "loss": 1.876, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.319800853729248, + "learning_rate": 4.520714865962632e-06, + "loss": 1.8294, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.28790020942688, + "learning_rate": 4.51868399675061e-06, + "loss": 1.8537, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.284822702407837, + "learning_rate": 4.516653127538587e-06, + "loss": 1.8087, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.273820638656616, + "learning_rate": 4.514622258326564e-06, + "loss": 1.7992, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.2025654315948486, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.8248, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 2.3218843936920166, + "learning_rate": 4.510560519902519e-06, + "loss": 1.8647, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.0048956871032715, + "eval_runtime": 44.5972, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.515766143798828, + "learning_rate": 4.508529650690496e-06, + "loss": 1.7673, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.4376487731933594, + "learning_rate": 4.506498781478473e-06, + "loss": 1.7755, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.2801272869110107, + "learning_rate": 4.504467912266451e-06, + "loss": 1.8273, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.4707891941070557, + "learning_rate": 4.502437043054428e-06, + "loss": 1.8971, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.301301956176758, + "learning_rate": 4.500406173842405e-06, + "loss": 1.8345, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.2317373752593994, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.8248, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.447373867034912, + "learning_rate": 4.49634443541836e-06, + "loss": 1.8148, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.337245225906372, + "learning_rate": 4.494313566206336e-06, + "loss": 1.8811, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3717031478881836, + "learning_rate": 4.492282696994314e-06, + "loss": 1.8311, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.287677526473999, + "learning_rate": 4.490251827782291e-06, + "loss": 1.7728, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.004889965057373, + "eval_runtime": 43.8132, + "eval_samples_per_second": 22.824, + "eval_steps_per_second": 0.959, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.441445827484131, + "learning_rate": 4.488220958570269e-06, + "loss": 1.7927, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.3191535472869873, + "learning_rate": 4.486190089358245e-06, + "loss": 1.759, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.3118820190429688, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8509, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.4711859226226807, + "learning_rate": 4.4821283509342e-06, + "loss": 1.8402, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.3142173290252686, + "learning_rate": 4.480097481722178e-06, + "loss": 1.8238, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.2665774822235107, + "learning_rate": 4.478066612510154e-06, + "loss": 1.7826, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.1692490577697754, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.8138, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.412783145904541, + "learning_rate": 4.474004874086109e-06, + "loss": 1.9421, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3692686557769775, + "learning_rate": 4.471974004874087e-06, + "loss": 1.7722, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.3224151134490967, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.8753, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.0016305446624756, + "eval_runtime": 45.564, + "eval_samples_per_second": 21.947, + "eval_steps_per_second": 0.922, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.51043438911438, + "learning_rate": 4.467912266450041e-06, + "loss": 1.8325, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.4568657875061035, + "learning_rate": 4.465881397238018e-06, + "loss": 1.8081, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.3279001712799072, + "learning_rate": 4.463850528025995e-06, + "loss": 1.8439, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.6041171550750732, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.8623, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.3780531883239746, + "learning_rate": 4.45978878960195e-06, + "loss": 1.8269, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.511500597000122, + "learning_rate": 4.457757920389927e-06, + "loss": 1.7749, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.707789659500122, + "learning_rate": 4.455727051177904e-06, + "loss": 1.8157, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.5495495796203613, + "learning_rate": 4.453696181965882e-06, + "loss": 1.8534, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.255876302719116, + "learning_rate": 4.451665312753859e-06, + "loss": 1.8377, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.415097951889038, + "learning_rate": 4.449634443541836e-06, + "loss": 1.8764, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 1.9943474531173706, + "eval_runtime": 42.2985, + "eval_samples_per_second": 23.642, + "eval_steps_per_second": 0.993, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.2604713439941406, + "learning_rate": 4.4476035743298135e-06, + "loss": 1.7898, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.5643842220306396, + "learning_rate": 4.445572705117791e-06, + "loss": 1.7953, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.412548780441284, + "learning_rate": 4.443541835905768e-06, + "loss": 1.8611, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.5423877239227295, + "learning_rate": 4.441510966693745e-06, + "loss": 1.8049, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.4986631870269775, + "learning_rate": 4.439480097481723e-06, + "loss": 1.8759, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.8349273204803467, + "learning_rate": 4.4374492282697e-06, + "loss": 1.7416, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.452017068862915, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8268, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.6425273418426514, + "learning_rate": 4.433387489845654e-06, + "loss": 1.8334, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.328740358352661, + "learning_rate": 4.431356620633632e-06, + "loss": 1.7931, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.262209415435791, + "learning_rate": 4.429325751421609e-06, + "loss": 1.8151, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 1.9917123317718506, + "eval_runtime": 42.4309, + "eval_samples_per_second": 23.568, + "eval_steps_per_second": 0.99, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.2984049320220947, + "learning_rate": 4.427294882209586e-06, + "loss": 1.8187, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.3910019397735596, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.7512, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.4494471549987793, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8202, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.3857171535491943, + "learning_rate": 4.421202274573518e-06, + "loss": 1.8279, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.4526500701904297, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.7136, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.2216501235961914, + "learning_rate": 4.417140536149473e-06, + "loss": 1.7574, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.7639758586883545, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8403, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.3177521228790283, + "learning_rate": 4.413078797725427e-06, + "loss": 1.7898, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.527902364730835, + "learning_rate": 4.411047928513404e-06, + "loss": 1.8184, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.34061336517334, + "learning_rate": 4.409017059301382e-06, + "loss": 1.811, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 1.9882216453552246, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.6462934017181396, + "learning_rate": 4.406986190089358e-06, + "loss": 1.8501, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.2725913524627686, + "learning_rate": 4.404955320877336e-06, + "loss": 1.8262, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.390084981918335, + "learning_rate": 4.402924451665313e-06, + "loss": 1.7842, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.314305305480957, + "learning_rate": 4.400893582453291e-06, + "loss": 1.7692, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.2828257083892822, + "learning_rate": 4.398862713241267e-06, + "loss": 1.8031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.3493599891662598, + "learning_rate": 4.396831844029245e-06, + "loss": 1.8183, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.423365831375122, + "learning_rate": 4.394800974817222e-06, + "loss": 1.8012, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.438015937805176, + "learning_rate": 4.392770105605199e-06, + "loss": 1.8217, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.357172727584839, + "learning_rate": 4.390739236393176e-06, + "loss": 1.8382, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.3383474349975586, + "learning_rate": 4.388708367181154e-06, + "loss": 1.841, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 1.9878787994384766, + "eval_runtime": 44.0694, + "eval_samples_per_second": 22.691, + "eval_steps_per_second": 0.953, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.459385633468628, + "learning_rate": 4.386677497969131e-06, + "loss": 1.8377, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.356788158416748, + "learning_rate": 4.384646628757108e-06, + "loss": 1.8718, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.3217222690582275, + "learning_rate": 4.3826157595450855e-06, + "loss": 1.8073, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.3134331703186035, + "learning_rate": 4.380584890333063e-06, + "loss": 1.7675, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.3273653984069824, + "learning_rate": 4.37855402112104e-06, + "loss": 1.7169, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.611034393310547, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8702, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.2077300548553467, + "learning_rate": 4.3744922826969946e-06, + "loss": 1.7951, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.3393311500549316, + "learning_rate": 4.372461413484972e-06, + "loss": 1.8136, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.3306660652160645, + "learning_rate": 4.370430544272949e-06, + "loss": 1.8372, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.4440250396728516, + "learning_rate": 4.368399675060926e-06, + "loss": 1.844, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 1.98439621925354, + "eval_runtime": 44.4595, + "eval_samples_per_second": 22.492, + "eval_steps_per_second": 0.945, + "step": 600 + }, + { + "epoch": 0.6586301369863014, + "grad_norm": 2.205542802810669, + "learning_rate": 4.366368805848904e-06, + "loss": 1.8182, + "step": 601 + }, + { + "epoch": 0.6597260273972603, + "grad_norm": 2.6030094623565674, + "learning_rate": 4.364337936636881e-06, + "loss": 1.7524, + "step": 602 + }, + { + "epoch": 0.6608219178082192, + "grad_norm": 2.4000444412231445, + "learning_rate": 4.362307067424858e-06, + "loss": 1.8269, + "step": 603 + }, + { + "epoch": 0.6619178082191781, + "grad_norm": 2.319526433944702, + "learning_rate": 4.3602761982128355e-06, + "loss": 1.7722, + "step": 604 + }, + { + "epoch": 0.663013698630137, + "grad_norm": 2.1624808311462402, + "learning_rate": 4.358245329000813e-06, + "loss": 1.8081, + "step": 605 + }, + { + "epoch": 0.6641095890410958, + "grad_norm": 2.26615309715271, + "learning_rate": 4.35621445978879e-06, + "loss": 1.8049, + "step": 606 + }, + { + "epoch": 0.6652054794520548, + "grad_norm": 2.4895036220550537, + "learning_rate": 4.354183590576767e-06, + "loss": 1.7932, + "step": 607 + }, + { + "epoch": 0.6663013698630137, + "grad_norm": 2.437161445617676, + "learning_rate": 4.352152721364745e-06, + "loss": 1.772, + "step": 608 + }, + { + "epoch": 0.6673972602739726, + "grad_norm": 2.344069480895996, + "learning_rate": 4.350121852152722e-06, + "loss": 1.8213, + "step": 609 + }, + { + "epoch": 0.6684931506849315, + "grad_norm": 2.360734224319458, + "learning_rate": 4.348090982940699e-06, + "loss": 1.8268, + "step": 610 + }, + { + "epoch": 0.6684931506849315, + "eval_loss": 1.9846479892730713, + "eval_runtime": 44.947, + "eval_samples_per_second": 22.248, + "eval_steps_per_second": 0.934, + "step": 610 + }, + { + "epoch": 0.6695890410958905, + "grad_norm": 2.59487247467041, + "learning_rate": 4.3460601137286764e-06, + "loss": 1.8303, + "step": 611 + }, + { + "epoch": 0.6706849315068493, + "grad_norm": 2.6512107849121094, + "learning_rate": 4.344029244516654e-06, + "loss": 1.7979, + "step": 612 + }, + { + "epoch": 0.6717808219178082, + "grad_norm": 2.2900476455688477, + "learning_rate": 4.341998375304631e-06, + "loss": 1.8707, + "step": 613 + }, + { + "epoch": 0.6728767123287671, + "grad_norm": 2.409494638442993, + "learning_rate": 4.339967506092607e-06, + "loss": 1.8075, + "step": 614 + }, + { + "epoch": 0.673972602739726, + "grad_norm": 2.527205467224121, + "learning_rate": 4.3379366368805855e-06, + "loss": 1.7884, + "step": 615 + }, + { + "epoch": 0.675068493150685, + "grad_norm": 2.33461332321167, + "learning_rate": 4.335905767668562e-06, + "loss": 1.7839, + "step": 616 + }, + { + "epoch": 0.6761643835616439, + "grad_norm": 2.3555047512054443, + "learning_rate": 4.33387489845654e-06, + "loss": 1.8245, + "step": 617 + }, + { + "epoch": 0.6772602739726027, + "grad_norm": 2.397148609161377, + "learning_rate": 4.3318440292445165e-06, + "loss": 1.8115, + "step": 618 + }, + { + "epoch": 0.6783561643835616, + "grad_norm": 2.232139825820923, + "learning_rate": 4.329813160032495e-06, + "loss": 1.7747, + "step": 619 + }, + { + "epoch": 0.6794520547945205, + "grad_norm": 2.1684389114379883, + "learning_rate": 4.327782290820471e-06, + "loss": 1.748, + "step": 620 + }, + { + "epoch": 0.6794520547945205, + "eval_loss": 1.980297565460205, + "eval_runtime": 42.9183, + "eval_samples_per_second": 23.3, + "eval_steps_per_second": 0.979, + "step": 620 + }, + { + "epoch": 0.6805479452054795, + "grad_norm": 2.2454638481140137, + "learning_rate": 4.325751421608449e-06, + "loss": 1.83, + "step": 621 + }, + { + "epoch": 0.6816438356164384, + "grad_norm": 2.3264827728271484, + "learning_rate": 4.323720552396426e-06, + "loss": 1.7856, + "step": 622 + }, + { + "epoch": 0.6827397260273973, + "grad_norm": 2.298804998397827, + "learning_rate": 4.321689683184404e-06, + "loss": 1.78, + "step": 623 + }, + { + "epoch": 0.6838356164383562, + "grad_norm": 2.2202727794647217, + "learning_rate": 4.31965881397238e-06, + "loss": 1.7875, + "step": 624 + }, + { + "epoch": 0.684931506849315, + "grad_norm": 2.494479179382324, + "learning_rate": 4.3176279447603574e-06, + "loss": 1.8363, + "step": 625 + }, + { + "epoch": 0.686027397260274, + "grad_norm": 2.3520524501800537, + "learning_rate": 4.315597075548335e-06, + "loss": 1.8266, + "step": 626 + }, + { + "epoch": 0.6871232876712329, + "grad_norm": 2.373878002166748, + "learning_rate": 4.313566206336312e-06, + "loss": 1.8348, + "step": 627 + }, + { + "epoch": 0.6882191780821918, + "grad_norm": 2.4232168197631836, + "learning_rate": 4.311535337124289e-06, + "loss": 1.77, + "step": 628 + }, + { + "epoch": 0.6893150684931507, + "grad_norm": 2.2572121620178223, + "learning_rate": 4.3095044679122665e-06, + "loss": 1.7914, + "step": 629 + }, + { + "epoch": 0.6904109589041096, + "grad_norm": 2.5817952156066895, + "learning_rate": 4.307473598700244e-06, + "loss": 1.8099, + "step": 630 + }, + { + "epoch": 0.6904109589041096, + "eval_loss": 1.9784427881240845, + "eval_runtime": 43.2981, + "eval_samples_per_second": 23.096, + "eval_steps_per_second": 0.97, + "step": 630 + }, + { + "epoch": 0.6915068493150685, + "grad_norm": 2.646040916442871, + "learning_rate": 4.305442729488221e-06, + "loss": 1.7452, + "step": 631 + }, + { + "epoch": 0.6926027397260274, + "grad_norm": 2.370220184326172, + "learning_rate": 4.303411860276198e-06, + "loss": 1.8399, + "step": 632 + }, + { + "epoch": 0.6936986301369863, + "grad_norm": 2.4599130153656006, + "learning_rate": 4.301380991064176e-06, + "loss": 1.8273, + "step": 633 + }, + { + "epoch": 0.6947945205479452, + "grad_norm": 2.4135093688964844, + "learning_rate": 4.299350121852153e-06, + "loss": 1.852, + "step": 634 + }, + { + "epoch": 0.6958904109589041, + "grad_norm": 2.371021270751953, + "learning_rate": 4.29731925264013e-06, + "loss": 1.8339, + "step": 635 + }, + { + "epoch": 0.6969863013698631, + "grad_norm": 2.6858229637145996, + "learning_rate": 4.2952883834281075e-06, + "loss": 1.792, + "step": 636 + }, + { + "epoch": 0.6980821917808219, + "grad_norm": 2.4101266860961914, + "learning_rate": 4.293257514216085e-06, + "loss": 1.7556, + "step": 637 + }, + { + "epoch": 0.6991780821917808, + "grad_norm": 2.769761800765991, + "learning_rate": 4.291226645004062e-06, + "loss": 1.8223, + "step": 638 + }, + { + "epoch": 0.7002739726027397, + "grad_norm": 2.5777804851531982, + "learning_rate": 4.289195775792039e-06, + "loss": 1.8406, + "step": 639 + }, + { + "epoch": 0.7013698630136986, + "grad_norm": 2.376864433288574, + "learning_rate": 4.287164906580017e-06, + "loss": 1.7846, + "step": 640 + }, + { + "epoch": 0.7013698630136986, + "eval_loss": 1.9758412837982178, + "eval_runtime": 43.9353, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 640 + }, + { + "epoch": 0.7024657534246576, + "grad_norm": 2.7461960315704346, + "learning_rate": 4.285134037367994e-06, + "loss": 1.776, + "step": 641 + }, + { + "epoch": 0.7035616438356165, + "grad_norm": 2.549576759338379, + "learning_rate": 4.283103168155971e-06, + "loss": 1.7644, + "step": 642 + }, + { + "epoch": 0.7046575342465753, + "grad_norm": 2.568074941635132, + "learning_rate": 4.281072298943948e-06, + "loss": 1.7571, + "step": 643 + }, + { + "epoch": 0.7057534246575342, + "grad_norm": 2.8195722103118896, + "learning_rate": 4.279041429731926e-06, + "loss": 1.7959, + "step": 644 + }, + { + "epoch": 0.7068493150684931, + "grad_norm": 2.471623182296753, + "learning_rate": 4.277010560519903e-06, + "loss": 1.7584, + "step": 645 + }, + { + "epoch": 0.707945205479452, + "grad_norm": 2.664689302444458, + "learning_rate": 4.27497969130788e-06, + "loss": 1.7909, + "step": 646 + }, + { + "epoch": 0.709041095890411, + "grad_norm": 2.684504985809326, + "learning_rate": 4.2729488220958575e-06, + "loss": 1.8205, + "step": 647 + }, + { + "epoch": 0.7101369863013699, + "grad_norm": 2.5247528553009033, + "learning_rate": 4.270917952883835e-06, + "loss": 1.8365, + "step": 648 + }, + { + "epoch": 0.7112328767123288, + "grad_norm": 2.757972478866577, + "learning_rate": 4.268887083671812e-06, + "loss": 1.7665, + "step": 649 + }, + { + "epoch": 0.7123287671232876, + "grad_norm": 2.6032614707946777, + "learning_rate": 4.266856214459789e-06, + "loss": 1.8702, + "step": 650 + }, + { + "epoch": 0.7123287671232876, + "eval_loss": 1.9716291427612305, + "eval_runtime": 43.5482, + "eval_samples_per_second": 22.963, + "eval_steps_per_second": 0.964, + "step": 650 + }, + { + "epoch": 0.7134246575342466, + "grad_norm": 2.5324370861053467, + "learning_rate": 4.264825345247766e-06, + "loss": 1.738, + "step": 651 + }, + { + "epoch": 0.7145205479452055, + "grad_norm": 2.730712890625, + "learning_rate": 4.262794476035744e-06, + "loss": 1.7918, + "step": 652 + }, + { + "epoch": 0.7156164383561644, + "grad_norm": 2.5554966926574707, + "learning_rate": 4.26076360682372e-06, + "loss": 1.7218, + "step": 653 + }, + { + "epoch": 0.7167123287671233, + "grad_norm": 2.4774436950683594, + "learning_rate": 4.2587327376116984e-06, + "loss": 1.7823, + "step": 654 + }, + { + "epoch": 0.7178082191780822, + "grad_norm": 2.77431058883667, + "learning_rate": 4.256701868399675e-06, + "loss": 1.794, + "step": 655 + }, + { + "epoch": 0.718904109589041, + "grad_norm": 2.258828639984131, + "learning_rate": 4.254670999187653e-06, + "loss": 1.805, + "step": 656 + }, + { + "epoch": 0.72, + "grad_norm": 2.384399175643921, + "learning_rate": 4.252640129975629e-06, + "loss": 1.818, + "step": 657 + }, + { + "epoch": 0.7210958904109589, + "grad_norm": 2.3278932571411133, + "learning_rate": 4.2506092607636075e-06, + "loss": 1.7719, + "step": 658 + }, + { + "epoch": 0.7221917808219178, + "grad_norm": 2.207348108291626, + "learning_rate": 4.248578391551584e-06, + "loss": 1.7362, + "step": 659 + }, + { + "epoch": 0.7232876712328767, + "grad_norm": 2.4525978565216064, + "learning_rate": 4.246547522339562e-06, + "loss": 1.778, + "step": 660 + }, + { + "epoch": 0.7232876712328767, + "eval_loss": 1.9694374799728394, + "eval_runtime": 42.5977, + "eval_samples_per_second": 23.475, + "eval_steps_per_second": 0.986, + "step": 660 + }, + { + "epoch": 0.7243835616438357, + "grad_norm": 2.377835273742676, + "learning_rate": 4.2445166531275385e-06, + "loss": 1.7926, + "step": 661 + }, + { + "epoch": 0.7254794520547945, + "grad_norm": 2.4623866081237793, + "learning_rate": 4.242485783915517e-06, + "loss": 1.7366, + "step": 662 + }, + { + "epoch": 0.7265753424657534, + "grad_norm": 2.542219400405884, + "learning_rate": 4.240454914703493e-06, + "loss": 1.7409, + "step": 663 + }, + { + "epoch": 0.7276712328767123, + "grad_norm": 2.1955819129943848, + "learning_rate": 4.23842404549147e-06, + "loss": 1.7596, + "step": 664 + }, + { + "epoch": 0.7287671232876712, + "grad_norm": 2.5834269523620605, + "learning_rate": 4.236393176279448e-06, + "loss": 1.7343, + "step": 665 + }, + { + "epoch": 0.7298630136986302, + "grad_norm": 2.219660520553589, + "learning_rate": 4.234362307067425e-06, + "loss": 1.8089, + "step": 666 + }, + { + "epoch": 0.7309589041095891, + "grad_norm": 2.283200740814209, + "learning_rate": 4.232331437855402e-06, + "loss": 1.8248, + "step": 667 + }, + { + "epoch": 0.7320547945205479, + "grad_norm": 2.274127244949341, + "learning_rate": 4.2303005686433795e-06, + "loss": 1.825, + "step": 668 + }, + { + "epoch": 0.7331506849315068, + "grad_norm": 2.3550686836242676, + "learning_rate": 4.228269699431357e-06, + "loss": 1.7957, + "step": 669 + }, + { + "epoch": 0.7342465753424657, + "grad_norm": 2.588027238845825, + "learning_rate": 4.226238830219334e-06, + "loss": 1.7865, + "step": 670 + }, + { + "epoch": 0.7342465753424657, + "eval_loss": 1.9662212133407593, + "eval_runtime": 43.2661, + "eval_samples_per_second": 23.113, + "eval_steps_per_second": 0.971, + "step": 670 + }, + { + "epoch": 0.7353424657534247, + "grad_norm": 2.2445170879364014, + "learning_rate": 4.224207961007311e-06, + "loss": 1.791, + "step": 671 + }, + { + "epoch": 0.7364383561643836, + "grad_norm": 2.399174690246582, + "learning_rate": 4.2221770917952886e-06, + "loss": 1.7944, + "step": 672 + }, + { + "epoch": 0.7375342465753425, + "grad_norm": 2.586184501647949, + "learning_rate": 4.220146222583266e-06, + "loss": 1.7856, + "step": 673 + }, + { + "epoch": 0.7386301369863014, + "grad_norm": 2.4174695014953613, + "learning_rate": 4.218115353371243e-06, + "loss": 1.7842, + "step": 674 + }, + { + "epoch": 0.7397260273972602, + "grad_norm": 2.2684640884399414, + "learning_rate": 4.21608448415922e-06, + "loss": 1.7482, + "step": 675 + }, + { + "epoch": 0.7408219178082192, + "grad_norm": 3.1782288551330566, + "learning_rate": 4.214053614947198e-06, + "loss": 1.8286, + "step": 676 + }, + { + "epoch": 0.7419178082191781, + "grad_norm": 2.4046335220336914, + "learning_rate": 4.212022745735175e-06, + "loss": 1.7228, + "step": 677 + }, + { + "epoch": 0.743013698630137, + "grad_norm": 3.3722083568573, + "learning_rate": 4.209991876523152e-06, + "loss": 1.8044, + "step": 678 + }, + { + "epoch": 0.7441095890410959, + "grad_norm": 2.293466329574585, + "learning_rate": 4.2079610073111295e-06, + "loss": 1.7415, + "step": 679 + }, + { + "epoch": 0.7452054794520548, + "grad_norm": 2.28663969039917, + "learning_rate": 4.205930138099107e-06, + "loss": 1.8282, + "step": 680 + }, + { + "epoch": 0.7452054794520548, + "eval_loss": 1.9627619981765747, + "eval_runtime": 43.6092, + "eval_samples_per_second": 22.931, + "eval_steps_per_second": 0.963, + "step": 680 + }, + { + "epoch": 0.7463013698630137, + "grad_norm": 2.255870819091797, + "learning_rate": 4.203899268887084e-06, + "loss": 1.7983, + "step": 681 + }, + { + "epoch": 0.7473972602739726, + "grad_norm": 2.4964773654937744, + "learning_rate": 4.201868399675061e-06, + "loss": 1.8141, + "step": 682 + }, + { + "epoch": 0.7484931506849315, + "grad_norm": 2.3372740745544434, + "learning_rate": 4.199837530463039e-06, + "loss": 1.7142, + "step": 683 + }, + { + "epoch": 0.7495890410958904, + "grad_norm": 2.3506968021392822, + "learning_rate": 4.197806661251016e-06, + "loss": 1.7898, + "step": 684 + }, + { + "epoch": 0.7506849315068493, + "grad_norm": 2.242410898208618, + "learning_rate": 4.195775792038993e-06, + "loss": 1.7677, + "step": 685 + }, + { + "epoch": 0.7517808219178083, + "grad_norm": 2.3864502906799316, + "learning_rate": 4.19374492282697e-06, + "loss": 1.7097, + "step": 686 + }, + { + "epoch": 0.7528767123287671, + "grad_norm": 2.482349157333374, + "learning_rate": 4.191714053614948e-06, + "loss": 1.8052, + "step": 687 + }, + { + "epoch": 0.753972602739726, + "grad_norm": 2.317652940750122, + "learning_rate": 4.189683184402924e-06, + "loss": 1.7692, + "step": 688 + }, + { + "epoch": 0.7550684931506849, + "grad_norm": 2.2777392864227295, + "learning_rate": 4.187652315190902e-06, + "loss": 1.8246, + "step": 689 + }, + { + "epoch": 0.7561643835616438, + "grad_norm": 2.3506035804748535, + "learning_rate": 4.185621445978879e-06, + "loss": 1.7576, + "step": 690 + }, + { + "epoch": 0.7561643835616438, + "eval_loss": 1.9610412120819092, + "eval_runtime": 43.3119, + "eval_samples_per_second": 23.088, + "eval_steps_per_second": 0.97, + "step": 690 + }, + { + "epoch": 0.7572602739726028, + "grad_norm": 2.3011536598205566, + "learning_rate": 4.183590576766857e-06, + "loss": 1.7891, + "step": 691 + }, + { + "epoch": 0.7583561643835617, + "grad_norm": 2.2614853382110596, + "learning_rate": 4.181559707554833e-06, + "loss": 1.8258, + "step": 692 + }, + { + "epoch": 0.7594520547945205, + "grad_norm": 2.219175100326538, + "learning_rate": 4.179528838342811e-06, + "loss": 1.7494, + "step": 693 + }, + { + "epoch": 0.7605479452054794, + "grad_norm": 2.4239423274993896, + "learning_rate": 4.177497969130788e-06, + "loss": 1.7611, + "step": 694 + }, + { + "epoch": 0.7616438356164383, + "grad_norm": 2.470712423324585, + "learning_rate": 4.175467099918766e-06, + "loss": 1.8281, + "step": 695 + }, + { + "epoch": 0.7627397260273973, + "grad_norm": 2.504060745239258, + "learning_rate": 4.173436230706742e-06, + "loss": 1.7641, + "step": 696 + }, + { + "epoch": 0.7638356164383562, + "grad_norm": 2.326601982116699, + "learning_rate": 4.1714053614947205e-06, + "loss": 1.7671, + "step": 697 + }, + { + "epoch": 0.7649315068493151, + "grad_norm": 2.4280030727386475, + "learning_rate": 4.169374492282697e-06, + "loss": 1.7573, + "step": 698 + }, + { + "epoch": 0.766027397260274, + "grad_norm": 2.455716371536255, + "learning_rate": 4.167343623070675e-06, + "loss": 1.809, + "step": 699 + }, + { + "epoch": 0.7671232876712328, + "grad_norm": 2.3029844760894775, + "learning_rate": 4.165312753858652e-06, + "loss": 1.77, + "step": 700 + }, + { + "epoch": 0.7671232876712328, + "eval_loss": 1.9604709148406982, + "eval_runtime": 44.4627, + "eval_samples_per_second": 22.491, + "eval_steps_per_second": 0.945, + "step": 700 + }, + { + "epoch": 0.7682191780821918, + "grad_norm": 2.3833982944488525, + "learning_rate": 4.163281884646629e-06, + "loss": 1.7652, + "step": 701 + }, + { + "epoch": 0.7693150684931507, + "grad_norm": 2.2392966747283936, + "learning_rate": 4.161251015434607e-06, + "loss": 1.7904, + "step": 702 + }, + { + "epoch": 0.7704109589041096, + "grad_norm": 2.4546353816986084, + "learning_rate": 4.159220146222583e-06, + "loss": 1.7324, + "step": 703 + }, + { + "epoch": 0.7715068493150685, + "grad_norm": 2.13592529296875, + "learning_rate": 4.157189277010561e-06, + "loss": 1.759, + "step": 704 + }, + { + "epoch": 0.7726027397260274, + "grad_norm": 2.381657361984253, + "learning_rate": 4.155158407798538e-06, + "loss": 1.7871, + "step": 705 + }, + { + "epoch": 0.7736986301369863, + "grad_norm": 2.4142940044403076, + "learning_rate": 4.153127538586516e-06, + "loss": 1.7832, + "step": 706 + }, + { + "epoch": 0.7747945205479452, + "grad_norm": 2.305697441101074, + "learning_rate": 4.151096669374492e-06, + "loss": 1.7619, + "step": 707 + }, + { + "epoch": 0.7758904109589041, + "grad_norm": 2.193429708480835, + "learning_rate": 4.1490658001624705e-06, + "loss": 1.7738, + "step": 708 + }, + { + "epoch": 0.776986301369863, + "grad_norm": 2.3007619380950928, + "learning_rate": 4.147034930950447e-06, + "loss": 1.7839, + "step": 709 + }, + { + "epoch": 0.7780821917808219, + "grad_norm": 2.6337807178497314, + "learning_rate": 4.145004061738425e-06, + "loss": 1.7748, + "step": 710 + }, + { + "epoch": 0.7780821917808219, + "eval_loss": 1.9597856998443604, + "eval_runtime": 43.8644, + "eval_samples_per_second": 22.798, + "eval_steps_per_second": 0.957, + "step": 710 + }, + { + "epoch": 0.7791780821917809, + "grad_norm": 2.4294793605804443, + "learning_rate": 4.1429731925264015e-06, + "loss": 1.7001, + "step": 711 + }, + { + "epoch": 0.7802739726027397, + "grad_norm": 2.512155532836914, + "learning_rate": 4.140942323314379e-06, + "loss": 1.7334, + "step": 712 + }, + { + "epoch": 0.7813698630136986, + "grad_norm": 2.290808916091919, + "learning_rate": 4.138911454102356e-06, + "loss": 1.7522, + "step": 713 + }, + { + "epoch": 0.7824657534246575, + "grad_norm": 2.483161211013794, + "learning_rate": 4.136880584890333e-06, + "loss": 1.798, + "step": 714 + }, + { + "epoch": 0.7835616438356164, + "grad_norm": 2.2973921298980713, + "learning_rate": 4.1348497156783106e-06, + "loss": 1.7847, + "step": 715 + }, + { + "epoch": 0.7846575342465754, + "grad_norm": 2.253084659576416, + "learning_rate": 4.132818846466288e-06, + "loss": 1.7812, + "step": 716 + }, + { + "epoch": 0.7857534246575343, + "grad_norm": 2.306342840194702, + "learning_rate": 4.130787977254265e-06, + "loss": 1.7886, + "step": 717 + }, + { + "epoch": 0.7868493150684932, + "grad_norm": 2.5153164863586426, + "learning_rate": 4.128757108042242e-06, + "loss": 1.7954, + "step": 718 + }, + { + "epoch": 0.787945205479452, + "grad_norm": 2.4122278690338135, + "learning_rate": 4.12672623883022e-06, + "loss": 1.8525, + "step": 719 + }, + { + "epoch": 0.7890410958904109, + "grad_norm": 2.4520716667175293, + "learning_rate": 4.124695369618197e-06, + "loss": 1.7648, + "step": 720 + }, + { + "epoch": 0.7890410958904109, + "eval_loss": 1.9547661542892456, + "eval_runtime": 45.3774, + "eval_samples_per_second": 22.037, + "eval_steps_per_second": 0.926, + "step": 720 + }, + { + "epoch": 0.7901369863013699, + "grad_norm": 2.2959611415863037, + "learning_rate": 4.122664500406174e-06, + "loss": 1.7626, + "step": 721 + }, + { + "epoch": 0.7912328767123288, + "grad_norm": 2.2280471324920654, + "learning_rate": 4.1206336311941515e-06, + "loss": 1.7954, + "step": 722 + }, + { + "epoch": 0.7923287671232877, + "grad_norm": 2.3870065212249756, + "learning_rate": 4.118602761982129e-06, + "loss": 1.7377, + "step": 723 + }, + { + "epoch": 0.7934246575342466, + "grad_norm": 2.292612075805664, + "learning_rate": 4.116571892770106e-06, + "loss": 1.7, + "step": 724 + }, + { + "epoch": 0.7945205479452054, + "grad_norm": 2.4607021808624268, + "learning_rate": 4.114541023558083e-06, + "loss": 1.8338, + "step": 725 + }, + { + "epoch": 0.7956164383561644, + "grad_norm": 2.532824993133545, + "learning_rate": 4.112510154346061e-06, + "loss": 1.7549, + "step": 726 + }, + { + "epoch": 0.7967123287671233, + "grad_norm": 2.2139623165130615, + "learning_rate": 4.110479285134038e-06, + "loss": 1.78, + "step": 727 + }, + { + "epoch": 0.7978082191780822, + "grad_norm": 2.4749462604522705, + "learning_rate": 4.108448415922015e-06, + "loss": 1.8034, + "step": 728 + }, + { + "epoch": 0.7989041095890411, + "grad_norm": 2.609265089035034, + "learning_rate": 4.1064175467099924e-06, + "loss": 1.7941, + "step": 729 + }, + { + "epoch": 0.8, + "grad_norm": 2.367753028869629, + "learning_rate": 4.10438667749797e-06, + "loss": 1.6653, + "step": 730 + }, + { + "epoch": 0.8, + "eval_loss": 1.9539440870285034, + "eval_runtime": 48.8438, + "eval_samples_per_second": 20.473, + "eval_steps_per_second": 0.86, + "step": 730 + }, + { + "epoch": 0.8010958904109589, + "grad_norm": 2.9723567962646484, + "learning_rate": 4.102355808285947e-06, + "loss": 1.8317, + "step": 731 + }, + { + "epoch": 0.8021917808219178, + "grad_norm": 2.388806104660034, + "learning_rate": 4.100324939073924e-06, + "loss": 1.7597, + "step": 732 + }, + { + "epoch": 0.8032876712328767, + "grad_norm": 2.4882986545562744, + "learning_rate": 4.0982940698619015e-06, + "loss": 1.7708, + "step": 733 + }, + { + "epoch": 0.8043835616438356, + "grad_norm": 2.7586395740509033, + "learning_rate": 4.096263200649879e-06, + "loss": 1.7804, + "step": 734 + }, + { + "epoch": 0.8054794520547945, + "grad_norm": 2.237025260925293, + "learning_rate": 4.094232331437856e-06, + "loss": 1.7415, + "step": 735 + }, + { + "epoch": 0.8065753424657535, + "grad_norm": 2.421330213546753, + "learning_rate": 4.092201462225833e-06, + "loss": 1.734, + "step": 736 + }, + { + "epoch": 0.8076712328767123, + "grad_norm": 2.7281360626220703, + "learning_rate": 4.090170593013811e-06, + "loss": 1.7716, + "step": 737 + }, + { + "epoch": 0.8087671232876712, + "grad_norm": 2.2200379371643066, + "learning_rate": 4.088139723801787e-06, + "loss": 1.7266, + "step": 738 + }, + { + "epoch": 0.8098630136986301, + "grad_norm": 2.465440034866333, + "learning_rate": 4.086108854589765e-06, + "loss": 1.741, + "step": 739 + }, + { + "epoch": 0.810958904109589, + "grad_norm": 2.39835262298584, + "learning_rate": 4.084077985377742e-06, + "loss": 1.757, + "step": 740 + }, + { + "epoch": 0.810958904109589, + "eval_loss": 1.9514222145080566, + "eval_runtime": 52.8068, + "eval_samples_per_second": 18.937, + "eval_steps_per_second": 0.795, + "step": 740 + }, + { + "epoch": 0.812054794520548, + "grad_norm": 2.526684522628784, + "learning_rate": 4.08204711616572e-06, + "loss": 1.8195, + "step": 741 + }, + { + "epoch": 0.8131506849315069, + "grad_norm": 2.341954469680786, + "learning_rate": 4.080016246953696e-06, + "loss": 1.7196, + "step": 742 + }, + { + "epoch": 0.8142465753424658, + "grad_norm": 2.3755977153778076, + "learning_rate": 4.077985377741674e-06, + "loss": 1.8115, + "step": 743 + }, + { + "epoch": 0.8153424657534246, + "grad_norm": 2.3952431678771973, + "learning_rate": 4.075954508529651e-06, + "loss": 1.812, + "step": 744 + }, + { + "epoch": 0.8164383561643835, + "grad_norm": 2.554089069366455, + "learning_rate": 4.073923639317629e-06, + "loss": 1.8184, + "step": 745 + }, + { + "epoch": 0.8175342465753425, + "grad_norm": 2.2319302558898926, + "learning_rate": 4.071892770105605e-06, + "loss": 1.7448, + "step": 746 + }, + { + "epoch": 0.8186301369863014, + "grad_norm": 2.284856081008911, + "learning_rate": 4.069861900893583e-06, + "loss": 1.7154, + "step": 747 + }, + { + "epoch": 0.8197260273972603, + "grad_norm": 2.4081835746765137, + "learning_rate": 4.06783103168156e-06, + "loss": 1.7311, + "step": 748 + }, + { + "epoch": 0.8208219178082192, + "grad_norm": 2.6115968227386475, + "learning_rate": 4.065800162469538e-06, + "loss": 1.7436, + "step": 749 + }, + { + "epoch": 0.821917808219178, + "grad_norm": 2.368004322052002, + "learning_rate": 4.063769293257514e-06, + "loss": 1.7811, + "step": 750 + }, + { + "epoch": 0.821917808219178, + "eval_loss": 1.9508955478668213, + "eval_runtime": 46.8711, + "eval_samples_per_second": 21.335, + "eval_steps_per_second": 0.896, + "step": 750 + }, + { + "epoch": 0.823013698630137, + "grad_norm": 2.4607419967651367, + "learning_rate": 4.061738424045492e-06, + "loss": 1.7563, + "step": 751 + }, + { + "epoch": 0.8241095890410959, + "grad_norm": 2.2978763580322266, + "learning_rate": 4.059707554833469e-06, + "loss": 1.7755, + "step": 752 + }, + { + "epoch": 0.8252054794520548, + "grad_norm": 2.3316237926483154, + "learning_rate": 4.057676685621446e-06, + "loss": 1.8226, + "step": 753 + }, + { + "epoch": 0.8263013698630137, + "grad_norm": 2.3699169158935547, + "learning_rate": 4.0556458164094235e-06, + "loss": 1.7755, + "step": 754 + }, + { + "epoch": 0.8273972602739726, + "grad_norm": 2.195596218109131, + "learning_rate": 4.053614947197401e-06, + "loss": 1.6916, + "step": 755 + }, + { + "epoch": 0.8284931506849315, + "grad_norm": 2.4524662494659424, + "learning_rate": 4.051584077985378e-06, + "loss": 1.774, + "step": 756 + }, + { + "epoch": 0.8295890410958904, + "grad_norm": 2.573888063430786, + "learning_rate": 4.049553208773355e-06, + "loss": 1.7953, + "step": 757 + }, + { + "epoch": 0.8306849315068493, + "grad_norm": 2.5157253742218018, + "learning_rate": 4.047522339561333e-06, + "loss": 1.7597, + "step": 758 + }, + { + "epoch": 0.8317808219178082, + "grad_norm": 2.5588185787200928, + "learning_rate": 4.04549147034931e-06, + "loss": 1.7553, + "step": 759 + }, + { + "epoch": 0.8328767123287671, + "grad_norm": 2.8950047492980957, + "learning_rate": 4.043460601137287e-06, + "loss": 1.8647, + "step": 760 + }, + { + "epoch": 0.8328767123287671, + "eval_loss": 1.945882797241211, + "eval_runtime": 45.481, + "eval_samples_per_second": 21.987, + "eval_steps_per_second": 0.923, + "step": 760 + }, + { + "epoch": 0.8339726027397261, + "grad_norm": 2.217116117477417, + "learning_rate": 4.041429731925264e-06, + "loss": 1.7465, + "step": 761 + }, + { + "epoch": 0.8350684931506849, + "grad_norm": 2.5601565837860107, + "learning_rate": 4.039398862713242e-06, + "loss": 1.7575, + "step": 762 + }, + { + "epoch": 0.8361643835616438, + "grad_norm": 2.4540457725524902, + "learning_rate": 4.037367993501219e-06, + "loss": 1.8291, + "step": 763 + }, + { + "epoch": 0.8372602739726027, + "grad_norm": 2.327181100845337, + "learning_rate": 4.035337124289196e-06, + "loss": 1.7409, + "step": 764 + }, + { + "epoch": 0.8383561643835616, + "grad_norm": 2.485457181930542, + "learning_rate": 4.0333062550771735e-06, + "loss": 1.7192, + "step": 765 + }, + { + "epoch": 0.8394520547945206, + "grad_norm": 2.4613442420959473, + "learning_rate": 4.031275385865151e-06, + "loss": 1.7613, + "step": 766 + }, + { + "epoch": 0.8405479452054795, + "grad_norm": 2.303217649459839, + "learning_rate": 4.029244516653128e-06, + "loss": 1.7799, + "step": 767 + }, + { + "epoch": 0.8416438356164384, + "grad_norm": 2.3830621242523193, + "learning_rate": 4.027213647441105e-06, + "loss": 1.7045, + "step": 768 + }, + { + "epoch": 0.8427397260273972, + "grad_norm": 2.342332601547241, + "learning_rate": 4.025182778229083e-06, + "loss": 1.7246, + "step": 769 + }, + { + "epoch": 0.8438356164383561, + "grad_norm": 2.4014711380004883, + "learning_rate": 4.02315190901706e-06, + "loss": 1.7621, + "step": 770 + }, + { + "epoch": 0.8438356164383561, + "eval_loss": 1.95199453830719, + "eval_runtime": 45.4754, + "eval_samples_per_second": 21.99, + "eval_steps_per_second": 0.924, + "step": 770 + }, + { + "epoch": 0.8449315068493151, + "grad_norm": 2.6015400886535645, + "learning_rate": 4.021121039805037e-06, + "loss": 1.7536, + "step": 771 + }, + { + "epoch": 0.846027397260274, + "grad_norm": 2.866978406906128, + "learning_rate": 4.0190901705930144e-06, + "loss": 1.8405, + "step": 772 + }, + { + "epoch": 0.8471232876712329, + "grad_norm": 2.389810085296631, + "learning_rate": 4.017059301380992e-06, + "loss": 1.7365, + "step": 773 + }, + { + "epoch": 0.8482191780821918, + "grad_norm": 2.280442953109741, + "learning_rate": 4.015028432168969e-06, + "loss": 1.8273, + "step": 774 + }, + { + "epoch": 0.8493150684931506, + "grad_norm": 2.509330987930298, + "learning_rate": 4.012997562956945e-06, + "loss": 1.7127, + "step": 775 + }, + { + "epoch": 0.8504109589041096, + "grad_norm": 2.452078104019165, + "learning_rate": 4.0109666937449235e-06, + "loss": 1.7457, + "step": 776 + }, + { + "epoch": 0.8515068493150685, + "grad_norm": 2.368337631225586, + "learning_rate": 4.0089358245329e-06, + "loss": 1.7412, + "step": 777 + }, + { + "epoch": 0.8526027397260274, + "grad_norm": 2.4529192447662354, + "learning_rate": 4.006904955320878e-06, + "loss": 1.7434, + "step": 778 + }, + { + "epoch": 0.8536986301369863, + "grad_norm": 2.647761821746826, + "learning_rate": 4.0048740861088545e-06, + "loss": 1.8013, + "step": 779 + }, + { + "epoch": 0.8547945205479452, + "grad_norm": 2.387424945831299, + "learning_rate": 4.002843216896833e-06, + "loss": 1.7948, + "step": 780 + }, + { + "epoch": 0.8547945205479452, + "eval_loss": 1.946029543876648, + "eval_runtime": 43.9666, + "eval_samples_per_second": 22.745, + "eval_steps_per_second": 0.955, + "step": 780 + }, + { + "epoch": 0.8558904109589041, + "grad_norm": 2.9217846393585205, + "learning_rate": 4.000812347684809e-06, + "loss": 1.7384, + "step": 781 + }, + { + "epoch": 0.856986301369863, + "grad_norm": 2.400040864944458, + "learning_rate": 3.998781478472787e-06, + "loss": 1.7496, + "step": 782 + }, + { + "epoch": 0.8580821917808219, + "grad_norm": 2.4050440788269043, + "learning_rate": 3.996750609260764e-06, + "loss": 1.7165, + "step": 783 + }, + { + "epoch": 0.8591780821917808, + "grad_norm": 3.6001529693603516, + "learning_rate": 3.994719740048742e-06, + "loss": 1.8679, + "step": 784 + }, + { + "epoch": 0.8602739726027397, + "grad_norm": 2.4040098190307617, + "learning_rate": 3.992688870836718e-06, + "loss": 1.7671, + "step": 785 + }, + { + "epoch": 0.8613698630136987, + "grad_norm": 2.5645527839660645, + "learning_rate": 3.990658001624696e-06, + "loss": 1.7374, + "step": 786 + }, + { + "epoch": 0.8624657534246576, + "grad_norm": 2.799394130706787, + "learning_rate": 3.988627132412673e-06, + "loss": 1.796, + "step": 787 + }, + { + "epoch": 0.8635616438356164, + "grad_norm": 2.5481607913970947, + "learning_rate": 3.98659626320065e-06, + "loss": 1.753, + "step": 788 + }, + { + "epoch": 0.8646575342465753, + "grad_norm": 2.8851330280303955, + "learning_rate": 3.984565393988627e-06, + "loss": 1.7505, + "step": 789 + }, + { + "epoch": 0.8657534246575342, + "grad_norm": 2.6668694019317627, + "learning_rate": 3.9825345247766046e-06, + "loss": 1.7604, + "step": 790 + }, + { + "epoch": 0.8657534246575342, + "eval_loss": 1.939759612083435, + "eval_runtime": 44.5019, + "eval_samples_per_second": 22.471, + "eval_steps_per_second": 0.944, + "step": 790 + }, + { + "epoch": 0.8668493150684932, + "grad_norm": 2.320935010910034, + "learning_rate": 3.980503655564582e-06, + "loss": 1.7749, + "step": 791 + }, + { + "epoch": 0.8679452054794521, + "grad_norm": 2.829897880554199, + "learning_rate": 3.978472786352559e-06, + "loss": 1.7763, + "step": 792 + }, + { + "epoch": 0.869041095890411, + "grad_norm": 2.8420908451080322, + "learning_rate": 3.976441917140536e-06, + "loss": 1.7598, + "step": 793 + }, + { + "epoch": 0.8701369863013698, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.8183, + "step": 794 + }, + { + "epoch": 0.8712328767123287, + "grad_norm": 2.5250179767608643, + "learning_rate": 3.974411047928514e-06, + "loss": 1.781, + "step": 795 + }, + { + "epoch": 0.8723287671232877, + "grad_norm": 3.9271891117095947, + "learning_rate": 3.972380178716491e-06, + "loss": 1.7949, + "step": 796 + }, + { + "epoch": 0.8734246575342466, + "grad_norm": 2.277897834777832, + "learning_rate": 3.970349309504468e-06, + "loss": 1.707, + "step": 797 + }, + { + "epoch": 0.8745205479452055, + "grad_norm": 3.241063117980957, + "learning_rate": 3.9683184402924455e-06, + "loss": 1.7175, + "step": 798 + }, + { + "epoch": 0.8756164383561644, + "grad_norm": 2.514458417892456, + "learning_rate": 3.966287571080423e-06, + "loss": 1.7764, + "step": 799 + }, + { + "epoch": 0.8767123287671232, + "grad_norm": 2.420212745666504, + "learning_rate": 3.9642567018684e-06, + "loss": 1.7618, + "step": 800 + }, + { + "epoch": 0.8767123287671232, + "eval_loss": 1.9401386976242065, + "eval_runtime": 44.1672, + "eval_samples_per_second": 22.641, + "eval_steps_per_second": 0.951, + "step": 800 + }, + { + "epoch": 0.8778082191780822, + "grad_norm": 2.3980538845062256, + "learning_rate": 3.962225832656377e-06, + "loss": 1.6922, + "step": 801 + }, + { + "epoch": 0.8789041095890411, + "grad_norm": 2.892646312713623, + "learning_rate": 3.960194963444355e-06, + "loss": 1.7571, + "step": 802 + }, + { + "epoch": 0.88, + "grad_norm": 2.499260902404785, + "learning_rate": 3.958164094232332e-06, + "loss": 1.7501, + "step": 803 + }, + { + "epoch": 0.8810958904109589, + "grad_norm": 2.409695625305176, + "learning_rate": 3.956133225020309e-06, + "loss": 1.7553, + "step": 804 + }, + { + "epoch": 0.8821917808219178, + "grad_norm": 2.640069007873535, + "learning_rate": 3.954102355808286e-06, + "loss": 1.7526, + "step": 805 + }, + { + "epoch": 0.8832876712328767, + "grad_norm": 2.9101760387420654, + "learning_rate": 3.952071486596264e-06, + "loss": 1.7483, + "step": 806 + }, + { + "epoch": 0.8843835616438356, + "grad_norm": 2.2579474449157715, + "learning_rate": 3.950040617384241e-06, + "loss": 1.7379, + "step": 807 + }, + { + "epoch": 0.8854794520547945, + "grad_norm": 4.039298057556152, + "learning_rate": 3.948009748172218e-06, + "loss": 1.7988, + "step": 808 + }, + { + "epoch": 0.8865753424657534, + "grad_norm": 2.3232030868530273, + "learning_rate": 3.9459788789601955e-06, + "loss": 1.7592, + "step": 809 + }, + { + "epoch": 0.8876712328767123, + "grad_norm": 2.697164297103882, + "learning_rate": 3.943948009748173e-06, + "loss": 1.7622, + "step": 810 + }, + { + "epoch": 0.8876712328767123, + "eval_loss": 1.9367746114730835, + "eval_runtime": 43.7013, + "eval_samples_per_second": 22.883, + "eval_steps_per_second": 0.961, + "step": 810 + }, + { + "epoch": 0.8887671232876713, + "grad_norm": 2.480297803878784, + "learning_rate": 3.94191714053615e-06, + "loss": 1.8006, + "step": 811 + }, + { + "epoch": 0.8898630136986302, + "grad_norm": 2.3624603748321533, + "learning_rate": 3.939886271324127e-06, + "loss": 1.7152, + "step": 812 + }, + { + "epoch": 0.890958904109589, + "grad_norm": 2.417187213897705, + "learning_rate": 3.937855402112104e-06, + "loss": 1.7077, + "step": 813 + }, + { + "epoch": 0.8920547945205479, + "grad_norm": 2.6575164794921875, + "learning_rate": 3.935824532900082e-06, + "loss": 1.7744, + "step": 814 + }, + { + "epoch": 0.8931506849315068, + "grad_norm": 2.4896557331085205, + "learning_rate": 3.933793663688058e-06, + "loss": 1.8052, + "step": 815 + }, + { + "epoch": 0.8942465753424658, + "grad_norm": 2.475693941116333, + "learning_rate": 3.9317627944760365e-06, + "loss": 1.7008, + "step": 816 + }, + { + "epoch": 0.8953424657534247, + "grad_norm": 2.3210372924804688, + "learning_rate": 3.929731925264013e-06, + "loss": 1.7339, + "step": 817 + }, + { + "epoch": 0.8964383561643836, + "grad_norm": 2.433785915374756, + "learning_rate": 3.927701056051991e-06, + "loss": 1.7544, + "step": 818 + }, + { + "epoch": 0.8975342465753424, + "grad_norm": 3.181574583053589, + "learning_rate": 3.9256701868399674e-06, + "loss": 1.7229, + "step": 819 + }, + { + "epoch": 0.8986301369863013, + "grad_norm": 2.704399347305298, + "learning_rate": 3.9236393176279456e-06, + "loss": 1.7086, + "step": 820 + }, + { + "epoch": 0.8986301369863013, + "eval_loss": 1.9335075616836548, + "eval_runtime": 47.0248, + "eval_samples_per_second": 21.265, + "eval_steps_per_second": 0.893, + "step": 820 + }, + { + "epoch": 0.8997260273972603, + "grad_norm": 3.5027623176574707, + "learning_rate": 3.921608448415922e-06, + "loss": 1.7622, + "step": 821 + }, + { + "epoch": 0.9008219178082192, + "grad_norm": 2.450943946838379, + "learning_rate": 3.9195775792039e-06, + "loss": 1.7286, + "step": 822 + }, + { + "epoch": 0.9019178082191781, + "grad_norm": 2.4603374004364014, + "learning_rate": 3.9175467099918765e-06, + "loss": 1.7725, + "step": 823 + }, + { + "epoch": 0.903013698630137, + "grad_norm": 2.8217649459838867, + "learning_rate": 3.915515840779855e-06, + "loss": 1.7859, + "step": 824 + }, + { + "epoch": 0.9041095890410958, + "grad_norm": 2.3308372497558594, + "learning_rate": 3.913484971567831e-06, + "loss": 1.6874, + "step": 825 + }, + { + "epoch": 0.9052054794520548, + "grad_norm": 2.624776840209961, + "learning_rate": 3.911454102355808e-06, + "loss": 1.7632, + "step": 826 + }, + { + "epoch": 0.9063013698630137, + "grad_norm": 2.4097838401794434, + "learning_rate": 3.909423233143786e-06, + "loss": 1.7576, + "step": 827 + }, + { + "epoch": 0.9073972602739726, + "grad_norm": 2.487410545349121, + "learning_rate": 3.907392363931763e-06, + "loss": 1.831, + "step": 828 + }, + { + "epoch": 0.9084931506849315, + "grad_norm": 2.647245168685913, + "learning_rate": 3.90536149471974e-06, + "loss": 1.8448, + "step": 829 + }, + { + "epoch": 0.9095890410958904, + "grad_norm": 2.397664785385132, + "learning_rate": 3.9033306255077175e-06, + "loss": 1.868, + "step": 830 + }, + { + "epoch": 0.9095890410958904, + "eval_loss": 1.930379033088684, + "eval_runtime": 51.1662, + "eval_samples_per_second": 19.544, + "eval_steps_per_second": 0.821, + "step": 830 + }, + { + "epoch": 0.9106849315068493, + "grad_norm": 2.277433395385742, + "learning_rate": 3.901299756295695e-06, + "loss": 1.7394, + "step": 831 + }, + { + "epoch": 0.9117808219178082, + "grad_norm": 3.1312496662139893, + "learning_rate": 3.899268887083672e-06, + "loss": 1.7422, + "step": 832 + }, + { + "epoch": 0.9128767123287671, + "grad_norm": 2.5011534690856934, + "learning_rate": 3.897238017871649e-06, + "loss": 1.7484, + "step": 833 + }, + { + "epoch": 0.913972602739726, + "grad_norm": 2.8716166019439697, + "learning_rate": 3.8952071486596266e-06, + "loss": 1.7253, + "step": 834 + }, + { + "epoch": 0.915068493150685, + "grad_norm": 2.4346914291381836, + "learning_rate": 3.893176279447604e-06, + "loss": 1.8197, + "step": 835 + }, + { + "epoch": 0.9161643835616439, + "grad_norm": 2.348475933074951, + "learning_rate": 3.891145410235581e-06, + "loss": 1.7737, + "step": 836 + }, + { + "epoch": 0.9172602739726028, + "grad_norm": 2.3996007442474365, + "learning_rate": 3.889114541023558e-06, + "loss": 1.7333, + "step": 837 + }, + { + "epoch": 0.9183561643835616, + "grad_norm": 2.355262279510498, + "learning_rate": 3.887083671811536e-06, + "loss": 1.6902, + "step": 838 + }, + { + "epoch": 0.9194520547945205, + "grad_norm": 2.335932731628418, + "learning_rate": 3.885052802599513e-06, + "loss": 1.8206, + "step": 839 + }, + { + "epoch": 0.9205479452054794, + "grad_norm": 2.2893190383911133, + "learning_rate": 3.88302193338749e-06, + "loss": 1.7533, + "step": 840 + }, + { + "epoch": 0.9205479452054794, + "eval_loss": 1.9290764331817627, + "eval_runtime": 46.9373, + "eval_samples_per_second": 21.305, + "eval_steps_per_second": 0.895, + "step": 840 + }, + { + "epoch": 0.9216438356164384, + "grad_norm": 2.739842176437378, + "learning_rate": 3.8809910641754675e-06, + "loss": 1.7318, + "step": 841 + }, + { + "epoch": 0.9227397260273973, + "grad_norm": 2.587916374206543, + "learning_rate": 3.878960194963445e-06, + "loss": 1.7979, + "step": 842 + }, + { + "epoch": 0.9238356164383562, + "grad_norm": 2.406548500061035, + "learning_rate": 3.876929325751422e-06, + "loss": 1.6899, + "step": 843 + }, + { + "epoch": 0.924931506849315, + "grad_norm": 2.440321922302246, + "learning_rate": 3.874898456539399e-06, + "loss": 1.7546, + "step": 844 + }, + { + "epoch": 0.9260273972602739, + "grad_norm": 2.320030450820923, + "learning_rate": 3.872867587327377e-06, + "loss": 1.7254, + "step": 845 + }, + { + "epoch": 0.9271232876712329, + "grad_norm": 2.3795316219329834, + "learning_rate": 3.870836718115354e-06, + "loss": 1.728, + "step": 846 + }, + { + "epoch": 0.9282191780821918, + "grad_norm": 2.299802541732788, + "learning_rate": 3.868805848903331e-06, + "loss": 1.7464, + "step": 847 + }, + { + "epoch": 0.9293150684931507, + "grad_norm": 2.469719409942627, + "learning_rate": 3.8667749796913084e-06, + "loss": 1.765, + "step": 848 + }, + { + "epoch": 0.9304109589041096, + "grad_norm": 2.612109899520874, + "learning_rate": 3.864744110479286e-06, + "loss": 1.7827, + "step": 849 + }, + { + "epoch": 0.9315068493150684, + "grad_norm": 2.623173952102661, + "learning_rate": 3.862713241267262e-06, + "loss": 1.8019, + "step": 850 + }, + { + "epoch": 0.9315068493150684, + "eval_loss": 1.9268505573272705, + "eval_runtime": 46.77, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 850 + }, + { + "epoch": 0.9326027397260274, + "grad_norm": 2.4146616458892822, + "learning_rate": 3.86068237205524e-06, + "loss": 1.7498, + "step": 851 + }, + { + "epoch": 0.9336986301369863, + "grad_norm": 2.361994743347168, + "learning_rate": 3.858651502843217e-06, + "loss": 1.6926, + "step": 852 + }, + { + "epoch": 0.9347945205479452, + "grad_norm": 2.493215560913086, + "learning_rate": 3.856620633631195e-06, + "loss": 1.736, + "step": 853 + }, + { + "epoch": 0.9358904109589041, + "grad_norm": 2.4341812133789062, + "learning_rate": 3.854589764419171e-06, + "loss": 1.8152, + "step": 854 + }, + { + "epoch": 0.936986301369863, + "grad_norm": 2.391331195831299, + "learning_rate": 3.852558895207149e-06, + "loss": 1.7717, + "step": 855 + }, + { + "epoch": 0.938082191780822, + "grad_norm": 2.419492244720459, + "learning_rate": 3.850528025995126e-06, + "loss": 1.7283, + "step": 856 + }, + { + "epoch": 0.9391780821917808, + "grad_norm": 2.349090576171875, + "learning_rate": 3.848497156783104e-06, + "loss": 1.7317, + "step": 857 + }, + { + "epoch": 0.9402739726027397, + "grad_norm": 2.298647880554199, + "learning_rate": 3.84646628757108e-06, + "loss": 1.7389, + "step": 858 + }, + { + "epoch": 0.9413698630136986, + "grad_norm": 2.4612667560577393, + "learning_rate": 3.8444354183590585e-06, + "loss": 1.7333, + "step": 859 + }, + { + "epoch": 0.9424657534246575, + "grad_norm": 2.3139944076538086, + "learning_rate": 3.842404549147035e-06, + "loss": 1.6944, + "step": 860 + }, + { + "epoch": 0.9424657534246575, + "eval_loss": 1.924438714981079, + "eval_runtime": 45.8276, + "eval_samples_per_second": 21.821, + "eval_steps_per_second": 0.916, + "step": 860 + }, + { + "epoch": 0.9435616438356165, + "grad_norm": 2.17958664894104, + "learning_rate": 3.840373679935013e-06, + "loss": 1.7474, + "step": 861 + }, + { + "epoch": 0.9446575342465754, + "grad_norm": 2.337859869003296, + "learning_rate": 3.8383428107229894e-06, + "loss": 1.7345, + "step": 862 + }, + { + "epoch": 0.9457534246575342, + "grad_norm": 2.4777672290802, + "learning_rate": 3.836311941510967e-06, + "loss": 1.77, + "step": 863 + }, + { + "epoch": 0.9468493150684931, + "grad_norm": 2.280158519744873, + "learning_rate": 3.834281072298944e-06, + "loss": 1.7069, + "step": 864 + }, + { + "epoch": 0.947945205479452, + "grad_norm": 2.2387664318084717, + "learning_rate": 3.832250203086921e-06, + "loss": 1.7889, + "step": 865 + }, + { + "epoch": 0.949041095890411, + "grad_norm": 2.373771905899048, + "learning_rate": 3.8302193338748985e-06, + "loss": 1.8268, + "step": 866 + }, + { + "epoch": 0.9501369863013699, + "grad_norm": 2.4950132369995117, + "learning_rate": 3.828188464662876e-06, + "loss": 1.7586, + "step": 867 + }, + { + "epoch": 0.9512328767123288, + "grad_norm": 2.3296444416046143, + "learning_rate": 3.826157595450853e-06, + "loss": 1.6991, + "step": 868 + }, + { + "epoch": 0.9523287671232876, + "grad_norm": 2.6653530597686768, + "learning_rate": 3.82412672623883e-06, + "loss": 1.7505, + "step": 869 + }, + { + "epoch": 0.9534246575342465, + "grad_norm": 2.485128402709961, + "learning_rate": 3.822095857026808e-06, + "loss": 1.7196, + "step": 870 + }, + { + "epoch": 0.9534246575342465, + "eval_loss": 1.9221889972686768, + "eval_runtime": 46.6581, + "eval_samples_per_second": 21.433, + "eval_steps_per_second": 0.9, + "step": 870 + }, + { + "epoch": 0.9545205479452055, + "grad_norm": 2.3726181983947754, + "learning_rate": 3.820064987814785e-06, + "loss": 1.7296, + "step": 871 + }, + { + "epoch": 0.9556164383561644, + "grad_norm": 2.233992099761963, + "learning_rate": 3.818034118602762e-06, + "loss": 1.7313, + "step": 872 + }, + { + "epoch": 0.9567123287671233, + "grad_norm": 2.5388147830963135, + "learning_rate": 3.8160032493907395e-06, + "loss": 1.7435, + "step": 873 + }, + { + "epoch": 0.9578082191780822, + "grad_norm": 2.4787333011627197, + "learning_rate": 3.813972380178717e-06, + "loss": 1.7733, + "step": 874 + }, + { + "epoch": 0.958904109589041, + "grad_norm": 2.232564926147461, + "learning_rate": 3.811941510966694e-06, + "loss": 1.693, + "step": 875 + }, + { + "epoch": 0.96, + "grad_norm": 2.3882386684417725, + "learning_rate": 3.809910641754671e-06, + "loss": 1.6529, + "step": 876 + }, + { + "epoch": 0.9610958904109589, + "grad_norm": 2.4785983562469482, + "learning_rate": 3.8078797725426486e-06, + "loss": 1.7456, + "step": 877 + }, + { + "epoch": 0.9621917808219178, + "grad_norm": 2.685098886489868, + "learning_rate": 3.8058489033306254e-06, + "loss": 1.7162, + "step": 878 + }, + { + "epoch": 0.9632876712328767, + "grad_norm": 2.353523015975952, + "learning_rate": 3.803818034118603e-06, + "loss": 1.7852, + "step": 879 + }, + { + "epoch": 0.9643835616438357, + "grad_norm": 2.342163562774658, + "learning_rate": 3.80178716490658e-06, + "loss": 1.7767, + "step": 880 + }, + { + "epoch": 0.9643835616438357, + "eval_loss": 1.9235820770263672, + "eval_runtime": 47.1327, + "eval_samples_per_second": 21.217, + "eval_steps_per_second": 0.891, + "step": 880 + }, + { + "epoch": 0.9654794520547946, + "grad_norm": 2.550818920135498, + "learning_rate": 3.7997562956945577e-06, + "loss": 1.7328, + "step": 881 + }, + { + "epoch": 0.9665753424657534, + "grad_norm": 2.3204798698425293, + "learning_rate": 3.7977254264825345e-06, + "loss": 1.7058, + "step": 882 + }, + { + "epoch": 0.9676712328767123, + "grad_norm": 2.4752719402313232, + "learning_rate": 3.7956945572705122e-06, + "loss": 1.8051, + "step": 883 + }, + { + "epoch": 0.9687671232876712, + "grad_norm": 2.5202794075012207, + "learning_rate": 3.793663688058489e-06, + "loss": 1.7966, + "step": 884 + }, + { + "epoch": 0.9698630136986301, + "grad_norm": 2.26521635055542, + "learning_rate": 3.791632818846467e-06, + "loss": 1.7879, + "step": 885 + }, + { + "epoch": 0.9709589041095891, + "grad_norm": 2.331515312194824, + "learning_rate": 3.7896019496344436e-06, + "loss": 1.7814, + "step": 886 + }, + { + "epoch": 0.972054794520548, + "grad_norm": 2.37178635597229, + "learning_rate": 3.787571080422421e-06, + "loss": 1.7635, + "step": 887 + }, + { + "epoch": 0.9731506849315068, + "grad_norm": 2.321417808532715, + "learning_rate": 3.785540211210398e-06, + "loss": 1.7513, + "step": 888 + }, + { + "epoch": 0.9742465753424657, + "grad_norm": 2.485989809036255, + "learning_rate": 3.7835093419983755e-06, + "loss": 1.7306, + "step": 889 + }, + { + "epoch": 0.9753424657534246, + "grad_norm": 2.7497379779815674, + "learning_rate": 3.7814784727863527e-06, + "loss": 1.8409, + "step": 890 + }, + { + "epoch": 0.9753424657534246, + "eval_loss": 1.9275448322296143, + "eval_runtime": 46.2088, + "eval_samples_per_second": 21.641, + "eval_steps_per_second": 0.909, + "step": 890 + }, + { + "epoch": 0.9764383561643836, + "grad_norm": 4.9214019775390625, + "learning_rate": 3.77944760357433e-06, + "loss": 1.7834, + "step": 891 + }, + { + "epoch": 0.9775342465753425, + "grad_norm": 4.505277156829834, + "learning_rate": 3.7774167343623073e-06, + "loss": 1.7179, + "step": 892 + }, + { + "epoch": 0.9786301369863014, + "grad_norm": 2.5772030353546143, + "learning_rate": 3.7753858651502846e-06, + "loss": 1.7766, + "step": 893 + }, + { + "epoch": 0.9797260273972602, + "grad_norm": 2.254185438156128, + "learning_rate": 3.773354995938262e-06, + "loss": 1.7697, + "step": 894 + }, + { + "epoch": 0.9808219178082191, + "grad_norm": 3.632233142852783, + "learning_rate": 3.771324126726239e-06, + "loss": 1.7968, + "step": 895 + }, + { + "epoch": 0.9819178082191781, + "grad_norm": 2.174461841583252, + "learning_rate": 3.7692932575142164e-06, + "loss": 1.7346, + "step": 896 + }, + { + "epoch": 0.983013698630137, + "grad_norm": 2.3983266353607178, + "learning_rate": 3.7672623883021937e-06, + "loss": 1.6534, + "step": 897 + }, + { + "epoch": 0.9841095890410959, + "grad_norm": 3.393791913986206, + "learning_rate": 3.765231519090171e-06, + "loss": 1.7561, + "step": 898 + }, + { + "epoch": 0.9852054794520548, + "grad_norm": 2.2853987216949463, + "learning_rate": 3.7632006498781482e-06, + "loss": 1.6963, + "step": 899 + }, + { + "epoch": 0.9863013698630136, + "grad_norm": 2.649789810180664, + "learning_rate": 3.761169780666125e-06, + "loss": 1.7395, + "step": 900 + }, + { + "epoch": 0.9863013698630136, + "eval_loss": 1.9242008924484253, + "eval_runtime": 46.6017, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 900 + }, + { + "epoch": 0.9873972602739726, + "grad_norm": 3.417872190475464, + "learning_rate": 3.7591389114541028e-06, + "loss": 1.784, + "step": 901 + }, + { + "epoch": 0.9884931506849315, + "grad_norm": 2.316666841506958, + "learning_rate": 3.7571080422420796e-06, + "loss": 1.7154, + "step": 902 + }, + { + "epoch": 0.9895890410958904, + "grad_norm": 2.486161231994629, + "learning_rate": 3.7550771730300573e-06, + "loss": 1.7903, + "step": 903 + }, + { + "epoch": 0.9906849315068493, + "grad_norm": 4.538443565368652, + "learning_rate": 3.753046303818034e-06, + "loss": 1.7711, + "step": 904 + }, + { + "epoch": 0.9917808219178083, + "grad_norm": 2.9772889614105225, + "learning_rate": 3.751015434606012e-06, + "loss": 1.8046, + "step": 905 + }, + { + "epoch": 0.9928767123287672, + "grad_norm": 7.2353715896606445, + "learning_rate": 3.7489845653939887e-06, + "loss": 1.7189, + "step": 906 + }, + { + "epoch": 0.993972602739726, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.736, + "step": 907 + }, + { + "epoch": 0.9950684931506849, + "grad_norm": 3.0353152751922607, + "learning_rate": 3.7469536961819664e-06, + "loss": 1.8026, + "step": 908 + }, + { + "epoch": 0.9961643835616438, + "grad_norm": 12.551762580871582, + "learning_rate": 3.7449228269699433e-06, + "loss": 1.7788, + "step": 909 + }, + { + "epoch": 0.9972602739726028, + "grad_norm": 3.0898046493530273, + "learning_rate": 3.742891957757921e-06, + "loss": 1.7878, + "step": 910 + }, + { + "epoch": 0.9972602739726028, + "eval_loss": 1.9646916389465332, + "eval_runtime": 44.1583, + "eval_samples_per_second": 22.646, + "eval_steps_per_second": 0.951, + "step": 910 + }, + { + "epoch": 0.9983561643835617, + "grad_norm": 8.120022773742676, + "learning_rate": 3.740861088545898e-06, + "loss": 1.768, + "step": 911 + }, + { + "epoch": 0.9994520547945206, + "grad_norm": 6.485774040222168, + "learning_rate": 3.7388302193338755e-06, + "loss": 1.746, + "step": 912 + }, + { + "epoch": 1.0005479452054795, + "grad_norm": 2.85135555267334, + "learning_rate": 3.7367993501218524e-06, + "loss": 1.6968, + "step": 913 + }, + { + "epoch": 1.0016438356164383, + "grad_norm": 5.8551435470581055, + "learning_rate": 3.7347684809098297e-06, + "loss": 1.6796, + "step": 914 + }, + { + "epoch": 1.0027397260273974, + "grad_norm": 4.45880126953125, + "learning_rate": 3.732737611697807e-06, + "loss": 1.7219, + "step": 915 + }, + { + "epoch": 1.0038356164383562, + "grad_norm": 3.0435831546783447, + "learning_rate": 3.7307067424857842e-06, + "loss": 1.6959, + "step": 916 + }, + { + "epoch": 1.004931506849315, + "grad_norm": 2.8194847106933594, + "learning_rate": 3.7286758732737615e-06, + "loss": 1.8152, + "step": 917 + }, + { + "epoch": 1.006027397260274, + "grad_norm": 3.1266117095947266, + "learning_rate": 3.7266450040617388e-06, + "loss": 1.6937, + "step": 918 + }, + { + "epoch": 1.0071232876712328, + "grad_norm": 3.642854690551758, + "learning_rate": 3.724614134849716e-06, + "loss": 1.7992, + "step": 919 + }, + { + "epoch": 1.0082191780821919, + "grad_norm": 3.344496965408325, + "learning_rate": 3.7225832656376933e-06, + "loss": 1.7583, + "step": 920 + }, + { + "epoch": 1.0082191780821919, + "eval_loss": 1.9293720722198486, + "eval_runtime": 50.8602, + "eval_samples_per_second": 19.662, + "eval_steps_per_second": 0.826, + "step": 920 + }, + { + "epoch": 1.0093150684931507, + "grad_norm": 2.7666447162628174, + "learning_rate": 3.7205523964256706e-06, + "loss": 1.7441, + "step": 921 + }, + { + "epoch": 1.0104109589041095, + "grad_norm": 2.8819239139556885, + "learning_rate": 3.718521527213648e-06, + "loss": 1.6314, + "step": 922 + }, + { + "epoch": 1.0115068493150685, + "grad_norm": 2.6030640602111816, + "learning_rate": 3.716490658001625e-06, + "loss": 1.7389, + "step": 923 + }, + { + "epoch": 1.0126027397260273, + "grad_norm": 2.705246925354004, + "learning_rate": 3.7144597887896024e-06, + "loss": 1.7408, + "step": 924 + }, + { + "epoch": 1.0136986301369864, + "grad_norm": 3.181328535079956, + "learning_rate": 3.7124289195775797e-06, + "loss": 1.7092, + "step": 925 + }, + { + "epoch": 1.0147945205479452, + "grad_norm": 3.0449445247650146, + "learning_rate": 3.710398050365557e-06, + "loss": 1.7399, + "step": 926 + }, + { + "epoch": 1.0158904109589042, + "grad_norm": 2.650517225265503, + "learning_rate": 3.708367181153534e-06, + "loss": 1.6861, + "step": 927 + }, + { + "epoch": 1.016986301369863, + "grad_norm": 2.7593672275543213, + "learning_rate": 3.7063363119415115e-06, + "loss": 1.6701, + "step": 928 + }, + { + "epoch": 1.0180821917808218, + "grad_norm": 3.0486466884613037, + "learning_rate": 3.7043054427294884e-06, + "loss": 1.7191, + "step": 929 + }, + { + "epoch": 1.0191780821917809, + "grad_norm": 2.502850294113159, + "learning_rate": 3.702274573517466e-06, + "loss": 1.6982, + "step": 930 + }, + { + "epoch": 1.0191780821917809, + "eval_loss": 1.9228183031082153, + "eval_runtime": 49.565, + "eval_samples_per_second": 20.176, + "eval_steps_per_second": 0.847, + "step": 930 + }, + { + "epoch": 1.0202739726027397, + "grad_norm": 2.400770664215088, + "learning_rate": 3.700243704305443e-06, + "loss": 1.6666, + "step": 931 + }, + { + "epoch": 1.0213698630136987, + "grad_norm": 2.5583295822143555, + "learning_rate": 3.6982128350934206e-06, + "loss": 1.7971, + "step": 932 + }, + { + "epoch": 1.0224657534246575, + "grad_norm": 2.519505739212036, + "learning_rate": 3.6961819658813975e-06, + "loss": 1.7076, + "step": 933 + }, + { + "epoch": 1.0235616438356165, + "grad_norm": 3.106309413909912, + "learning_rate": 3.694151096669375e-06, + "loss": 1.7279, + "step": 934 + }, + { + "epoch": 1.0246575342465754, + "grad_norm": 2.516906499862671, + "learning_rate": 3.692120227457352e-06, + "loss": 1.6265, + "step": 935 + }, + { + "epoch": 1.0257534246575342, + "grad_norm": 2.603166103363037, + "learning_rate": 3.6900893582453297e-06, + "loss": 1.7122, + "step": 936 + }, + { + "epoch": 1.0268493150684932, + "grad_norm": 2.842773199081421, + "learning_rate": 3.6880584890333066e-06, + "loss": 1.7312, + "step": 937 + }, + { + "epoch": 1.027945205479452, + "grad_norm": 2.8860342502593994, + "learning_rate": 3.6860276198212834e-06, + "loss": 1.7229, + "step": 938 + }, + { + "epoch": 1.029041095890411, + "grad_norm": 2.517254114151001, + "learning_rate": 3.683996750609261e-06, + "loss": 1.7342, + "step": 939 + }, + { + "epoch": 1.0301369863013699, + "grad_norm": 2.430025100708008, + "learning_rate": 3.681965881397238e-06, + "loss": 1.6817, + "step": 940 + }, + { + "epoch": 1.0301369863013699, + "eval_loss": 1.9156773090362549, + "eval_runtime": 44.6903, + "eval_samples_per_second": 22.376, + "eval_steps_per_second": 0.94, + "step": 940 + }, + { + "epoch": 1.0312328767123287, + "grad_norm": 2.340075969696045, + "learning_rate": 3.6799350121852157e-06, + "loss": 1.7044, + "step": 941 + }, + { + "epoch": 1.0323287671232877, + "grad_norm": 2.4758968353271484, + "learning_rate": 3.6779041429731925e-06, + "loss": 1.6935, + "step": 942 + }, + { + "epoch": 1.0334246575342465, + "grad_norm": 2.878903865814209, + "learning_rate": 3.6758732737611702e-06, + "loss": 1.7021, + "step": 943 + }, + { + "epoch": 1.0345205479452055, + "grad_norm": 2.388366222381592, + "learning_rate": 3.673842404549147e-06, + "loss": 1.7229, + "step": 944 + }, + { + "epoch": 1.0356164383561643, + "grad_norm": 2.5789167881011963, + "learning_rate": 3.671811535337125e-06, + "loss": 1.8112, + "step": 945 + }, + { + "epoch": 1.0367123287671234, + "grad_norm": 2.743072509765625, + "learning_rate": 3.6697806661251016e-06, + "loss": 1.7239, + "step": 946 + }, + { + "epoch": 1.0378082191780822, + "grad_norm": 2.5485000610351562, + "learning_rate": 3.6677497969130793e-06, + "loss": 1.7641, + "step": 947 + }, + { + "epoch": 1.038904109589041, + "grad_norm": 2.53190541267395, + "learning_rate": 3.665718927701056e-06, + "loss": 1.6996, + "step": 948 + }, + { + "epoch": 1.04, + "grad_norm": 2.3354969024658203, + "learning_rate": 3.663688058489034e-06, + "loss": 1.6623, + "step": 949 + }, + { + "epoch": 1.0410958904109588, + "grad_norm": 2.37371826171875, + "learning_rate": 3.6616571892770107e-06, + "loss": 1.7109, + "step": 950 + }, + { + "epoch": 1.0410958904109588, + "eval_loss": 1.9122159481048584, + "eval_runtime": 43.8844, + "eval_samples_per_second": 22.787, + "eval_steps_per_second": 0.957, + "step": 950 + }, + { + "epoch": 1.0421917808219179, + "grad_norm": 2.406137228012085, + "learning_rate": 3.659626320064988e-06, + "loss": 1.6958, + "step": 951 + }, + { + "epoch": 1.0432876712328767, + "grad_norm": 2.4220657348632812, + "learning_rate": 3.6575954508529653e-06, + "loss": 1.7196, + "step": 952 + }, + { + "epoch": 1.0443835616438357, + "grad_norm": 2.6203134059906006, + "learning_rate": 3.6555645816409426e-06, + "loss": 1.6014, + "step": 953 + }, + { + "epoch": 1.0454794520547945, + "grad_norm": 2.5388472080230713, + "learning_rate": 3.65353371242892e-06, + "loss": 1.7272, + "step": 954 + }, + { + "epoch": 1.0465753424657533, + "grad_norm": 2.672658681869507, + "learning_rate": 3.651502843216897e-06, + "loss": 1.6786, + "step": 955 + }, + { + "epoch": 1.0476712328767124, + "grad_norm": 2.4366531372070312, + "learning_rate": 3.6494719740048744e-06, + "loss": 1.7198, + "step": 956 + }, + { + "epoch": 1.0487671232876712, + "grad_norm": 2.4872887134552, + "learning_rate": 3.6474411047928517e-06, + "loss": 1.7082, + "step": 957 + }, + { + "epoch": 1.0498630136986302, + "grad_norm": 2.371206283569336, + "learning_rate": 3.645410235580829e-06, + "loss": 1.6964, + "step": 958 + }, + { + "epoch": 1.050958904109589, + "grad_norm": 2.496103286743164, + "learning_rate": 3.6433793663688062e-06, + "loss": 1.716, + "step": 959 + }, + { + "epoch": 1.0520547945205478, + "grad_norm": 2.658316135406494, + "learning_rate": 3.6413484971567835e-06, + "loss": 1.7263, + "step": 960 + }, + { + "epoch": 1.0520547945205478, + "eval_loss": 1.9115047454833984, + "eval_runtime": 43.8788, + "eval_samples_per_second": 22.79, + "eval_steps_per_second": 0.957, + "step": 960 + }, + { + "epoch": 1.0531506849315069, + "grad_norm": 2.5931882858276367, + "learning_rate": 3.6393176279447608e-06, + "loss": 1.7586, + "step": 961 + }, + { + "epoch": 1.0542465753424657, + "grad_norm": 2.4697139263153076, + "learning_rate": 3.637286758732738e-06, + "loss": 1.7232, + "step": 962 + }, + { + "epoch": 1.0553424657534247, + "grad_norm": 2.7274153232574463, + "learning_rate": 3.6352558895207153e-06, + "loss": 1.6917, + "step": 963 + }, + { + "epoch": 1.0564383561643835, + "grad_norm": 2.683138132095337, + "learning_rate": 3.633225020308692e-06, + "loss": 1.7518, + "step": 964 + }, + { + "epoch": 1.0575342465753426, + "grad_norm": 2.4510817527770996, + "learning_rate": 3.63119415109667e-06, + "loss": 1.6533, + "step": 965 + }, + { + "epoch": 1.0586301369863014, + "grad_norm": 2.6939656734466553, + "learning_rate": 3.6291632818846467e-06, + "loss": 1.6545, + "step": 966 + }, + { + "epoch": 1.0597260273972602, + "grad_norm": 2.572864532470703, + "learning_rate": 3.6271324126726244e-06, + "loss": 1.6965, + "step": 967 + }, + { + "epoch": 1.0608219178082192, + "grad_norm": 2.4816951751708984, + "learning_rate": 3.6251015434606013e-06, + "loss": 1.6851, + "step": 968 + }, + { + "epoch": 1.061917808219178, + "grad_norm": 2.640394926071167, + "learning_rate": 3.623070674248579e-06, + "loss": 1.7336, + "step": 969 + }, + { + "epoch": 1.063013698630137, + "grad_norm": 2.6707329750061035, + "learning_rate": 3.621039805036556e-06, + "loss": 1.7182, + "step": 970 + }, + { + "epoch": 1.063013698630137, + "eval_loss": 1.91274094581604, + "eval_runtime": 44.35, + "eval_samples_per_second": 22.548, + "eval_steps_per_second": 0.947, + "step": 970 + }, + { + "epoch": 1.0641095890410959, + "grad_norm": 2.6947290897369385, + "learning_rate": 3.6190089358245335e-06, + "loss": 1.7255, + "step": 971 + }, + { + "epoch": 1.0652054794520547, + "grad_norm": 2.5496068000793457, + "learning_rate": 3.6169780666125104e-06, + "loss": 1.6209, + "step": 972 + }, + { + "epoch": 1.0663013698630137, + "grad_norm": 2.6196374893188477, + "learning_rate": 3.614947197400488e-06, + "loss": 1.7277, + "step": 973 + }, + { + "epoch": 1.0673972602739725, + "grad_norm": 2.4534807205200195, + "learning_rate": 3.612916328188465e-06, + "loss": 1.7404, + "step": 974 + }, + { + "epoch": 1.0684931506849316, + "grad_norm": 2.5428082942962646, + "learning_rate": 3.610885458976442e-06, + "loss": 1.6618, + "step": 975 + }, + { + "epoch": 1.0695890410958904, + "grad_norm": 2.5431482791900635, + "learning_rate": 3.6088545897644195e-06, + "loss": 1.6743, + "step": 976 + }, + { + "epoch": 1.0706849315068494, + "grad_norm": 2.707956552505493, + "learning_rate": 3.6068237205523963e-06, + "loss": 1.7483, + "step": 977 + }, + { + "epoch": 1.0717808219178082, + "grad_norm": 2.5643982887268066, + "learning_rate": 3.604792851340374e-06, + "loss": 1.7304, + "step": 978 + }, + { + "epoch": 1.072876712328767, + "grad_norm": 2.4612927436828613, + "learning_rate": 3.602761982128351e-06, + "loss": 1.6561, + "step": 979 + }, + { + "epoch": 1.073972602739726, + "grad_norm": 2.3140358924865723, + "learning_rate": 3.6007311129163286e-06, + "loss": 1.6834, + "step": 980 + }, + { + "epoch": 1.073972602739726, + "eval_loss": 1.9124692678451538, + "eval_runtime": 44.0522, + "eval_samples_per_second": 22.7, + "eval_steps_per_second": 0.953, + "step": 980 + }, + { + "epoch": 1.0750684931506849, + "grad_norm": 2.4636526107788086, + "learning_rate": 3.5987002437043054e-06, + "loss": 1.7395, + "step": 981 + }, + { + "epoch": 1.076164383561644, + "grad_norm": 2.541180372238159, + "learning_rate": 3.596669374492283e-06, + "loss": 1.6579, + "step": 982 + }, + { + "epoch": 1.0772602739726027, + "grad_norm": 2.621335983276367, + "learning_rate": 3.59463850528026e-06, + "loss": 1.7332, + "step": 983 + }, + { + "epoch": 1.0783561643835617, + "grad_norm": 2.363450050354004, + "learning_rate": 3.5926076360682377e-06, + "loss": 1.7047, + "step": 984 + }, + { + "epoch": 1.0794520547945206, + "grad_norm": 2.8609447479248047, + "learning_rate": 3.5905767668562145e-06, + "loss": 1.7416, + "step": 985 + }, + { + "epoch": 1.0805479452054794, + "grad_norm": 2.457338571548462, + "learning_rate": 3.5885458976441922e-06, + "loss": 1.7516, + "step": 986 + }, + { + "epoch": 1.0816438356164384, + "grad_norm": 2.3943030834198, + "learning_rate": 3.586515028432169e-06, + "loss": 1.6436, + "step": 987 + }, + { + "epoch": 1.0827397260273972, + "grad_norm": 2.4719340801239014, + "learning_rate": 3.5844841592201464e-06, + "loss": 1.6988, + "step": 988 + }, + { + "epoch": 1.0838356164383562, + "grad_norm": 2.3442330360412598, + "learning_rate": 3.5824532900081237e-06, + "loss": 1.7165, + "step": 989 + }, + { + "epoch": 1.084931506849315, + "grad_norm": 2.3391900062561035, + "learning_rate": 3.580422420796101e-06, + "loss": 1.6263, + "step": 990 + }, + { + "epoch": 1.084931506849315, + "eval_loss": 1.9100929498672485, + "eval_runtime": 44.2504, + "eval_samples_per_second": 22.599, + "eval_steps_per_second": 0.949, + "step": 990 + }, + { + "epoch": 1.086027397260274, + "grad_norm": 2.533566474914551, + "learning_rate": 3.578391551584078e-06, + "loss": 1.7141, + "step": 991 + }, + { + "epoch": 1.087123287671233, + "grad_norm": 2.464540481567383, + "learning_rate": 3.5763606823720555e-06, + "loss": 1.7335, + "step": 992 + }, + { + "epoch": 1.0882191780821917, + "grad_norm": 2.4789979457855225, + "learning_rate": 3.5743298131600328e-06, + "loss": 1.7074, + "step": 993 + }, + { + "epoch": 1.0893150684931507, + "grad_norm": 2.8252034187316895, + "learning_rate": 3.57229894394801e-06, + "loss": 1.6808, + "step": 994 + }, + { + "epoch": 1.0904109589041096, + "grad_norm": 2.5870537757873535, + "learning_rate": 3.5702680747359873e-06, + "loss": 1.7171, + "step": 995 + }, + { + "epoch": 1.0915068493150686, + "grad_norm": 2.349393129348755, + "learning_rate": 3.5682372055239646e-06, + "loss": 1.6687, + "step": 996 + }, + { + "epoch": 1.0926027397260274, + "grad_norm": 2.3343207836151123, + "learning_rate": 3.566206336311942e-06, + "loss": 1.6351, + "step": 997 + }, + { + "epoch": 1.0936986301369862, + "grad_norm": 2.558645248413086, + "learning_rate": 3.564175467099919e-06, + "loss": 1.708, + "step": 998 + }, + { + "epoch": 1.0947945205479452, + "grad_norm": 2.7598931789398193, + "learning_rate": 3.5621445978878964e-06, + "loss": 1.6797, + "step": 999 + }, + { + "epoch": 1.095890410958904, + "grad_norm": 2.6048543453216553, + "learning_rate": 3.5601137286758737e-06, + "loss": 1.7117, + "step": 1000 + }, + { + "epoch": 1.095890410958904, + "eval_loss": 1.906927466392517, + "eval_runtime": 44.5189, + "eval_samples_per_second": 22.462, + "eval_steps_per_second": 0.943, + "step": 1000 + }, + { + "epoch": 1.096986301369863, + "grad_norm": 2.6124815940856934, + "learning_rate": 3.5580828594638505e-06, + "loss": 1.7328, + "step": 1001 + }, + { + "epoch": 1.098082191780822, + "grad_norm": 2.3410215377807617, + "learning_rate": 3.5560519902518282e-06, + "loss": 1.6932, + "step": 1002 + }, + { + "epoch": 1.099178082191781, + "grad_norm": 2.5247764587402344, + "learning_rate": 3.554021121039805e-06, + "loss": 1.6628, + "step": 1003 + }, + { + "epoch": 1.1002739726027397, + "grad_norm": 2.642958164215088, + "learning_rate": 3.551990251827783e-06, + "loss": 1.7035, + "step": 1004 + }, + { + "epoch": 1.1013698630136985, + "grad_norm": 2.3978517055511475, + "learning_rate": 3.5499593826157596e-06, + "loss": 1.6796, + "step": 1005 + }, + { + "epoch": 1.1024657534246576, + "grad_norm": 2.3969287872314453, + "learning_rate": 3.5479285134037373e-06, + "loss": 1.6626, + "step": 1006 + }, + { + "epoch": 1.1035616438356164, + "grad_norm": 2.5723509788513184, + "learning_rate": 3.545897644191714e-06, + "loss": 1.7135, + "step": 1007 + }, + { + "epoch": 1.1046575342465754, + "grad_norm": 2.545154571533203, + "learning_rate": 3.543866774979692e-06, + "loss": 1.6626, + "step": 1008 + }, + { + "epoch": 1.1057534246575342, + "grad_norm": 2.501058340072632, + "learning_rate": 3.5418359057676687e-06, + "loss": 1.7112, + "step": 1009 + }, + { + "epoch": 1.106849315068493, + "grad_norm": 2.695483684539795, + "learning_rate": 3.5398050365556464e-06, + "loss": 1.6773, + "step": 1010 + }, + { + "epoch": 1.106849315068493, + "eval_loss": 1.9063464403152466, + "eval_runtime": 44.2034, + "eval_samples_per_second": 22.623, + "eval_steps_per_second": 0.95, + "step": 1010 + }, + { + "epoch": 1.107945205479452, + "grad_norm": 3.0584375858306885, + "learning_rate": 3.5377741673436233e-06, + "loss": 1.7242, + "step": 1011 + }, + { + "epoch": 1.1090410958904109, + "grad_norm": 2.445220947265625, + "learning_rate": 3.535743298131601e-06, + "loss": 1.7349, + "step": 1012 + }, + { + "epoch": 1.11013698630137, + "grad_norm": 2.427220106124878, + "learning_rate": 3.533712428919578e-06, + "loss": 1.677, + "step": 1013 + }, + { + "epoch": 1.1112328767123287, + "grad_norm": 2.5148212909698486, + "learning_rate": 3.5316815597075547e-06, + "loss": 1.7007, + "step": 1014 + }, + { + "epoch": 1.1123287671232878, + "grad_norm": 2.644319534301758, + "learning_rate": 3.5296506904955324e-06, + "loss": 1.7442, + "step": 1015 + }, + { + "epoch": 1.1134246575342466, + "grad_norm": 2.6118004322052, + "learning_rate": 3.5276198212835093e-06, + "loss": 1.7509, + "step": 1016 + }, + { + "epoch": 1.1145205479452054, + "grad_norm": 2.5224156379699707, + "learning_rate": 3.525588952071487e-06, + "loss": 1.7279, + "step": 1017 + }, + { + "epoch": 1.1156164383561644, + "grad_norm": 2.4668405055999756, + "learning_rate": 3.523558082859464e-06, + "loss": 1.6754, + "step": 1018 + }, + { + "epoch": 1.1167123287671232, + "grad_norm": 3.8330910205841064, + "learning_rate": 3.5215272136474415e-06, + "loss": 1.6815, + "step": 1019 + }, + { + "epoch": 1.1178082191780823, + "grad_norm": 2.569058895111084, + "learning_rate": 3.5194963444354184e-06, + "loss": 1.6518, + "step": 1020 + }, + { + "epoch": 1.1178082191780823, + "eval_loss": 1.906802773475647, + "eval_runtime": 50.8521, + "eval_samples_per_second": 19.665, + "eval_steps_per_second": 0.826, + "step": 1020 + }, + { + "epoch": 1.118904109589041, + "grad_norm": 2.705265522003174, + "learning_rate": 3.517465475223396e-06, + "loss": 1.7172, + "step": 1021 + }, + { + "epoch": 1.12, + "grad_norm": 3.2667057514190674, + "learning_rate": 3.515434606011373e-06, + "loss": 1.7641, + "step": 1022 + }, + { + "epoch": 1.121095890410959, + "grad_norm": 3.5119211673736572, + "learning_rate": 3.5134037367993506e-06, + "loss": 1.7178, + "step": 1023 + }, + { + "epoch": 1.1221917808219177, + "grad_norm": 2.7348034381866455, + "learning_rate": 3.5113728675873275e-06, + "loss": 1.6815, + "step": 1024 + }, + { + "epoch": 1.1232876712328768, + "grad_norm": 3.019892454147339, + "learning_rate": 3.5093419983753047e-06, + "loss": 1.6417, + "step": 1025 + }, + { + "epoch": 1.1243835616438356, + "grad_norm": 3.2486684322357178, + "learning_rate": 3.507311129163282e-06, + "loss": 1.6888, + "step": 1026 + }, + { + "epoch": 1.1254794520547946, + "grad_norm": 2.897071361541748, + "learning_rate": 3.5052802599512593e-06, + "loss": 1.7121, + "step": 1027 + }, + { + "epoch": 1.1265753424657534, + "grad_norm": 2.7342352867126465, + "learning_rate": 3.5032493907392366e-06, + "loss": 1.6339, + "step": 1028 + }, + { + "epoch": 1.1276712328767124, + "grad_norm": 2.691178560256958, + "learning_rate": 3.501218521527214e-06, + "loss": 1.7069, + "step": 1029 + }, + { + "epoch": 1.1287671232876713, + "grad_norm": 2.548682689666748, + "learning_rate": 3.499187652315191e-06, + "loss": 1.7218, + "step": 1030 + }, + { + "epoch": 1.1287671232876713, + "eval_loss": 1.9029825925827026, + "eval_runtime": 48.287, + "eval_samples_per_second": 20.71, + "eval_steps_per_second": 0.87, + "step": 1030 + }, + { + "epoch": 1.12986301369863, + "grad_norm": 2.422056198120117, + "learning_rate": 3.4971567831031684e-06, + "loss": 1.6486, + "step": 1031 + }, + { + "epoch": 1.130958904109589, + "grad_norm": 2.4367074966430664, + "learning_rate": 3.4951259138911457e-06, + "loss": 1.759, + "step": 1032 + }, + { + "epoch": 1.132054794520548, + "grad_norm": 2.48002290725708, + "learning_rate": 3.493095044679123e-06, + "loss": 1.5902, + "step": 1033 + }, + { + "epoch": 1.133150684931507, + "grad_norm": 2.5030295848846436, + "learning_rate": 3.4910641754671002e-06, + "loss": 1.7108, + "step": 1034 + }, + { + "epoch": 1.1342465753424658, + "grad_norm": 2.5892179012298584, + "learning_rate": 3.4890333062550775e-06, + "loss": 1.6154, + "step": 1035 + }, + { + "epoch": 1.1353424657534246, + "grad_norm": 2.4282920360565186, + "learning_rate": 3.4870024370430548e-06, + "loss": 1.6807, + "step": 1036 + }, + { + "epoch": 1.1364383561643836, + "grad_norm": 2.4826700687408447, + "learning_rate": 3.484971567831032e-06, + "loss": 1.7284, + "step": 1037 + }, + { + "epoch": 1.1375342465753424, + "grad_norm": 2.416597843170166, + "learning_rate": 3.482940698619009e-06, + "loss": 1.6972, + "step": 1038 + }, + { + "epoch": 1.1386301369863014, + "grad_norm": 2.4664933681488037, + "learning_rate": 3.4809098294069866e-06, + "loss": 1.7119, + "step": 1039 + }, + { + "epoch": 1.1397260273972603, + "grad_norm": 2.541856288909912, + "learning_rate": 3.4788789601949634e-06, + "loss": 1.6752, + "step": 1040 + }, + { + "epoch": 1.1397260273972603, + "eval_loss": 1.8999215364456177, + "eval_runtime": 46.3528, + "eval_samples_per_second": 21.574, + "eval_steps_per_second": 0.906, + "step": 1040 + }, + { + "epoch": 1.140821917808219, + "grad_norm": 2.4803109169006348, + "learning_rate": 3.476848090982941e-06, + "loss": 1.776, + "step": 1041 + }, + { + "epoch": 1.141917808219178, + "grad_norm": 2.4363205432891846, + "learning_rate": 3.474817221770918e-06, + "loss": 1.7196, + "step": 1042 + }, + { + "epoch": 1.143013698630137, + "grad_norm": 2.4377009868621826, + "learning_rate": 3.4727863525588957e-06, + "loss": 1.6937, + "step": 1043 + }, + { + "epoch": 1.144109589041096, + "grad_norm": 2.425791025161743, + "learning_rate": 3.4707554833468725e-06, + "loss": 1.6547, + "step": 1044 + }, + { + "epoch": 1.1452054794520548, + "grad_norm": 2.3216712474823, + "learning_rate": 3.4687246141348502e-06, + "loss": 1.6789, + "step": 1045 + }, + { + "epoch": 1.1463013698630138, + "grad_norm": 2.65527081489563, + "learning_rate": 3.466693744922827e-06, + "loss": 1.7231, + "step": 1046 + }, + { + "epoch": 1.1473972602739726, + "grad_norm": 2.3415145874023438, + "learning_rate": 3.464662875710805e-06, + "loss": 1.6805, + "step": 1047 + }, + { + "epoch": 1.1484931506849314, + "grad_norm": 2.3864822387695312, + "learning_rate": 3.4626320064987817e-06, + "loss": 1.7153, + "step": 1048 + }, + { + "epoch": 1.1495890410958904, + "grad_norm": 2.490896224975586, + "learning_rate": 3.4606011372867594e-06, + "loss": 1.6866, + "step": 1049 + }, + { + "epoch": 1.1506849315068493, + "grad_norm": 2.339711904525757, + "learning_rate": 3.458570268074736e-06, + "loss": 1.6732, + "step": 1050 + }, + { + "epoch": 1.1506849315068493, + "eval_loss": 1.8985264301300049, + "eval_runtime": 46.7854, + "eval_samples_per_second": 21.374, + "eval_steps_per_second": 0.898, + "step": 1050 + }, + { + "epoch": 1.1517808219178083, + "grad_norm": 2.393709659576416, + "learning_rate": 3.4565393988627135e-06, + "loss": 1.6549, + "step": 1051 + }, + { + "epoch": 1.152876712328767, + "grad_norm": 2.564114809036255, + "learning_rate": 3.4545085296506908e-06, + "loss": 1.7019, + "step": 1052 + }, + { + "epoch": 1.1539726027397261, + "grad_norm": 2.3826403617858887, + "learning_rate": 3.452477660438668e-06, + "loss": 1.6427, + "step": 1053 + }, + { + "epoch": 1.155068493150685, + "grad_norm": 2.3185465335845947, + "learning_rate": 3.4504467912266453e-06, + "loss": 1.6916, + "step": 1054 + }, + { + "epoch": 1.1561643835616437, + "grad_norm": 2.4503891468048096, + "learning_rate": 3.4484159220146226e-06, + "loss": 1.6793, + "step": 1055 + }, + { + "epoch": 1.1572602739726028, + "grad_norm": 2.386446714401245, + "learning_rate": 3.4463850528026e-06, + "loss": 1.7186, + "step": 1056 + }, + { + "epoch": 1.1583561643835616, + "grad_norm": 2.6404714584350586, + "learning_rate": 3.444354183590577e-06, + "loss": 1.6605, + "step": 1057 + }, + { + "epoch": 1.1594520547945206, + "grad_norm": 2.342374324798584, + "learning_rate": 3.4423233143785544e-06, + "loss": 1.6219, + "step": 1058 + }, + { + "epoch": 1.1605479452054794, + "grad_norm": 2.7503273487091064, + "learning_rate": 3.4402924451665317e-06, + "loss": 1.693, + "step": 1059 + }, + { + "epoch": 1.1616438356164385, + "grad_norm": 2.5416243076324463, + "learning_rate": 3.438261575954509e-06, + "loss": 1.7323, + "step": 1060 + }, + { + "epoch": 1.1616438356164385, + "eval_loss": 1.9016659259796143, + "eval_runtime": 48.5456, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1060 + }, + { + "epoch": 1.1627397260273973, + "grad_norm": 2.5787603855133057, + "learning_rate": 3.4362307067424862e-06, + "loss": 1.7057, + "step": 1061 + }, + { + "epoch": 1.163835616438356, + "grad_norm": 2.649528980255127, + "learning_rate": 3.434199837530463e-06, + "loss": 1.6348, + "step": 1062 + }, + { + "epoch": 1.1649315068493151, + "grad_norm": 2.504089832305908, + "learning_rate": 3.432168968318441e-06, + "loss": 1.7202, + "step": 1063 + }, + { + "epoch": 1.166027397260274, + "grad_norm": 2.5624678134918213, + "learning_rate": 3.4301380991064176e-06, + "loss": 1.7351, + "step": 1064 + }, + { + "epoch": 1.167123287671233, + "grad_norm": 2.641331911087036, + "learning_rate": 3.4281072298943953e-06, + "loss": 1.7453, + "step": 1065 + }, + { + "epoch": 1.1682191780821918, + "grad_norm": 2.599500894546509, + "learning_rate": 3.426076360682372e-06, + "loss": 1.6942, + "step": 1066 + }, + { + "epoch": 1.1693150684931506, + "grad_norm": 2.668599843978882, + "learning_rate": 3.42404549147035e-06, + "loss": 1.6337, + "step": 1067 + }, + { + "epoch": 1.1704109589041096, + "grad_norm": 2.75667667388916, + "learning_rate": 3.4220146222583267e-06, + "loss": 1.6988, + "step": 1068 + }, + { + "epoch": 1.1715068493150684, + "grad_norm": 2.641526937484741, + "learning_rate": 3.4199837530463044e-06, + "loss": 1.7641, + "step": 1069 + }, + { + "epoch": 1.1726027397260275, + "grad_norm": 2.4972548484802246, + "learning_rate": 3.4179528838342813e-06, + "loss": 1.7269, + "step": 1070 + }, + { + "epoch": 1.1726027397260275, + "eval_loss": 1.8979382514953613, + "eval_runtime": 45.6555, + "eval_samples_per_second": 21.903, + "eval_steps_per_second": 0.92, + "step": 1070 + }, + { + "epoch": 1.1736986301369863, + "grad_norm": 2.5454585552215576, + "learning_rate": 3.415922014622259e-06, + "loss": 1.7294, + "step": 1071 + }, + { + "epoch": 1.174794520547945, + "grad_norm": 2.5921928882598877, + "learning_rate": 3.413891145410236e-06, + "loss": 1.7145, + "step": 1072 + }, + { + "epoch": 1.1758904109589041, + "grad_norm": 2.4028453826904297, + "learning_rate": 3.4118602761982135e-06, + "loss": 1.6985, + "step": 1073 + }, + { + "epoch": 1.176986301369863, + "grad_norm": 2.604522705078125, + "learning_rate": 3.4098294069861904e-06, + "loss": 1.6586, + "step": 1074 + }, + { + "epoch": 1.178082191780822, + "grad_norm": 2.528970241546631, + "learning_rate": 3.4077985377741673e-06, + "loss": 1.734, + "step": 1075 + }, + { + "epoch": 1.1791780821917808, + "grad_norm": 2.565030574798584, + "learning_rate": 3.405767668562145e-06, + "loss": 1.6997, + "step": 1076 + }, + { + "epoch": 1.1802739726027398, + "grad_norm": 2.7735602855682373, + "learning_rate": 3.403736799350122e-06, + "loss": 1.6543, + "step": 1077 + }, + { + "epoch": 1.1813698630136986, + "grad_norm": 2.5116474628448486, + "learning_rate": 3.4017059301380995e-06, + "loss": 1.7711, + "step": 1078 + }, + { + "epoch": 1.1824657534246574, + "grad_norm": 2.4827239513397217, + "learning_rate": 3.3996750609260764e-06, + "loss": 1.6831, + "step": 1079 + }, + { + "epoch": 1.1835616438356165, + "grad_norm": 2.544309377670288, + "learning_rate": 3.397644191714054e-06, + "loss": 1.6905, + "step": 1080 + }, + { + "epoch": 1.1835616438356165, + "eval_loss": 1.8945367336273193, + "eval_runtime": 46.6027, + "eval_samples_per_second": 21.458, + "eval_steps_per_second": 0.901, + "step": 1080 + }, + { + "epoch": 1.1846575342465753, + "grad_norm": 2.456636905670166, + "learning_rate": 3.395613322502031e-06, + "loss": 1.646, + "step": 1081 + }, + { + "epoch": 1.1857534246575343, + "grad_norm": 2.5900352001190186, + "learning_rate": 3.3935824532900086e-06, + "loss": 1.6967, + "step": 1082 + }, + { + "epoch": 1.1868493150684931, + "grad_norm": 3.3344573974609375, + "learning_rate": 3.3915515840779855e-06, + "loss": 1.6947, + "step": 1083 + }, + { + "epoch": 1.1879452054794521, + "grad_norm": 2.5621678829193115, + "learning_rate": 3.389520714865963e-06, + "loss": 1.7084, + "step": 1084 + }, + { + "epoch": 1.189041095890411, + "grad_norm": 2.5676684379577637, + "learning_rate": 3.38748984565394e-06, + "loss": 1.6788, + "step": 1085 + }, + { + "epoch": 1.1901369863013698, + "grad_norm": 2.600691318511963, + "learning_rate": 3.3854589764419177e-06, + "loss": 1.6811, + "step": 1086 + }, + { + "epoch": 1.1912328767123288, + "grad_norm": 2.5324196815490723, + "learning_rate": 3.3834281072298946e-06, + "loss": 1.661, + "step": 1087 + }, + { + "epoch": 1.1923287671232876, + "grad_norm": 2.7559397220611572, + "learning_rate": 3.381397238017872e-06, + "loss": 1.7473, + "step": 1088 + }, + { + "epoch": 1.1934246575342466, + "grad_norm": 2.526397466659546, + "learning_rate": 3.379366368805849e-06, + "loss": 1.7, + "step": 1089 + }, + { + "epoch": 1.1945205479452055, + "grad_norm": 2.3752379417419434, + "learning_rate": 3.3773354995938264e-06, + "loss": 1.689, + "step": 1090 + }, + { + "epoch": 1.1945205479452055, + "eval_loss": 1.8995447158813477, + "eval_runtime": 48.0083, + "eval_samples_per_second": 20.83, + "eval_steps_per_second": 0.875, + "step": 1090 + }, + { + "epoch": 1.1956164383561645, + "grad_norm": 2.5902793407440186, + "learning_rate": 3.3753046303818037e-06, + "loss": 1.6899, + "step": 1091 + }, + { + "epoch": 1.1967123287671233, + "grad_norm": 3.157573938369751, + "learning_rate": 3.373273761169781e-06, + "loss": 1.6792, + "step": 1092 + }, + { + "epoch": 1.1978082191780821, + "grad_norm": 2.628035545349121, + "learning_rate": 3.3712428919577582e-06, + "loss": 1.7003, + "step": 1093 + }, + { + "epoch": 1.1989041095890411, + "grad_norm": 2.4810822010040283, + "learning_rate": 3.3692120227457355e-06, + "loss": 1.6489, + "step": 1094 + }, + { + "epoch": 1.2, + "grad_norm": 2.4859938621520996, + "learning_rate": 3.3671811535337128e-06, + "loss": 1.6768, + "step": 1095 + }, + { + "epoch": 1.201095890410959, + "grad_norm": 2.6773345470428467, + "learning_rate": 3.36515028432169e-06, + "loss": 1.7046, + "step": 1096 + }, + { + "epoch": 1.2021917808219178, + "grad_norm": 2.9695260524749756, + "learning_rate": 3.3631194151096673e-06, + "loss": 1.7341, + "step": 1097 + }, + { + "epoch": 1.2032876712328768, + "grad_norm": 2.8146679401397705, + "learning_rate": 3.3610885458976446e-06, + "loss": 1.7472, + "step": 1098 + }, + { + "epoch": 1.2043835616438356, + "grad_norm": 2.534985065460205, + "learning_rate": 3.359057676685622e-06, + "loss": 1.6199, + "step": 1099 + }, + { + "epoch": 1.2054794520547945, + "grad_norm": 2.9351718425750732, + "learning_rate": 3.357026807473599e-06, + "loss": 1.6656, + "step": 1100 + }, + { + "epoch": 1.2054794520547945, + "eval_loss": 1.8962808847427368, + "eval_runtime": 47.0443, + "eval_samples_per_second": 21.257, + "eval_steps_per_second": 0.893, + "step": 1100 + }, + { + "epoch": 1.2065753424657535, + "grad_norm": 2.4404256343841553, + "learning_rate": 3.354995938261576e-06, + "loss": 1.6207, + "step": 1101 + }, + { + "epoch": 1.2076712328767123, + "grad_norm": 2.6691436767578125, + "learning_rate": 3.3529650690495537e-06, + "loss": 1.6627, + "step": 1102 + }, + { + "epoch": 1.208767123287671, + "grad_norm": 2.5436172485351562, + "learning_rate": 3.3509341998375305e-06, + "loss": 1.6919, + "step": 1103 + }, + { + "epoch": 1.2098630136986301, + "grad_norm": 2.7167909145355225, + "learning_rate": 3.3489033306255082e-06, + "loss": 1.6434, + "step": 1104 + }, + { + "epoch": 1.210958904109589, + "grad_norm": 2.575446367263794, + "learning_rate": 3.346872461413485e-06, + "loss": 1.6554, + "step": 1105 + }, + { + "epoch": 1.212054794520548, + "grad_norm": 2.3874504566192627, + "learning_rate": 3.344841592201463e-06, + "loss": 1.6691, + "step": 1106 + }, + { + "epoch": 1.2131506849315068, + "grad_norm": 2.66317081451416, + "learning_rate": 3.3428107229894397e-06, + "loss": 1.6864, + "step": 1107 + }, + { + "epoch": 1.2142465753424658, + "grad_norm": 2.4396398067474365, + "learning_rate": 3.3407798537774174e-06, + "loss": 1.7217, + "step": 1108 + }, + { + "epoch": 1.2153424657534246, + "grad_norm": 2.5002803802490234, + "learning_rate": 3.338748984565394e-06, + "loss": 1.7556, + "step": 1109 + }, + { + "epoch": 1.2164383561643834, + "grad_norm": 2.679577350616455, + "learning_rate": 3.336718115353372e-06, + "loss": 1.6822, + "step": 1110 + }, + { + "epoch": 1.2164383561643834, + "eval_loss": 1.8942477703094482, + "eval_runtime": 44.3804, + "eval_samples_per_second": 22.532, + "eval_steps_per_second": 0.946, + "step": 1110 + }, + { + "epoch": 1.2175342465753425, + "grad_norm": 2.6119370460510254, + "learning_rate": 3.3346872461413488e-06, + "loss": 1.662, + "step": 1111 + }, + { + "epoch": 1.2186301369863013, + "grad_norm": 2.6024818420410156, + "learning_rate": 3.3326563769293256e-06, + "loss": 1.687, + "step": 1112 + }, + { + "epoch": 1.2197260273972603, + "grad_norm": 4.128854751586914, + "learning_rate": 3.3306255077173033e-06, + "loss": 1.6898, + "step": 1113 + }, + { + "epoch": 1.2208219178082191, + "grad_norm": 2.3910412788391113, + "learning_rate": 3.32859463850528e-06, + "loss": 1.6274, + "step": 1114 + }, + { + "epoch": 1.2219178082191782, + "grad_norm": 2.8630096912384033, + "learning_rate": 3.326563769293258e-06, + "loss": 1.6554, + "step": 1115 + }, + { + "epoch": 1.223013698630137, + "grad_norm": 4.201050281524658, + "learning_rate": 3.3245329000812347e-06, + "loss": 1.6676, + "step": 1116 + }, + { + "epoch": 1.2241095890410958, + "grad_norm": 2.492168664932251, + "learning_rate": 3.3225020308692124e-06, + "loss": 1.6319, + "step": 1117 + }, + { + "epoch": 1.2252054794520548, + "grad_norm": 3.6807360649108887, + "learning_rate": 3.3204711616571893e-06, + "loss": 1.6776, + "step": 1118 + }, + { + "epoch": 1.2263013698630136, + "grad_norm": 2.7271926403045654, + "learning_rate": 3.318440292445167e-06, + "loss": 1.6383, + "step": 1119 + }, + { + "epoch": 1.2273972602739727, + "grad_norm": 2.5567307472229004, + "learning_rate": 3.316409423233144e-06, + "loss": 1.7325, + "step": 1120 + }, + { + "epoch": 1.2273972602739727, + "eval_loss": 1.892768383026123, + "eval_runtime": 50.909, + "eval_samples_per_second": 19.643, + "eval_steps_per_second": 0.825, + "step": 1120 + }, + { + "epoch": 1.2284931506849315, + "grad_norm": 2.5373101234436035, + "learning_rate": 3.3143785540211215e-06, + "loss": 1.689, + "step": 1121 + }, + { + "epoch": 1.2295890410958905, + "grad_norm": 2.602923631668091, + "learning_rate": 3.3123476848090984e-06, + "loss": 1.6861, + "step": 1122 + }, + { + "epoch": 1.2306849315068493, + "grad_norm": 2.507110118865967, + "learning_rate": 3.310316815597076e-06, + "loss": 1.7575, + "step": 1123 + }, + { + "epoch": 1.2317808219178081, + "grad_norm": 2.361786127090454, + "learning_rate": 3.308285946385053e-06, + "loss": 1.6267, + "step": 1124 + }, + { + "epoch": 1.2328767123287672, + "grad_norm": 2.5360615253448486, + "learning_rate": 3.30625507717303e-06, + "loss": 1.6911, + "step": 1125 + }, + { + "epoch": 1.233972602739726, + "grad_norm": 2.6238744258880615, + "learning_rate": 3.3042242079610075e-06, + "loss": 1.7193, + "step": 1126 + }, + { + "epoch": 1.235068493150685, + "grad_norm": 2.5358500480651855, + "learning_rate": 3.3021933387489847e-06, + "loss": 1.6691, + "step": 1127 + }, + { + "epoch": 1.2361643835616438, + "grad_norm": 2.7285404205322266, + "learning_rate": 3.300162469536962e-06, + "loss": 1.7249, + "step": 1128 + }, + { + "epoch": 1.2372602739726029, + "grad_norm": 2.346160650253296, + "learning_rate": 3.2981316003249393e-06, + "loss": 1.6958, + "step": 1129 + }, + { + "epoch": 1.2383561643835617, + "grad_norm": 2.5415399074554443, + "learning_rate": 3.2961007311129166e-06, + "loss": 1.7544, + "step": 1130 + }, + { + "epoch": 1.2383561643835617, + "eval_loss": 1.8925012350082397, + "eval_runtime": 46.4957, + "eval_samples_per_second": 21.507, + "eval_steps_per_second": 0.903, + "step": 1130 + }, + { + "epoch": 1.2394520547945205, + "grad_norm": 2.5924015045166016, + "learning_rate": 3.294069861900894e-06, + "loss": 1.6901, + "step": 1131 + }, + { + "epoch": 1.2405479452054795, + "grad_norm": 2.4649693965911865, + "learning_rate": 3.292038992688871e-06, + "loss": 1.7234, + "step": 1132 + }, + { + "epoch": 1.2416438356164383, + "grad_norm": 2.3648035526275635, + "learning_rate": 3.2900081234768484e-06, + "loss": 1.7324, + "step": 1133 + }, + { + "epoch": 1.2427397260273974, + "grad_norm": 2.5146145820617676, + "learning_rate": 3.2879772542648257e-06, + "loss": 1.631, + "step": 1134 + }, + { + "epoch": 1.2438356164383562, + "grad_norm": 2.5919511318206787, + "learning_rate": 3.285946385052803e-06, + "loss": 1.706, + "step": 1135 + }, + { + "epoch": 1.244931506849315, + "grad_norm": 2.299527168273926, + "learning_rate": 3.2839155158407802e-06, + "loss": 1.608, + "step": 1136 + }, + { + "epoch": 1.246027397260274, + "grad_norm": 2.480637788772583, + "learning_rate": 3.2818846466287575e-06, + "loss": 1.7415, + "step": 1137 + }, + { + "epoch": 1.2471232876712328, + "grad_norm": 2.458183526992798, + "learning_rate": 3.2798537774167344e-06, + "loss": 1.6806, + "step": 1138 + }, + { + "epoch": 1.2482191780821918, + "grad_norm": 2.4864983558654785, + "learning_rate": 3.277822908204712e-06, + "loss": 1.6373, + "step": 1139 + }, + { + "epoch": 1.2493150684931507, + "grad_norm": 2.3432068824768066, + "learning_rate": 3.275792038992689e-06, + "loss": 1.6832, + "step": 1140 + }, + { + "epoch": 1.2493150684931507, + "eval_loss": 1.8899296522140503, + "eval_runtime": 44.3309, + "eval_samples_per_second": 22.558, + "eval_steps_per_second": 0.947, + "step": 1140 + }, + { + "epoch": 1.2504109589041095, + "grad_norm": 2.365891218185425, + "learning_rate": 3.2737611697806666e-06, + "loss": 1.6492, + "step": 1141 + }, + { + "epoch": 1.2515068493150685, + "grad_norm": 2.463261842727661, + "learning_rate": 3.2717303005686435e-06, + "loss": 1.6847, + "step": 1142 + }, + { + "epoch": 1.2526027397260273, + "grad_norm": 2.3977205753326416, + "learning_rate": 3.269699431356621e-06, + "loss": 1.6789, + "step": 1143 + }, + { + "epoch": 1.2536986301369863, + "grad_norm": 2.4172558784484863, + "learning_rate": 3.267668562144598e-06, + "loss": 1.6495, + "step": 1144 + }, + { + "epoch": 1.2547945205479452, + "grad_norm": 2.663381814956665, + "learning_rate": 3.2656376929325757e-06, + "loss": 1.7139, + "step": 1145 + }, + { + "epoch": 1.2558904109589042, + "grad_norm": 2.3702306747436523, + "learning_rate": 3.2636068237205526e-06, + "loss": 1.6876, + "step": 1146 + }, + { + "epoch": 1.256986301369863, + "grad_norm": 3.1759073734283447, + "learning_rate": 3.2615759545085303e-06, + "loss": 1.7465, + "step": 1147 + }, + { + "epoch": 1.2580821917808218, + "grad_norm": 2.6344592571258545, + "learning_rate": 3.259545085296507e-06, + "loss": 1.6339, + "step": 1148 + }, + { + "epoch": 1.2591780821917808, + "grad_norm": 2.364490032196045, + "learning_rate": 3.257514216084484e-06, + "loss": 1.7394, + "step": 1149 + }, + { + "epoch": 1.2602739726027397, + "grad_norm": 2.5639748573303223, + "learning_rate": 3.2554833468724617e-06, + "loss": 1.6928, + "step": 1150 + }, + { + "epoch": 1.2602739726027397, + "eval_loss": 1.8937300443649292, + "eval_runtime": 43.9356, + "eval_samples_per_second": 22.761, + "eval_steps_per_second": 0.956, + "step": 1150 + }, + { + "epoch": 1.2613698630136987, + "grad_norm": 3.152968645095825, + "learning_rate": 3.2534524776604385e-06, + "loss": 1.6693, + "step": 1151 + }, + { + "epoch": 1.2624657534246575, + "grad_norm": 2.45776104927063, + "learning_rate": 3.2514216084484162e-06, + "loss": 1.6999, + "step": 1152 + }, + { + "epoch": 1.2635616438356165, + "grad_norm": 2.5289807319641113, + "learning_rate": 3.249390739236393e-06, + "loss": 1.6523, + "step": 1153 + }, + { + "epoch": 1.2646575342465753, + "grad_norm": 2.4168646335601807, + "learning_rate": 3.2473598700243708e-06, + "loss": 1.6823, + "step": 1154 + }, + { + "epoch": 1.2657534246575342, + "grad_norm": 2.5115208625793457, + "learning_rate": 3.2453290008123476e-06, + "loss": 1.6816, + "step": 1155 + }, + { + "epoch": 1.2668493150684932, + "grad_norm": 2.685535192489624, + "learning_rate": 3.2432981316003253e-06, + "loss": 1.6307, + "step": 1156 + }, + { + "epoch": 1.267945205479452, + "grad_norm": 2.253084897994995, + "learning_rate": 3.241267262388302e-06, + "loss": 1.662, + "step": 1157 + }, + { + "epoch": 1.269041095890411, + "grad_norm": 2.4825408458709717, + "learning_rate": 3.23923639317628e-06, + "loss": 1.6444, + "step": 1158 + }, + { + "epoch": 1.2701369863013698, + "grad_norm": 2.691168785095215, + "learning_rate": 3.2372055239642567e-06, + "loss": 1.6912, + "step": 1159 + }, + { + "epoch": 1.2712328767123289, + "grad_norm": 2.978756904602051, + "learning_rate": 3.2351746547522344e-06, + "loss": 1.7536, + "step": 1160 + }, + { + "epoch": 1.2712328767123289, + "eval_loss": 1.8903727531433105, + "eval_runtime": 43.4099, + "eval_samples_per_second": 23.036, + "eval_steps_per_second": 0.968, + "step": 1160 + }, + { + "epoch": 1.2723287671232877, + "grad_norm": 2.317105531692505, + "learning_rate": 3.2331437855402113e-06, + "loss": 1.6441, + "step": 1161 + }, + { + "epoch": 1.2734246575342465, + "grad_norm": 3.1030166149139404, + "learning_rate": 3.2311129163281885e-06, + "loss": 1.7205, + "step": 1162 + }, + { + "epoch": 1.2745205479452055, + "grad_norm": 2.595792770385742, + "learning_rate": 3.229082047116166e-06, + "loss": 1.677, + "step": 1163 + }, + { + "epoch": 1.2756164383561643, + "grad_norm": 2.3315582275390625, + "learning_rate": 3.227051177904143e-06, + "loss": 1.6779, + "step": 1164 + }, + { + "epoch": 1.2767123287671232, + "grad_norm": 2.4262566566467285, + "learning_rate": 3.2250203086921204e-06, + "loss": 1.7003, + "step": 1165 + }, + { + "epoch": 1.2778082191780822, + "grad_norm": 2.4438798427581787, + "learning_rate": 3.2229894394800977e-06, + "loss": 1.6081, + "step": 1166 + }, + { + "epoch": 1.2789041095890412, + "grad_norm": 2.573531150817871, + "learning_rate": 3.220958570268075e-06, + "loss": 1.6489, + "step": 1167 + }, + { + "epoch": 1.28, + "grad_norm": 2.477809429168701, + "learning_rate": 3.218927701056052e-06, + "loss": 1.6025, + "step": 1168 + }, + { + "epoch": 1.2810958904109588, + "grad_norm": 2.5499629974365234, + "learning_rate": 3.2168968318440295e-06, + "loss": 1.7337, + "step": 1169 + }, + { + "epoch": 1.2821917808219179, + "grad_norm": 2.4699153900146484, + "learning_rate": 3.2148659626320068e-06, + "loss": 1.6443, + "step": 1170 + }, + { + "epoch": 1.2821917808219179, + "eval_loss": 1.8867121934890747, + "eval_runtime": 42.9343, + "eval_samples_per_second": 23.291, + "eval_steps_per_second": 0.978, + "step": 1170 + }, + { + "epoch": 1.2832876712328767, + "grad_norm": 2.4200708866119385, + "learning_rate": 3.212835093419984e-06, + "loss": 1.6743, + "step": 1171 + }, + { + "epoch": 1.2843835616438355, + "grad_norm": 2.418069839477539, + "learning_rate": 3.2108042242079613e-06, + "loss": 1.723, + "step": 1172 + }, + { + "epoch": 1.2854794520547945, + "grad_norm": 2.452284336090088, + "learning_rate": 3.2087733549959386e-06, + "loss": 1.6358, + "step": 1173 + }, + { + "epoch": 1.2865753424657536, + "grad_norm": 2.4241347312927246, + "learning_rate": 3.206742485783916e-06, + "loss": 1.638, + "step": 1174 + }, + { + "epoch": 1.2876712328767124, + "grad_norm": 2.461860418319702, + "learning_rate": 3.2047116165718927e-06, + "loss": 1.7017, + "step": 1175 + }, + { + "epoch": 1.2887671232876712, + "grad_norm": 2.591986894607544, + "learning_rate": 3.2026807473598704e-06, + "loss": 1.7065, + "step": 1176 + }, + { + "epoch": 1.2898630136986302, + "grad_norm": 2.478470802307129, + "learning_rate": 3.2006498781478473e-06, + "loss": 1.6262, + "step": 1177 + }, + { + "epoch": 1.290958904109589, + "grad_norm": 2.49831223487854, + "learning_rate": 3.198619008935825e-06, + "loss": 1.6871, + "step": 1178 + }, + { + "epoch": 1.2920547945205478, + "grad_norm": 2.485332727432251, + "learning_rate": 3.196588139723802e-06, + "loss": 1.6864, + "step": 1179 + }, + { + "epoch": 1.2931506849315069, + "grad_norm": 2.533710479736328, + "learning_rate": 3.1945572705117795e-06, + "loss": 1.6794, + "step": 1180 + }, + { + "epoch": 1.2931506849315069, + "eval_loss": 1.88768470287323, + "eval_runtime": 43.5948, + "eval_samples_per_second": 22.938, + "eval_steps_per_second": 0.963, + "step": 1180 + }, + { + "epoch": 1.2942465753424657, + "grad_norm": 2.368609666824341, + "learning_rate": 3.1925264012997564e-06, + "loss": 1.6877, + "step": 1181 + }, + { + "epoch": 1.2953424657534247, + "grad_norm": 2.4875223636627197, + "learning_rate": 3.190495532087734e-06, + "loss": 1.7406, + "step": 1182 + }, + { + "epoch": 1.2964383561643835, + "grad_norm": 2.5075156688690186, + "learning_rate": 3.188464662875711e-06, + "loss": 1.6545, + "step": 1183 + }, + { + "epoch": 1.2975342465753426, + "grad_norm": 2.428731918334961, + "learning_rate": 3.1864337936636886e-06, + "loss": 1.6213, + "step": 1184 + }, + { + "epoch": 1.2986301369863014, + "grad_norm": 2.467618465423584, + "learning_rate": 3.1844029244516655e-06, + "loss": 1.6263, + "step": 1185 + }, + { + "epoch": 1.2997260273972602, + "grad_norm": 2.561612129211426, + "learning_rate": 3.182372055239643e-06, + "loss": 1.6987, + "step": 1186 + }, + { + "epoch": 1.3008219178082192, + "grad_norm": 2.5035271644592285, + "learning_rate": 3.18034118602762e-06, + "loss": 1.6902, + "step": 1187 + }, + { + "epoch": 1.301917808219178, + "grad_norm": 2.538311243057251, + "learning_rate": 3.1783103168155973e-06, + "loss": 1.6523, + "step": 1188 + }, + { + "epoch": 1.303013698630137, + "grad_norm": 2.486321210861206, + "learning_rate": 3.1762794476035746e-06, + "loss": 1.6665, + "step": 1189 + }, + { + "epoch": 1.3041095890410959, + "grad_norm": 2.433173894882202, + "learning_rate": 3.174248578391552e-06, + "loss": 1.6605, + "step": 1190 + }, + { + "epoch": 1.3041095890410959, + "eval_loss": 1.8853652477264404, + "eval_runtime": 43.3417, + "eval_samples_per_second": 23.072, + "eval_steps_per_second": 0.969, + "step": 1190 + }, + { + "epoch": 1.305205479452055, + "grad_norm": 2.4658470153808594, + "learning_rate": 3.172217709179529e-06, + "loss": 1.6615, + "step": 1191 + }, + { + "epoch": 1.3063013698630137, + "grad_norm": 2.452450752258301, + "learning_rate": 3.1701868399675064e-06, + "loss": 1.6332, + "step": 1192 + }, + { + "epoch": 1.3073972602739725, + "grad_norm": 2.5310161113739014, + "learning_rate": 3.1681559707554837e-06, + "loss": 1.6846, + "step": 1193 + }, + { + "epoch": 1.3084931506849315, + "grad_norm": 2.4504501819610596, + "learning_rate": 3.166125101543461e-06, + "loss": 1.6996, + "step": 1194 + }, + { + "epoch": 1.3095890410958904, + "grad_norm": 2.6004488468170166, + "learning_rate": 3.1640942323314382e-06, + "loss": 1.6302, + "step": 1195 + }, + { + "epoch": 1.3106849315068494, + "grad_norm": 2.5000431537628174, + "learning_rate": 3.1620633631194155e-06, + "loss": 1.6742, + "step": 1196 + }, + { + "epoch": 1.3117808219178082, + "grad_norm": 2.4869067668914795, + "learning_rate": 3.1600324939073928e-06, + "loss": 1.6797, + "step": 1197 + }, + { + "epoch": 1.3128767123287672, + "grad_norm": 2.5018959045410156, + "learning_rate": 3.15800162469537e-06, + "loss": 1.673, + "step": 1198 + }, + { + "epoch": 1.313972602739726, + "grad_norm": 2.6645195484161377, + "learning_rate": 3.155970755483347e-06, + "loss": 1.6908, + "step": 1199 + }, + { + "epoch": 1.3150684931506849, + "grad_norm": 2.3708560466766357, + "learning_rate": 3.1539398862713246e-06, + "loss": 1.683, + "step": 1200 + }, + { + "epoch": 1.3150684931506849, + "eval_loss": 1.8829350471496582, + "eval_runtime": 43.5383, + "eval_samples_per_second": 22.968, + "eval_steps_per_second": 0.965, + "step": 1200 + }, + { + "epoch": 1.316164383561644, + "grad_norm": 2.574108362197876, + "learning_rate": 3.1519090170593015e-06, + "loss": 1.7155, + "step": 1201 + }, + { + "epoch": 1.3172602739726027, + "grad_norm": 2.4669814109802246, + "learning_rate": 3.149878147847279e-06, + "loss": 1.7529, + "step": 1202 + }, + { + "epoch": 1.3183561643835615, + "grad_norm": 2.4713149070739746, + "learning_rate": 3.147847278635256e-06, + "loss": 1.6491, + "step": 1203 + }, + { + "epoch": 1.3194520547945205, + "grad_norm": 2.5128800868988037, + "learning_rate": 3.1458164094232337e-06, + "loss": 1.6276, + "step": 1204 + }, + { + "epoch": 1.3205479452054796, + "grad_norm": 2.740370035171509, + "learning_rate": 3.1437855402112106e-06, + "loss": 1.6778, + "step": 1205 + }, + { + "epoch": 1.3216438356164384, + "grad_norm": 2.6038050651550293, + "learning_rate": 3.1417546709991883e-06, + "loss": 1.6527, + "step": 1206 + }, + { + "epoch": 1.3227397260273972, + "grad_norm": 2.782407283782959, + "learning_rate": 3.139723801787165e-06, + "loss": 1.6989, + "step": 1207 + }, + { + "epoch": 1.3238356164383562, + "grad_norm": 2.7673134803771973, + "learning_rate": 3.137692932575143e-06, + "loss": 1.6291, + "step": 1208 + }, + { + "epoch": 1.324931506849315, + "grad_norm": 2.7472870349884033, + "learning_rate": 3.1356620633631197e-06, + "loss": 1.6776, + "step": 1209 + }, + { + "epoch": 1.3260273972602739, + "grad_norm": 2.4887819290161133, + "learning_rate": 3.1336311941510974e-06, + "loss": 1.7009, + "step": 1210 + }, + { + "epoch": 1.3260273972602739, + "eval_loss": 1.8792160749435425, + "eval_runtime": 43.7334, + "eval_samples_per_second": 22.866, + "eval_steps_per_second": 0.96, + "step": 1210 + }, + { + "epoch": 1.3271232876712329, + "grad_norm": 2.4528491497039795, + "learning_rate": 3.1316003249390742e-06, + "loss": 1.7245, + "step": 1211 + }, + { + "epoch": 1.3282191780821917, + "grad_norm": 2.5177690982818604, + "learning_rate": 3.129569455727051e-06, + "loss": 1.6852, + "step": 1212 + }, + { + "epoch": 1.3293150684931507, + "grad_norm": 2.9238510131835938, + "learning_rate": 3.1275385865150288e-06, + "loss": 1.6994, + "step": 1213 + }, + { + "epoch": 1.3304109589041095, + "grad_norm": 2.446660041809082, + "learning_rate": 3.1255077173030056e-06, + "loss": 1.6636, + "step": 1214 + }, + { + "epoch": 1.3315068493150686, + "grad_norm": 2.644211530685425, + "learning_rate": 3.1234768480909833e-06, + "loss": 1.6983, + "step": 1215 + }, + { + "epoch": 1.3326027397260274, + "grad_norm": 2.697342872619629, + "learning_rate": 3.12144597887896e-06, + "loss": 1.7172, + "step": 1216 + }, + { + "epoch": 1.3336986301369862, + "grad_norm": 2.714172124862671, + "learning_rate": 3.119415109666938e-06, + "loss": 1.6702, + "step": 1217 + }, + { + "epoch": 1.3347945205479452, + "grad_norm": 2.471735715866089, + "learning_rate": 3.1173842404549147e-06, + "loss": 1.6945, + "step": 1218 + }, + { + "epoch": 1.335890410958904, + "grad_norm": 2.5089669227600098, + "learning_rate": 3.1153533712428924e-06, + "loss": 1.7181, + "step": 1219 + }, + { + "epoch": 1.336986301369863, + "grad_norm": 2.349229335784912, + "learning_rate": 3.1133225020308693e-06, + "loss": 1.7297, + "step": 1220 + }, + { + "epoch": 1.336986301369863, + "eval_loss": 1.880064845085144, + "eval_runtime": 47.0636, + "eval_samples_per_second": 21.248, + "eval_steps_per_second": 0.892, + "step": 1220 + }, + { + "epoch": 1.3380821917808219, + "grad_norm": 2.6843974590301514, + "learning_rate": 3.111291632818847e-06, + "loss": 1.6698, + "step": 1221 + }, + { + "epoch": 1.339178082191781, + "grad_norm": 2.4325790405273438, + "learning_rate": 3.109260763606824e-06, + "loss": 1.6973, + "step": 1222 + }, + { + "epoch": 1.3402739726027397, + "grad_norm": 2.488459825515747, + "learning_rate": 3.1072298943948015e-06, + "loss": 1.6552, + "step": 1223 + }, + { + "epoch": 1.3413698630136985, + "grad_norm": 2.4992082118988037, + "learning_rate": 3.1051990251827784e-06, + "loss": 1.7114, + "step": 1224 + }, + { + "epoch": 1.3424657534246576, + "grad_norm": 2.4742603302001953, + "learning_rate": 3.1031681559707557e-06, + "loss": 1.7073, + "step": 1225 + }, + { + "epoch": 1.3435616438356164, + "grad_norm": 2.556330442428589, + "learning_rate": 3.101137286758733e-06, + "loss": 1.7108, + "step": 1226 + }, + { + "epoch": 1.3446575342465754, + "grad_norm": 2.5711283683776855, + "learning_rate": 3.09910641754671e-06, + "loss": 1.6811, + "step": 1227 + }, + { + "epoch": 1.3457534246575342, + "grad_norm": 2.6467158794403076, + "learning_rate": 3.0970755483346875e-06, + "loss": 1.7395, + "step": 1228 + }, + { + "epoch": 1.3468493150684933, + "grad_norm": 2.465963363647461, + "learning_rate": 3.0950446791226648e-06, + "loss": 1.7006, + "step": 1229 + }, + { + "epoch": 1.347945205479452, + "grad_norm": 2.622716188430786, + "learning_rate": 3.093013809910642e-06, + "loss": 1.7385, + "step": 1230 + }, + { + "epoch": 1.347945205479452, + "eval_loss": 1.8808013200759888, + "eval_runtime": 45.5126, + "eval_samples_per_second": 21.972, + "eval_steps_per_second": 0.923, + "step": 1230 + }, + { + "epoch": 1.3490410958904109, + "grad_norm": 2.489701509475708, + "learning_rate": 3.0909829406986193e-06, + "loss": 1.6949, + "step": 1231 + }, + { + "epoch": 1.35013698630137, + "grad_norm": 2.4398739337921143, + "learning_rate": 3.0889520714865966e-06, + "loss": 1.6802, + "step": 1232 + }, + { + "epoch": 1.3512328767123287, + "grad_norm": 2.3654873371124268, + "learning_rate": 3.086921202274574e-06, + "loss": 1.6452, + "step": 1233 + }, + { + "epoch": 1.3523287671232875, + "grad_norm": 2.518768548965454, + "learning_rate": 3.084890333062551e-06, + "loss": 1.6913, + "step": 1234 + }, + { + "epoch": 1.3534246575342466, + "grad_norm": 2.63173246383667, + "learning_rate": 3.0828594638505284e-06, + "loss": 1.6611, + "step": 1235 + }, + { + "epoch": 1.3545205479452056, + "grad_norm": 2.531484603881836, + "learning_rate": 3.0808285946385057e-06, + "loss": 1.611, + "step": 1236 + }, + { + "epoch": 1.3556164383561644, + "grad_norm": 2.4796738624572754, + "learning_rate": 3.078797725426483e-06, + "loss": 1.6764, + "step": 1237 + }, + { + "epoch": 1.3567123287671232, + "grad_norm": 2.39241099357605, + "learning_rate": 3.07676685621446e-06, + "loss": 1.6417, + "step": 1238 + }, + { + "epoch": 1.3578082191780823, + "grad_norm": 2.3984735012054443, + "learning_rate": 3.0747359870024375e-06, + "loss": 1.6546, + "step": 1239 + }, + { + "epoch": 1.358904109589041, + "grad_norm": 2.5202760696411133, + "learning_rate": 3.0727051177904144e-06, + "loss": 1.6879, + "step": 1240 + }, + { + "epoch": 1.358904109589041, + "eval_loss": 1.879974365234375, + "eval_runtime": 44.395, + "eval_samples_per_second": 22.525, + "eval_steps_per_second": 0.946, + "step": 1240 + }, + { + "epoch": 1.3599999999999999, + "grad_norm": 2.5556983947753906, + "learning_rate": 3.070674248578392e-06, + "loss": 1.6733, + "step": 1241 + }, + { + "epoch": 1.361095890410959, + "grad_norm": 2.3853955268859863, + "learning_rate": 3.068643379366369e-06, + "loss": 1.6274, + "step": 1242 + }, + { + "epoch": 1.362191780821918, + "grad_norm": 2.4503982067108154, + "learning_rate": 3.0666125101543466e-06, + "loss": 1.6896, + "step": 1243 + }, + { + "epoch": 1.3632876712328768, + "grad_norm": 2.5618021488189697, + "learning_rate": 3.0645816409423235e-06, + "loss": 1.6592, + "step": 1244 + }, + { + "epoch": 1.3643835616438356, + "grad_norm": 2.6265292167663574, + "learning_rate": 3.062550771730301e-06, + "loss": 1.6634, + "step": 1245 + }, + { + "epoch": 1.3654794520547946, + "grad_norm": 2.4617390632629395, + "learning_rate": 3.060519902518278e-06, + "loss": 1.6032, + "step": 1246 + }, + { + "epoch": 1.3665753424657534, + "grad_norm": 2.4191603660583496, + "learning_rate": 3.0584890333062557e-06, + "loss": 1.6871, + "step": 1247 + }, + { + "epoch": 1.3676712328767122, + "grad_norm": 2.5509119033813477, + "learning_rate": 3.0564581640942326e-06, + "loss": 1.653, + "step": 1248 + }, + { + "epoch": 1.3687671232876713, + "grad_norm": 2.5399646759033203, + "learning_rate": 3.0544272948822094e-06, + "loss": 1.7046, + "step": 1249 + }, + { + "epoch": 1.36986301369863, + "grad_norm": 2.446967124938965, + "learning_rate": 3.052396425670187e-06, + "loss": 1.6547, + "step": 1250 + }, + { + "epoch": 1.36986301369863, + "eval_loss": 1.8803874254226685, + "eval_runtime": 45.361, + "eval_samples_per_second": 22.045, + "eval_steps_per_second": 0.926, + "step": 1250 + }, + { + "epoch": 1.370958904109589, + "grad_norm": 2.419206142425537, + "learning_rate": 3.050365556458164e-06, + "loss": 1.6517, + "step": 1251 + }, + { + "epoch": 1.372054794520548, + "grad_norm": 2.4592065811157227, + "learning_rate": 3.0483346872461417e-06, + "loss": 1.693, + "step": 1252 + }, + { + "epoch": 1.373150684931507, + "grad_norm": 2.5429770946502686, + "learning_rate": 3.0463038180341185e-06, + "loss": 1.7124, + "step": 1253 + }, + { + "epoch": 1.3742465753424657, + "grad_norm": 2.5141642093658447, + "learning_rate": 3.0442729488220962e-06, + "loss": 1.7103, + "step": 1254 + }, + { + "epoch": 1.3753424657534246, + "grad_norm": 2.362539291381836, + "learning_rate": 3.042242079610073e-06, + "loss": 1.6205, + "step": 1255 + }, + { + "epoch": 1.3764383561643836, + "grad_norm": 2.401175022125244, + "learning_rate": 3.0402112103980508e-06, + "loss": 1.6479, + "step": 1256 + }, + { + "epoch": 1.3775342465753424, + "grad_norm": 2.5883190631866455, + "learning_rate": 3.0381803411860276e-06, + "loss": 1.6732, + "step": 1257 + }, + { + "epoch": 1.3786301369863014, + "grad_norm": 2.5273513793945312, + "learning_rate": 3.0361494719740053e-06, + "loss": 1.6673, + "step": 1258 + }, + { + "epoch": 1.3797260273972602, + "grad_norm": 2.6569600105285645, + "learning_rate": 3.034118602761982e-06, + "loss": 1.6396, + "step": 1259 + }, + { + "epoch": 1.3808219178082193, + "grad_norm": 2.8843557834625244, + "learning_rate": 3.03208773354996e-06, + "loss": 1.7136, + "step": 1260 + }, + { + "epoch": 1.3808219178082193, + "eval_loss": 1.8763896226882935, + "eval_runtime": 46.1886, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1260 + }, + { + "epoch": 1.381917808219178, + "grad_norm": 2.6459810733795166, + "learning_rate": 3.0300568643379367e-06, + "loss": 1.6939, + "step": 1261 + }, + { + "epoch": 1.383013698630137, + "grad_norm": 2.4619321823120117, + "learning_rate": 3.028025995125914e-06, + "loss": 1.6197, + "step": 1262 + }, + { + "epoch": 1.384109589041096, + "grad_norm": 2.390430450439453, + "learning_rate": 3.0259951259138913e-06, + "loss": 1.6805, + "step": 1263 + }, + { + "epoch": 1.3852054794520547, + "grad_norm": 2.5494728088378906, + "learning_rate": 3.0239642567018686e-06, + "loss": 1.6116, + "step": 1264 + }, + { + "epoch": 1.3863013698630138, + "grad_norm": 2.5397655963897705, + "learning_rate": 3.021933387489846e-06, + "loss": 1.6753, + "step": 1265 + }, + { + "epoch": 1.3873972602739726, + "grad_norm": 2.634716510772705, + "learning_rate": 3.019902518277823e-06, + "loss": 1.6892, + "step": 1266 + }, + { + "epoch": 1.3884931506849316, + "grad_norm": 2.5844202041625977, + "learning_rate": 3.0178716490658004e-06, + "loss": 1.6401, + "step": 1267 + }, + { + "epoch": 1.3895890410958904, + "grad_norm": 2.577362060546875, + "learning_rate": 3.0158407798537777e-06, + "loss": 1.6405, + "step": 1268 + }, + { + "epoch": 1.3906849315068492, + "grad_norm": 2.3554489612579346, + "learning_rate": 3.013809910641755e-06, + "loss": 1.6427, + "step": 1269 + }, + { + "epoch": 1.3917808219178083, + "grad_norm": 2.893918037414551, + "learning_rate": 3.0117790414297322e-06, + "loss": 1.6739, + "step": 1270 + }, + { + "epoch": 1.3917808219178083, + "eval_loss": 1.8772739171981812, + "eval_runtime": 44.9972, + "eval_samples_per_second": 22.224, + "eval_steps_per_second": 0.933, + "step": 1270 + }, + { + "epoch": 1.392876712328767, + "grad_norm": 2.4818382263183594, + "learning_rate": 3.0097481722177095e-06, + "loss": 1.644, + "step": 1271 + }, + { + "epoch": 1.393972602739726, + "grad_norm": 2.4990177154541016, + "learning_rate": 3.0077173030056868e-06, + "loss": 1.6962, + "step": 1272 + }, + { + "epoch": 1.395068493150685, + "grad_norm": 2.627502918243408, + "learning_rate": 3.005686433793664e-06, + "loss": 1.639, + "step": 1273 + }, + { + "epoch": 1.396164383561644, + "grad_norm": 2.5949175357818604, + "learning_rate": 3.0036555645816413e-06, + "loss": 1.7095, + "step": 1274 + }, + { + "epoch": 1.3972602739726028, + "grad_norm": 2.422765016555786, + "learning_rate": 3.001624695369618e-06, + "loss": 1.6572, + "step": 1275 + }, + { + "epoch": 1.3983561643835616, + "grad_norm": 2.708702564239502, + "learning_rate": 2.999593826157596e-06, + "loss": 1.6747, + "step": 1276 + }, + { + "epoch": 1.3994520547945206, + "grad_norm": 2.624866247177124, + "learning_rate": 2.9975629569455727e-06, + "loss": 1.6643, + "step": 1277 + }, + { + "epoch": 1.4005479452054794, + "grad_norm": 2.5831358432769775, + "learning_rate": 2.9955320877335504e-06, + "loss": 1.7036, + "step": 1278 + }, + { + "epoch": 1.4016438356164382, + "grad_norm": 2.5799436569213867, + "learning_rate": 2.9935012185215273e-06, + "loss": 1.6628, + "step": 1279 + }, + { + "epoch": 1.4027397260273973, + "grad_norm": 2.7047722339630127, + "learning_rate": 2.991470349309505e-06, + "loss": 1.6591, + "step": 1280 + }, + { + "epoch": 1.4027397260273973, + "eval_loss": 1.8736577033996582, + "eval_runtime": 44.7095, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1280 + }, + { + "epoch": 1.403835616438356, + "grad_norm": 2.411628246307373, + "learning_rate": 2.989439480097482e-06, + "loss": 1.6644, + "step": 1281 + }, + { + "epoch": 1.4049315068493151, + "grad_norm": 2.8774638175964355, + "learning_rate": 2.9874086108854595e-06, + "loss": 1.7192, + "step": 1282 + }, + { + "epoch": 1.406027397260274, + "grad_norm": 2.857028007507324, + "learning_rate": 2.9853777416734364e-06, + "loss": 1.7079, + "step": 1283 + }, + { + "epoch": 1.407123287671233, + "grad_norm": 2.4619641304016113, + "learning_rate": 2.983346872461414e-06, + "loss": 1.6846, + "step": 1284 + }, + { + "epoch": 1.4082191780821918, + "grad_norm": 2.6566193103790283, + "learning_rate": 2.981316003249391e-06, + "loss": 1.6768, + "step": 1285 + }, + { + "epoch": 1.4093150684931506, + "grad_norm": 2.681217670440674, + "learning_rate": 2.9792851340373678e-06, + "loss": 1.7002, + "step": 1286 + }, + { + "epoch": 1.4104109589041096, + "grad_norm": 2.39505672454834, + "learning_rate": 2.9772542648253455e-06, + "loss": 1.65, + "step": 1287 + }, + { + "epoch": 1.4115068493150684, + "grad_norm": 2.6062734127044678, + "learning_rate": 2.9752233956133223e-06, + "loss": 1.6659, + "step": 1288 + }, + { + "epoch": 1.4126027397260275, + "grad_norm": 2.6402504444122314, + "learning_rate": 2.9731925264013e-06, + "loss": 1.6491, + "step": 1289 + }, + { + "epoch": 1.4136986301369863, + "grad_norm": 2.497014284133911, + "learning_rate": 2.971161657189277e-06, + "loss": 1.6383, + "step": 1290 + }, + { + "epoch": 1.4136986301369863, + "eval_loss": 1.874670147895813, + "eval_runtime": 42.9696, + "eval_samples_per_second": 23.272, + "eval_steps_per_second": 0.977, + "step": 1290 + }, + { + "epoch": 1.4147945205479453, + "grad_norm": 2.9070584774017334, + "learning_rate": 2.9691307879772546e-06, + "loss": 1.6388, + "step": 1291 + }, + { + "epoch": 1.415890410958904, + "grad_norm": 2.580744981765747, + "learning_rate": 2.9670999187652314e-06, + "loss": 1.7078, + "step": 1292 + }, + { + "epoch": 1.416986301369863, + "grad_norm": 2.4637644290924072, + "learning_rate": 2.965069049553209e-06, + "loss": 1.6266, + "step": 1293 + }, + { + "epoch": 1.418082191780822, + "grad_norm": 2.4489941596984863, + "learning_rate": 2.963038180341186e-06, + "loss": 1.705, + "step": 1294 + }, + { + "epoch": 1.4191780821917808, + "grad_norm": 2.664654016494751, + "learning_rate": 2.9610073111291637e-06, + "loss": 1.6964, + "step": 1295 + }, + { + "epoch": 1.4202739726027398, + "grad_norm": 2.7017202377319336, + "learning_rate": 2.9589764419171405e-06, + "loss": 1.6356, + "step": 1296 + }, + { + "epoch": 1.4213698630136986, + "grad_norm": 2.3691024780273438, + "learning_rate": 2.9569455727051182e-06, + "loss": 1.6566, + "step": 1297 + }, + { + "epoch": 1.4224657534246576, + "grad_norm": 2.540384292602539, + "learning_rate": 2.954914703493095e-06, + "loss": 1.6334, + "step": 1298 + }, + { + "epoch": 1.4235616438356165, + "grad_norm": 2.5240535736083984, + "learning_rate": 2.9528838342810724e-06, + "loss": 1.6316, + "step": 1299 + }, + { + "epoch": 1.4246575342465753, + "grad_norm": 3.105684518814087, + "learning_rate": 2.9508529650690496e-06, + "loss": 1.6769, + "step": 1300 + }, + { + "epoch": 1.4246575342465753, + "eval_loss": 1.8741132020950317, + "eval_runtime": 42.6795, + "eval_samples_per_second": 23.43, + "eval_steps_per_second": 0.984, + "step": 1300 + }, + { + "epoch": 1.4257534246575343, + "grad_norm": 2.494252920150757, + "learning_rate": 2.948822095857027e-06, + "loss": 1.6359, + "step": 1301 + }, + { + "epoch": 1.426849315068493, + "grad_norm": 4.069019317626953, + "learning_rate": 2.946791226645004e-06, + "loss": 1.621, + "step": 1302 + }, + { + "epoch": 1.427945205479452, + "grad_norm": 2.641447067260742, + "learning_rate": 2.9447603574329815e-06, + "loss": 1.6502, + "step": 1303 + }, + { + "epoch": 1.429041095890411, + "grad_norm": 2.6700453758239746, + "learning_rate": 2.9427294882209587e-06, + "loss": 1.6593, + "step": 1304 + }, + { + "epoch": 1.43013698630137, + "grad_norm": 2.686068296432495, + "learning_rate": 2.940698619008936e-06, + "loss": 1.67, + "step": 1305 + }, + { + "epoch": 1.4312328767123288, + "grad_norm": 2.409055471420288, + "learning_rate": 2.9386677497969133e-06, + "loss": 1.6809, + "step": 1306 + }, + { + "epoch": 1.4323287671232876, + "grad_norm": 2.4707298278808594, + "learning_rate": 2.9366368805848906e-06, + "loss": 1.69, + "step": 1307 + }, + { + "epoch": 1.4334246575342466, + "grad_norm": 2.4735982418060303, + "learning_rate": 2.934606011372868e-06, + "loss": 1.666, + "step": 1308 + }, + { + "epoch": 1.4345205479452054, + "grad_norm": 2.633183479309082, + "learning_rate": 2.932575142160845e-06, + "loss": 1.6597, + "step": 1309 + }, + { + "epoch": 1.4356164383561643, + "grad_norm": 2.4808642864227295, + "learning_rate": 2.9305442729488224e-06, + "loss": 1.6667, + "step": 1310 + }, + { + "epoch": 1.4356164383561643, + "eval_loss": 1.8731157779693604, + "eval_runtime": 41.3423, + "eval_samples_per_second": 24.188, + "eval_steps_per_second": 1.016, + "step": 1310 + }, + { + "epoch": 1.4367123287671233, + "grad_norm": 2.4044833183288574, + "learning_rate": 2.9285134037367997e-06, + "loss": 1.7028, + "step": 1311 + }, + { + "epoch": 1.4378082191780823, + "grad_norm": 2.942699909210205, + "learning_rate": 2.9264825345247765e-06, + "loss": 1.6678, + "step": 1312 + }, + { + "epoch": 1.4389041095890411, + "grad_norm": 2.505645990371704, + "learning_rate": 2.9244516653127542e-06, + "loss": 1.5972, + "step": 1313 + }, + { + "epoch": 1.44, + "grad_norm": 2.700967788696289, + "learning_rate": 2.922420796100731e-06, + "loss": 1.7064, + "step": 1314 + }, + { + "epoch": 1.441095890410959, + "grad_norm": 2.4805755615234375, + "learning_rate": 2.9203899268887088e-06, + "loss": 1.6201, + "step": 1315 + }, + { + "epoch": 1.4421917808219178, + "grad_norm": 2.753181219100952, + "learning_rate": 2.9183590576766856e-06, + "loss": 1.6704, + "step": 1316 + }, + { + "epoch": 1.4432876712328766, + "grad_norm": 2.488865613937378, + "learning_rate": 2.9163281884646633e-06, + "loss": 1.7057, + "step": 1317 + }, + { + "epoch": 1.4443835616438356, + "grad_norm": 2.5464885234832764, + "learning_rate": 2.91429731925264e-06, + "loss": 1.6678, + "step": 1318 + }, + { + "epoch": 1.4454794520547944, + "grad_norm": 2.4997060298919678, + "learning_rate": 2.912266450040618e-06, + "loss": 1.662, + "step": 1319 + }, + { + "epoch": 1.4465753424657535, + "grad_norm": 2.5162436962127686, + "learning_rate": 2.9102355808285947e-06, + "loss": 1.6501, + "step": 1320 + }, + { + "epoch": 1.4465753424657535, + "eval_loss": 1.869848370552063, + "eval_runtime": 43.3583, + "eval_samples_per_second": 23.064, + "eval_steps_per_second": 0.969, + "step": 1320 + }, + { + "epoch": 1.4476712328767123, + "grad_norm": 2.616129159927368, + "learning_rate": 2.9082047116165724e-06, + "loss": 1.6206, + "step": 1321 + }, + { + "epoch": 1.4487671232876713, + "grad_norm": 2.761080503463745, + "learning_rate": 2.9061738424045493e-06, + "loss": 1.7131, + "step": 1322 + }, + { + "epoch": 1.4498630136986301, + "grad_norm": 2.5994064807891846, + "learning_rate": 2.904142973192527e-06, + "loss": 1.6693, + "step": 1323 + }, + { + "epoch": 1.450958904109589, + "grad_norm": 3.4927968978881836, + "learning_rate": 2.902112103980504e-06, + "loss": 1.6694, + "step": 1324 + }, + { + "epoch": 1.452054794520548, + "grad_norm": 2.6050939559936523, + "learning_rate": 2.900081234768481e-06, + "loss": 1.6419, + "step": 1325 + }, + { + "epoch": 1.4531506849315068, + "grad_norm": 2.796086311340332, + "learning_rate": 2.8980503655564584e-06, + "loss": 1.6354, + "step": 1326 + }, + { + "epoch": 1.4542465753424658, + "grad_norm": 2.5658411979675293, + "learning_rate": 2.8960194963444357e-06, + "loss": 1.652, + "step": 1327 + }, + { + "epoch": 1.4553424657534246, + "grad_norm": 2.3802385330200195, + "learning_rate": 2.893988627132413e-06, + "loss": 1.5902, + "step": 1328 + }, + { + "epoch": 1.4564383561643837, + "grad_norm": 2.479189872741699, + "learning_rate": 2.8919577579203902e-06, + "loss": 1.6617, + "step": 1329 + }, + { + "epoch": 1.4575342465753425, + "grad_norm": 2.4367568492889404, + "learning_rate": 2.8899268887083675e-06, + "loss": 1.6691, + "step": 1330 + }, + { + "epoch": 1.4575342465753425, + "eval_loss": 1.868497371673584, + "eval_runtime": 45.172, + "eval_samples_per_second": 22.138, + "eval_steps_per_second": 0.93, + "step": 1330 + }, + { + "epoch": 1.4586301369863013, + "grad_norm": 2.51888370513916, + "learning_rate": 2.8878960194963448e-06, + "loss": 1.6872, + "step": 1331 + }, + { + "epoch": 1.4597260273972603, + "grad_norm": 2.5192768573760986, + "learning_rate": 2.885865150284322e-06, + "loss": 1.6584, + "step": 1332 + }, + { + "epoch": 1.4608219178082191, + "grad_norm": 2.5898501873016357, + "learning_rate": 2.8838342810722993e-06, + "loss": 1.6179, + "step": 1333 + }, + { + "epoch": 1.4619178082191782, + "grad_norm": 2.698101043701172, + "learning_rate": 2.8818034118602766e-06, + "loss": 1.663, + "step": 1334 + }, + { + "epoch": 1.463013698630137, + "grad_norm": 2.889942169189453, + "learning_rate": 2.879772542648254e-06, + "loss": 1.5954, + "step": 1335 + }, + { + "epoch": 1.464109589041096, + "grad_norm": 2.7909929752349854, + "learning_rate": 2.8777416734362307e-06, + "loss": 1.7264, + "step": 1336 + }, + { + "epoch": 1.4652054794520548, + "grad_norm": 3.1931703090667725, + "learning_rate": 2.8757108042242084e-06, + "loss": 1.7091, + "step": 1337 + }, + { + "epoch": 1.4663013698630136, + "grad_norm": 2.7934632301330566, + "learning_rate": 2.8736799350121853e-06, + "loss": 1.7204, + "step": 1338 + }, + { + "epoch": 1.4673972602739727, + "grad_norm": 2.470545530319214, + "learning_rate": 2.871649065800163e-06, + "loss": 1.644, + "step": 1339 + }, + { + "epoch": 1.4684931506849315, + "grad_norm": 3.011662006378174, + "learning_rate": 2.86961819658814e-06, + "loss": 1.6743, + "step": 1340 + }, + { + "epoch": 1.4684931506849315, + "eval_loss": 1.8678737878799438, + "eval_runtime": 44.7875, + "eval_samples_per_second": 22.328, + "eval_steps_per_second": 0.938, + "step": 1340 + }, + { + "epoch": 1.4695890410958903, + "grad_norm": 2.60823655128479, + "learning_rate": 2.8675873273761175e-06, + "loss": 1.6467, + "step": 1341 + }, + { + "epoch": 1.4706849315068493, + "grad_norm": 2.5347230434417725, + "learning_rate": 2.8655564581640944e-06, + "loss": 1.7044, + "step": 1342 + }, + { + "epoch": 1.4717808219178083, + "grad_norm": 2.481715440750122, + "learning_rate": 2.863525588952072e-06, + "loss": 1.5969, + "step": 1343 + }, + { + "epoch": 1.4728767123287672, + "grad_norm": 2.4966487884521484, + "learning_rate": 2.861494719740049e-06, + "loss": 1.6231, + "step": 1344 + }, + { + "epoch": 1.473972602739726, + "grad_norm": 2.8809802532196045, + "learning_rate": 2.8594638505280266e-06, + "loss": 1.6562, + "step": 1345 + }, + { + "epoch": 1.475068493150685, + "grad_norm": 2.570204734802246, + "learning_rate": 2.8574329813160035e-06, + "loss": 1.6511, + "step": 1346 + }, + { + "epoch": 1.4761643835616438, + "grad_norm": 2.6073813438415527, + "learning_rate": 2.855402112103981e-06, + "loss": 1.6456, + "step": 1347 + }, + { + "epoch": 1.4772602739726026, + "grad_norm": 2.8125052452087402, + "learning_rate": 2.853371242891958e-06, + "loss": 1.6722, + "step": 1348 + }, + { + "epoch": 1.4783561643835617, + "grad_norm": 5.341272354125977, + "learning_rate": 2.851340373679935e-06, + "loss": 1.7025, + "step": 1349 + }, + { + "epoch": 1.4794520547945205, + "grad_norm": 3.138002395629883, + "learning_rate": 2.8493095044679126e-06, + "loss": 1.6676, + "step": 1350 + }, + { + "epoch": 1.4794520547945205, + "eval_loss": 1.8733265399932861, + "eval_runtime": 45.4915, + "eval_samples_per_second": 21.982, + "eval_steps_per_second": 0.923, + "step": 1350 + }, + { + "epoch": 1.4805479452054795, + "grad_norm": 7.264789581298828, + "learning_rate": 2.8472786352558894e-06, + "loss": 1.6335, + "step": 1351 + }, + { + "epoch": 1.4816438356164383, + "grad_norm": 3.8533623218536377, + "learning_rate": 2.845247766043867e-06, + "loss": 1.6029, + "step": 1352 + }, + { + "epoch": 1.4827397260273973, + "grad_norm": 4.1789164543151855, + "learning_rate": 2.843216896831844e-06, + "loss": 1.6368, + "step": 1353 + }, + { + "epoch": 1.4838356164383562, + "grad_norm": 3.117408037185669, + "learning_rate": 2.8411860276198217e-06, + "loss": 1.6924, + "step": 1354 + }, + { + "epoch": 1.484931506849315, + "grad_norm": 2.721306085586548, + "learning_rate": 2.8391551584077985e-06, + "loss": 1.6914, + "step": 1355 + }, + { + "epoch": 1.486027397260274, + "grad_norm": 2.7099668979644775, + "learning_rate": 2.8371242891957762e-06, + "loss": 1.6508, + "step": 1356 + }, + { + "epoch": 1.4871232876712328, + "grad_norm": 3.7823967933654785, + "learning_rate": 2.835093419983753e-06, + "loss": 1.636, + "step": 1357 + }, + { + "epoch": 1.4882191780821918, + "grad_norm": 2.696370840072632, + "learning_rate": 2.833062550771731e-06, + "loss": 1.6175, + "step": 1358 + }, + { + "epoch": 1.4893150684931507, + "grad_norm": 2.7303690910339355, + "learning_rate": 2.8310316815597076e-06, + "loss": 1.6654, + "step": 1359 + }, + { + "epoch": 1.4904109589041097, + "grad_norm": 2.867727041244507, + "learning_rate": 2.8290008123476853e-06, + "loss": 1.6717, + "step": 1360 + }, + { + "epoch": 1.4904109589041097, + "eval_loss": 1.8720332384109497, + "eval_runtime": 44.164, + "eval_samples_per_second": 22.643, + "eval_steps_per_second": 0.951, + "step": 1360 + }, + { + "epoch": 1.4915068493150685, + "grad_norm": 3.1209630966186523, + "learning_rate": 2.826969943135662e-06, + "loss": 1.6635, + "step": 1361 + }, + { + "epoch": 1.4926027397260273, + "grad_norm": 2.5057783126831055, + "learning_rate": 2.8249390739236395e-06, + "loss": 1.6309, + "step": 1362 + }, + { + "epoch": 1.4936986301369863, + "grad_norm": 2.6650168895721436, + "learning_rate": 2.8229082047116167e-06, + "loss": 1.6365, + "step": 1363 + }, + { + "epoch": 1.4947945205479451, + "grad_norm": 2.371305465698242, + "learning_rate": 2.820877335499594e-06, + "loss": 1.603, + "step": 1364 + }, + { + "epoch": 1.4958904109589042, + "grad_norm": 2.387404203414917, + "learning_rate": 2.8188464662875713e-06, + "loss": 1.6189, + "step": 1365 + }, + { + "epoch": 1.496986301369863, + "grad_norm": 2.6265947818756104, + "learning_rate": 2.8168155970755486e-06, + "loss": 1.6511, + "step": 1366 + }, + { + "epoch": 1.498082191780822, + "grad_norm": 2.6108627319335938, + "learning_rate": 2.814784727863526e-06, + "loss": 1.6727, + "step": 1367 + }, + { + "epoch": 1.4991780821917808, + "grad_norm": 2.7021450996398926, + "learning_rate": 2.812753858651503e-06, + "loss": 1.6334, + "step": 1368 + }, + { + "epoch": 1.5002739726027396, + "grad_norm": 2.470118999481201, + "learning_rate": 2.8107229894394804e-06, + "loss": 1.6155, + "step": 1369 + }, + { + "epoch": 1.5013698630136987, + "grad_norm": 2.5722644329071045, + "learning_rate": 2.8086921202274577e-06, + "loss": 1.6788, + "step": 1370 + }, + { + "epoch": 1.5013698630136987, + "eval_loss": 1.8727209568023682, + "eval_runtime": 44.6253, + "eval_samples_per_second": 22.409, + "eval_steps_per_second": 0.941, + "step": 1370 + }, + { + "epoch": 1.5024657534246575, + "grad_norm": 2.7143876552581787, + "learning_rate": 2.806661251015435e-06, + "loss": 1.6956, + "step": 1371 + }, + { + "epoch": 1.5035616438356163, + "grad_norm": 2.8522002696990967, + "learning_rate": 2.8046303818034122e-06, + "loss": 1.7558, + "step": 1372 + }, + { + "epoch": 1.5046575342465753, + "grad_norm": 2.5352110862731934, + "learning_rate": 2.802599512591389e-06, + "loss": 1.6592, + "step": 1373 + }, + { + "epoch": 1.5057534246575344, + "grad_norm": 2.7065908908843994, + "learning_rate": 2.8005686433793668e-06, + "loss": 1.6121, + "step": 1374 + }, + { + "epoch": 1.5068493150684932, + "grad_norm": 2.616450786590576, + "learning_rate": 2.7985377741673436e-06, + "loss": 1.6643, + "step": 1375 + }, + { + "epoch": 1.507945205479452, + "grad_norm": 2.528763771057129, + "learning_rate": 2.7965069049553213e-06, + "loss": 1.7104, + "step": 1376 + }, + { + "epoch": 1.509041095890411, + "grad_norm": 3.725569009780884, + "learning_rate": 2.794476035743298e-06, + "loss": 1.6713, + "step": 1377 + }, + { + "epoch": 1.5101369863013698, + "grad_norm": 2.487497091293335, + "learning_rate": 2.792445166531276e-06, + "loss": 1.692, + "step": 1378 + }, + { + "epoch": 1.5112328767123286, + "grad_norm": 2.534766912460327, + "learning_rate": 2.7904142973192527e-06, + "loss": 1.6257, + "step": 1379 + }, + { + "epoch": 1.5123287671232877, + "grad_norm": 2.8553240299224854, + "learning_rate": 2.7883834281072304e-06, + "loss": 1.6147, + "step": 1380 + }, + { + "epoch": 1.5123287671232877, + "eval_loss": 1.8675988912582397, + "eval_runtime": 46.4445, + "eval_samples_per_second": 21.531, + "eval_steps_per_second": 0.904, + "step": 1380 + }, + { + "epoch": 1.5134246575342467, + "grad_norm": 2.5054166316986084, + "learning_rate": 2.7863525588952073e-06, + "loss": 1.6082, + "step": 1381 + }, + { + "epoch": 1.5145205479452055, + "grad_norm": 2.4133870601654053, + "learning_rate": 2.784321689683185e-06, + "loss": 1.6148, + "step": 1382 + }, + { + "epoch": 1.5156164383561643, + "grad_norm": 2.5863375663757324, + "learning_rate": 2.782290820471162e-06, + "loss": 1.7032, + "step": 1383 + }, + { + "epoch": 1.5167123287671234, + "grad_norm": 2.58491849899292, + "learning_rate": 2.7802599512591395e-06, + "loss": 1.679, + "step": 1384 + }, + { + "epoch": 1.5178082191780822, + "grad_norm": 2.5225989818573, + "learning_rate": 2.7782290820471164e-06, + "loss": 1.6688, + "step": 1385 + }, + { + "epoch": 1.518904109589041, + "grad_norm": 2.366861343383789, + "learning_rate": 2.7761982128350932e-06, + "loss": 1.7076, + "step": 1386 + }, + { + "epoch": 1.52, + "grad_norm": 2.4159810543060303, + "learning_rate": 2.774167343623071e-06, + "loss": 1.6301, + "step": 1387 + }, + { + "epoch": 1.521095890410959, + "grad_norm": 2.5394275188446045, + "learning_rate": 2.772136474411048e-06, + "loss": 1.5758, + "step": 1388 + }, + { + "epoch": 1.5221917808219179, + "grad_norm": 2.409821033477783, + "learning_rate": 2.7701056051990255e-06, + "loss": 1.664, + "step": 1389 + }, + { + "epoch": 1.5232876712328767, + "grad_norm": 2.4548892974853516, + "learning_rate": 2.7680747359870023e-06, + "loss": 1.643, + "step": 1390 + }, + { + "epoch": 1.5232876712328767, + "eval_loss": 1.8672850131988525, + "eval_runtime": 46.3149, + "eval_samples_per_second": 21.591, + "eval_steps_per_second": 0.907, + "step": 1390 + }, + { + "epoch": 1.5243835616438357, + "grad_norm": 2.7483904361724854, + "learning_rate": 2.76604386677498e-06, + "loss": 1.6898, + "step": 1391 + }, + { + "epoch": 1.5254794520547945, + "grad_norm": 2.538804769515991, + "learning_rate": 2.764012997562957e-06, + "loss": 1.6019, + "step": 1392 + }, + { + "epoch": 1.5265753424657533, + "grad_norm": 2.7153313159942627, + "learning_rate": 2.7619821283509346e-06, + "loss": 1.691, + "step": 1393 + }, + { + "epoch": 1.5276712328767124, + "grad_norm": 2.6892242431640625, + "learning_rate": 2.7599512591389114e-06, + "loss": 1.6386, + "step": 1394 + }, + { + "epoch": 1.5287671232876714, + "grad_norm": 2.550957679748535, + "learning_rate": 2.757920389926889e-06, + "loss": 1.6237, + "step": 1395 + }, + { + "epoch": 1.52986301369863, + "grad_norm": 2.3658559322357178, + "learning_rate": 2.755889520714866e-06, + "loss": 1.6015, + "step": 1396 + }, + { + "epoch": 1.530958904109589, + "grad_norm": 2.6524102687835693, + "learning_rate": 2.7538586515028437e-06, + "loss": 1.6644, + "step": 1397 + }, + { + "epoch": 1.532054794520548, + "grad_norm": 2.7067437171936035, + "learning_rate": 2.7518277822908205e-06, + "loss": 1.687, + "step": 1398 + }, + { + "epoch": 1.5331506849315069, + "grad_norm": 2.4044861793518066, + "learning_rate": 2.749796913078798e-06, + "loss": 1.6535, + "step": 1399 + }, + { + "epoch": 1.5342465753424657, + "grad_norm": 2.4933829307556152, + "learning_rate": 2.747766043866775e-06, + "loss": 1.7017, + "step": 1400 + }, + { + "epoch": 1.5342465753424657, + "eval_loss": 1.8652688264846802, + "eval_runtime": 46.8502, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1400 + }, + { + "epoch": 1.5353424657534247, + "grad_norm": 2.7062129974365234, + "learning_rate": 2.7457351746547524e-06, + "loss": 1.7398, + "step": 1401 + }, + { + "epoch": 1.5364383561643835, + "grad_norm": 2.5704996585845947, + "learning_rate": 2.7437043054427297e-06, + "loss": 1.6561, + "step": 1402 + }, + { + "epoch": 1.5375342465753423, + "grad_norm": 2.5875790119171143, + "learning_rate": 2.741673436230707e-06, + "loss": 1.643, + "step": 1403 + }, + { + "epoch": 1.5386301369863014, + "grad_norm": 2.7498223781585693, + "learning_rate": 2.739642567018684e-06, + "loss": 1.6652, + "step": 1404 + }, + { + "epoch": 1.5397260273972604, + "grad_norm": 2.7275257110595703, + "learning_rate": 2.7376116978066615e-06, + "loss": 1.6559, + "step": 1405 + }, + { + "epoch": 1.5408219178082192, + "grad_norm": 2.6349947452545166, + "learning_rate": 2.7355808285946388e-06, + "loss": 1.6489, + "step": 1406 + }, + { + "epoch": 1.541917808219178, + "grad_norm": 2.7058634757995605, + "learning_rate": 2.733549959382616e-06, + "loss": 1.7176, + "step": 1407 + }, + { + "epoch": 1.543013698630137, + "grad_norm": 2.481344223022461, + "learning_rate": 2.7315190901705933e-06, + "loss": 1.685, + "step": 1408 + }, + { + "epoch": 1.5441095890410959, + "grad_norm": 2.468277931213379, + "learning_rate": 2.7294882209585706e-06, + "loss": 1.6305, + "step": 1409 + }, + { + "epoch": 1.5452054794520547, + "grad_norm": 2.6183228492736816, + "learning_rate": 2.727457351746548e-06, + "loss": 1.6151, + "step": 1410 + }, + { + "epoch": 1.5452054794520547, + "eval_loss": 1.864329218864441, + "eval_runtime": 43.0443, + "eval_samples_per_second": 23.232, + "eval_steps_per_second": 0.976, + "step": 1410 + }, + { + "epoch": 1.5463013698630137, + "grad_norm": 2.50679349899292, + "learning_rate": 2.725426482534525e-06, + "loss": 1.6468, + "step": 1411 + }, + { + "epoch": 1.5473972602739727, + "grad_norm": 2.6565401554107666, + "learning_rate": 2.723395613322502e-06, + "loss": 1.6496, + "step": 1412 + }, + { + "epoch": 1.5484931506849315, + "grad_norm": 2.5332367420196533, + "learning_rate": 2.7213647441104797e-06, + "loss": 1.6424, + "step": 1413 + }, + { + "epoch": 1.5495890410958904, + "grad_norm": 2.6645619869232178, + "learning_rate": 2.7193338748984565e-06, + "loss": 1.8012, + "step": 1414 + }, + { + "epoch": 1.5506849315068494, + "grad_norm": 2.7359142303466797, + "learning_rate": 2.7173030056864342e-06, + "loss": 1.6025, + "step": 1415 + }, + { + "epoch": 1.5517808219178082, + "grad_norm": 2.660728693008423, + "learning_rate": 2.715272136474411e-06, + "loss": 1.6601, + "step": 1416 + }, + { + "epoch": 1.552876712328767, + "grad_norm": 2.5184221267700195, + "learning_rate": 2.713241267262389e-06, + "loss": 1.6383, + "step": 1417 + }, + { + "epoch": 1.553972602739726, + "grad_norm": 2.643712043762207, + "learning_rate": 2.7112103980503656e-06, + "loss": 1.6834, + "step": 1418 + }, + { + "epoch": 1.555068493150685, + "grad_norm": 2.504753351211548, + "learning_rate": 2.7091795288383433e-06, + "loss": 1.6292, + "step": 1419 + }, + { + "epoch": 1.5561643835616439, + "grad_norm": 2.3439948558807373, + "learning_rate": 2.70714865962632e-06, + "loss": 1.6558, + "step": 1420 + }, + { + "epoch": 1.5561643835616439, + "eval_loss": 1.861581563949585, + "eval_runtime": 43.9224, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 1420 + }, + { + "epoch": 1.5572602739726027, + "grad_norm": 2.4820125102996826, + "learning_rate": 2.705117790414298e-06, + "loss": 1.69, + "step": 1421 + }, + { + "epoch": 1.5583561643835617, + "grad_norm": 2.6156861782073975, + "learning_rate": 2.7030869212022747e-06, + "loss": 1.6304, + "step": 1422 + }, + { + "epoch": 1.5594520547945205, + "grad_norm": 2.490123987197876, + "learning_rate": 2.7010560519902516e-06, + "loss": 1.6573, + "step": 1423 + }, + { + "epoch": 1.5605479452054793, + "grad_norm": 2.4439024925231934, + "learning_rate": 2.6990251827782293e-06, + "loss": 1.6368, + "step": 1424 + }, + { + "epoch": 1.5616438356164384, + "grad_norm": 2.891603708267212, + "learning_rate": 2.696994313566206e-06, + "loss": 1.7111, + "step": 1425 + }, + { + "epoch": 1.5627397260273974, + "grad_norm": 2.481175661087036, + "learning_rate": 2.694963444354184e-06, + "loss": 1.6408, + "step": 1426 + }, + { + "epoch": 1.563835616438356, + "grad_norm": 2.5378847122192383, + "learning_rate": 2.6929325751421607e-06, + "loss": 1.6782, + "step": 1427 + }, + { + "epoch": 1.564931506849315, + "grad_norm": 2.592477321624756, + "learning_rate": 2.6909017059301384e-06, + "loss": 1.6085, + "step": 1428 + }, + { + "epoch": 1.566027397260274, + "grad_norm": 2.9336471557617188, + "learning_rate": 2.6888708367181153e-06, + "loss": 1.6719, + "step": 1429 + }, + { + "epoch": 1.5671232876712329, + "grad_norm": 2.3115012645721436, + "learning_rate": 2.686839967506093e-06, + "loss": 1.5797, + "step": 1430 + }, + { + "epoch": 1.5671232876712329, + "eval_loss": 1.8649041652679443, + "eval_runtime": 43.711, + "eval_samples_per_second": 22.878, + "eval_steps_per_second": 0.961, + "step": 1430 + }, + { + "epoch": 1.5682191780821917, + "grad_norm": 2.627028465270996, + "learning_rate": 2.68480909829407e-06, + "loss": 1.5784, + "step": 1431 + }, + { + "epoch": 1.5693150684931507, + "grad_norm": 2.917670488357544, + "learning_rate": 2.6827782290820475e-06, + "loss": 1.7007, + "step": 1432 + }, + { + "epoch": 1.5704109589041098, + "grad_norm": 2.418285608291626, + "learning_rate": 2.6807473598700244e-06, + "loss": 1.6613, + "step": 1433 + }, + { + "epoch": 1.5715068493150683, + "grad_norm": 2.607773780822754, + "learning_rate": 2.678716490658002e-06, + "loss": 1.6561, + "step": 1434 + }, + { + "epoch": 1.5726027397260274, + "grad_norm": 2.569714307785034, + "learning_rate": 2.676685621445979e-06, + "loss": 1.6541, + "step": 1435 + }, + { + "epoch": 1.5736986301369864, + "grad_norm": 2.359494209289551, + "learning_rate": 2.674654752233956e-06, + "loss": 1.6467, + "step": 1436 + }, + { + "epoch": 1.5747945205479452, + "grad_norm": 2.5155694484710693, + "learning_rate": 2.6726238830219335e-06, + "loss": 1.7003, + "step": 1437 + }, + { + "epoch": 1.575890410958904, + "grad_norm": 2.439147710800171, + "learning_rate": 2.6705930138099107e-06, + "loss": 1.5679, + "step": 1438 + }, + { + "epoch": 1.576986301369863, + "grad_norm": 2.5044116973876953, + "learning_rate": 2.668562144597888e-06, + "loss": 1.6626, + "step": 1439 + }, + { + "epoch": 1.5780821917808219, + "grad_norm": 2.4475765228271484, + "learning_rate": 2.6665312753858653e-06, + "loss": 1.656, + "step": 1440 + }, + { + "epoch": 1.5780821917808219, + "eval_loss": 1.8636877536773682, + "eval_runtime": 43.5138, + "eval_samples_per_second": 22.981, + "eval_steps_per_second": 0.965, + "step": 1440 + }, + { + "epoch": 1.5791780821917807, + "grad_norm": 2.4616832733154297, + "learning_rate": 2.6645004061738426e-06, + "loss": 1.6101, + "step": 1441 + }, + { + "epoch": 1.5802739726027397, + "grad_norm": 2.4125123023986816, + "learning_rate": 2.66246953696182e-06, + "loss": 1.6021, + "step": 1442 + }, + { + "epoch": 1.5813698630136988, + "grad_norm": 2.571326494216919, + "learning_rate": 2.660438667749797e-06, + "loss": 1.6482, + "step": 1443 + }, + { + "epoch": 1.5824657534246576, + "grad_norm": 2.424332618713379, + "learning_rate": 2.6584077985377744e-06, + "loss": 1.653, + "step": 1444 + }, + { + "epoch": 1.5835616438356164, + "grad_norm": 2.3034703731536865, + "learning_rate": 2.6563769293257517e-06, + "loss": 1.6096, + "step": 1445 + }, + { + "epoch": 1.5846575342465754, + "grad_norm": 2.4758505821228027, + "learning_rate": 2.654346060113729e-06, + "loss": 1.6274, + "step": 1446 + }, + { + "epoch": 1.5857534246575342, + "grad_norm": 3.800152540206909, + "learning_rate": 2.6523151909017062e-06, + "loss": 1.6217, + "step": 1447 + }, + { + "epoch": 1.586849315068493, + "grad_norm": 2.479827880859375, + "learning_rate": 2.6502843216896835e-06, + "loss": 1.7202, + "step": 1448 + }, + { + "epoch": 1.587945205479452, + "grad_norm": 2.604523181915283, + "learning_rate": 2.6482534524776603e-06, + "loss": 1.6778, + "step": 1449 + }, + { + "epoch": 1.589041095890411, + "grad_norm": 3.595973491668701, + "learning_rate": 2.646222583265638e-06, + "loss": 1.6678, + "step": 1450 + }, + { + "epoch": 1.589041095890411, + "eval_loss": 1.8647090196609497, + "eval_runtime": 43.0415, + "eval_samples_per_second": 23.233, + "eval_steps_per_second": 0.976, + "step": 1450 + }, + { + "epoch": 1.59013698630137, + "grad_norm": 3.297319173812866, + "learning_rate": 2.644191714053615e-06, + "loss": 1.637, + "step": 1451 + }, + { + "epoch": 1.5912328767123287, + "grad_norm": 2.423567056655884, + "learning_rate": 2.6421608448415926e-06, + "loss": 1.6018, + "step": 1452 + }, + { + "epoch": 1.5923287671232877, + "grad_norm": 4.4547953605651855, + "learning_rate": 2.6401299756295694e-06, + "loss": 1.6594, + "step": 1453 + }, + { + "epoch": 1.5934246575342466, + "grad_norm": 3.0775582790374756, + "learning_rate": 2.638099106417547e-06, + "loss": 1.6259, + "step": 1454 + }, + { + "epoch": 1.5945205479452054, + "grad_norm": 2.747938871383667, + "learning_rate": 2.636068237205524e-06, + "loss": 1.7208, + "step": 1455 + }, + { + "epoch": 1.5956164383561644, + "grad_norm": 2.5350255966186523, + "learning_rate": 2.6340373679935017e-06, + "loss": 1.5705, + "step": 1456 + }, + { + "epoch": 1.5967123287671234, + "grad_norm": 3.0630743503570557, + "learning_rate": 2.6320064987814785e-06, + "loss": 1.6507, + "step": 1457 + }, + { + "epoch": 1.5978082191780822, + "grad_norm": 3.5475759506225586, + "learning_rate": 2.6299756295694562e-06, + "loss": 1.7045, + "step": 1458 + }, + { + "epoch": 1.598904109589041, + "grad_norm": 2.534303903579712, + "learning_rate": 2.627944760357433e-06, + "loss": 1.7173, + "step": 1459 + }, + { + "epoch": 1.6, + "grad_norm": 2.6964452266693115, + "learning_rate": 2.6259138911454104e-06, + "loss": 1.6823, + "step": 1460 + }, + { + "epoch": 1.6, + "eval_loss": 1.865207552909851, + "eval_runtime": 42.9435, + "eval_samples_per_second": 23.286, + "eval_steps_per_second": 0.978, + "step": 1460 + }, + { + "epoch": 1.601095890410959, + "grad_norm": 2.924957752227783, + "learning_rate": 2.6238830219333877e-06, + "loss": 1.6036, + "step": 1461 + }, + { + "epoch": 1.6021917808219177, + "grad_norm": 2.7058141231536865, + "learning_rate": 2.621852152721365e-06, + "loss": 1.6701, + "step": 1462 + }, + { + "epoch": 1.6032876712328767, + "grad_norm": 2.735339641571045, + "learning_rate": 2.619821283509342e-06, + "loss": 1.607, + "step": 1463 + }, + { + "epoch": 1.6043835616438358, + "grad_norm": 2.573958396911621, + "learning_rate": 2.6177904142973195e-06, + "loss": 1.667, + "step": 1464 + }, + { + "epoch": 1.6054794520547944, + "grad_norm": 2.5776426792144775, + "learning_rate": 2.6157595450852968e-06, + "loss": 1.6938, + "step": 1465 + }, + { + "epoch": 1.6065753424657534, + "grad_norm": 4.454444408416748, + "learning_rate": 2.613728675873274e-06, + "loss": 1.7139, + "step": 1466 + }, + { + "epoch": 1.6076712328767124, + "grad_norm": 2.4883337020874023, + "learning_rate": 2.6116978066612513e-06, + "loss": 1.6524, + "step": 1467 + }, + { + "epoch": 1.6087671232876712, + "grad_norm": 2.672116994857788, + "learning_rate": 2.6096669374492286e-06, + "loss": 1.6197, + "step": 1468 + }, + { + "epoch": 1.60986301369863, + "grad_norm": 3.511932611465454, + "learning_rate": 2.607636068237206e-06, + "loss": 1.6739, + "step": 1469 + }, + { + "epoch": 1.610958904109589, + "grad_norm": 2.5960047245025635, + "learning_rate": 2.605605199025183e-06, + "loss": 1.6516, + "step": 1470 + }, + { + "epoch": 1.610958904109589, + "eval_loss": 1.8604435920715332, + "eval_runtime": 42.9247, + "eval_samples_per_second": 23.297, + "eval_steps_per_second": 0.978, + "step": 1470 + }, + { + "epoch": 1.612054794520548, + "grad_norm": 2.565376043319702, + "learning_rate": 2.6035743298131604e-06, + "loss": 1.6884, + "step": 1471 + }, + { + "epoch": 1.6131506849315067, + "grad_norm": 2.676114797592163, + "learning_rate": 2.6015434606011377e-06, + "loss": 1.6693, + "step": 1472 + }, + { + "epoch": 1.6142465753424657, + "grad_norm": 2.582751750946045, + "learning_rate": 2.5995125913891145e-06, + "loss": 1.6499, + "step": 1473 + }, + { + "epoch": 1.6153424657534248, + "grad_norm": 2.748908281326294, + "learning_rate": 2.5974817221770922e-06, + "loss": 1.6837, + "step": 1474 + }, + { + "epoch": 1.6164383561643836, + "grad_norm": 2.5578067302703857, + "learning_rate": 2.595450852965069e-06, + "loss": 1.6959, + "step": 1475 + }, + { + "epoch": 1.6175342465753424, + "grad_norm": 2.5702805519104004, + "learning_rate": 2.593419983753047e-06, + "loss": 1.6429, + "step": 1476 + }, + { + "epoch": 1.6186301369863014, + "grad_norm": 2.4748787879943848, + "learning_rate": 2.5913891145410236e-06, + "loss": 1.6878, + "step": 1477 + }, + { + "epoch": 1.6197260273972602, + "grad_norm": 2.6157331466674805, + "learning_rate": 2.5893582453290013e-06, + "loss": 1.6246, + "step": 1478 + }, + { + "epoch": 1.620821917808219, + "grad_norm": 2.8377959728240967, + "learning_rate": 2.587327376116978e-06, + "loss": 1.6451, + "step": 1479 + }, + { + "epoch": 1.621917808219178, + "grad_norm": 2.485398292541504, + "learning_rate": 2.585296506904956e-06, + "loss": 1.6288, + "step": 1480 + }, + { + "epoch": 1.621917808219178, + "eval_loss": 1.8593864440917969, + "eval_runtime": 44.6341, + "eval_samples_per_second": 22.404, + "eval_steps_per_second": 0.941, + "step": 1480 + }, + { + "epoch": 1.6230136986301371, + "grad_norm": 2.46850323677063, + "learning_rate": 2.5832656376929327e-06, + "loss": 1.5766, + "step": 1481 + }, + { + "epoch": 1.624109589041096, + "grad_norm": 2.673283576965332, + "learning_rate": 2.5812347684809104e-06, + "loss": 1.7383, + "step": 1482 + }, + { + "epoch": 1.6252054794520547, + "grad_norm": 2.935293197631836, + "learning_rate": 2.5792038992688873e-06, + "loss": 1.5742, + "step": 1483 + }, + { + "epoch": 1.6263013698630138, + "grad_norm": 2.531733751296997, + "learning_rate": 2.577173030056865e-06, + "loss": 1.6714, + "step": 1484 + }, + { + "epoch": 1.6273972602739726, + "grad_norm": 2.528425455093384, + "learning_rate": 2.575142160844842e-06, + "loss": 1.6802, + "step": 1485 + }, + { + "epoch": 1.6284931506849314, + "grad_norm": 2.5644803047180176, + "learning_rate": 2.5731112916328187e-06, + "loss": 1.6201, + "step": 1486 + }, + { + "epoch": 1.6295890410958904, + "grad_norm": 2.382451295852661, + "learning_rate": 2.5710804224207964e-06, + "loss": 1.5824, + "step": 1487 + }, + { + "epoch": 1.6306849315068495, + "grad_norm": 2.63320255279541, + "learning_rate": 2.5690495532087733e-06, + "loss": 1.6685, + "step": 1488 + }, + { + "epoch": 1.6317808219178083, + "grad_norm": 2.488844394683838, + "learning_rate": 2.567018683996751e-06, + "loss": 1.6078, + "step": 1489 + }, + { + "epoch": 1.632876712328767, + "grad_norm": 2.7845425605773926, + "learning_rate": 2.564987814784728e-06, + "loss": 1.638, + "step": 1490 + }, + { + "epoch": 1.632876712328767, + "eval_loss": 1.8572624921798706, + "eval_runtime": 43.2181, + "eval_samples_per_second": 23.138, + "eval_steps_per_second": 0.972, + "step": 1490 + }, + { + "epoch": 1.633972602739726, + "grad_norm": 2.709656000137329, + "learning_rate": 2.5629569455727055e-06, + "loss": 1.6425, + "step": 1491 + }, + { + "epoch": 1.635068493150685, + "grad_norm": 2.651277542114258, + "learning_rate": 2.5609260763606824e-06, + "loss": 1.7401, + "step": 1492 + }, + { + "epoch": 1.6361643835616437, + "grad_norm": 2.5424256324768066, + "learning_rate": 2.55889520714866e-06, + "loss": 1.6272, + "step": 1493 + }, + { + "epoch": 1.6372602739726028, + "grad_norm": 2.6508235931396484, + "learning_rate": 2.556864337936637e-06, + "loss": 1.6384, + "step": 1494 + }, + { + "epoch": 1.6383561643835618, + "grad_norm": 2.882636308670044, + "learning_rate": 2.5548334687246146e-06, + "loss": 1.6965, + "step": 1495 + }, + { + "epoch": 1.6394520547945204, + "grad_norm": 2.518367290496826, + "learning_rate": 2.5528025995125915e-06, + "loss": 1.6551, + "step": 1496 + }, + { + "epoch": 1.6405479452054794, + "grad_norm": 2.7977044582366943, + "learning_rate": 2.550771730300569e-06, + "loss": 1.679, + "step": 1497 + }, + { + "epoch": 1.6416438356164385, + "grad_norm": 2.353041172027588, + "learning_rate": 2.548740861088546e-06, + "loss": 1.5745, + "step": 1498 + }, + { + "epoch": 1.6427397260273973, + "grad_norm": 2.5718655586242676, + "learning_rate": 2.5467099918765233e-06, + "loss": 1.5655, + "step": 1499 + }, + { + "epoch": 1.643835616438356, + "grad_norm": 2.6955313682556152, + "learning_rate": 2.5446791226645006e-06, + "loss": 1.6764, + "step": 1500 + }, + { + "epoch": 1.643835616438356, + "eval_loss": 1.8569272756576538, + "eval_runtime": 45.3844, + "eval_samples_per_second": 22.034, + "eval_steps_per_second": 0.925, + "step": 1500 + }, + { + "epoch": 1.644931506849315, + "grad_norm": 2.4756696224212646, + "learning_rate": 2.542648253452478e-06, + "loss": 1.6522, + "step": 1501 + }, + { + "epoch": 1.6460273972602741, + "grad_norm": 2.573193311691284, + "learning_rate": 2.540617384240455e-06, + "loss": 1.6079, + "step": 1502 + }, + { + "epoch": 1.6471232876712327, + "grad_norm": 2.5967671871185303, + "learning_rate": 2.5385865150284324e-06, + "loss": 1.6341, + "step": 1503 + }, + { + "epoch": 1.6482191780821918, + "grad_norm": 2.7896342277526855, + "learning_rate": 2.5365556458164097e-06, + "loss": 1.6449, + "step": 1504 + }, + { + "epoch": 1.6493150684931508, + "grad_norm": 2.519728183746338, + "learning_rate": 2.534524776604387e-06, + "loss": 1.549, + "step": 1505 + }, + { + "epoch": 1.6504109589041096, + "grad_norm": 2.543139934539795, + "learning_rate": 2.5324939073923642e-06, + "loss": 1.6335, + "step": 1506 + }, + { + "epoch": 1.6515068493150684, + "grad_norm": 2.5580177307128906, + "learning_rate": 2.5304630381803415e-06, + "loss": 1.6402, + "step": 1507 + }, + { + "epoch": 1.6526027397260274, + "grad_norm": 2.443547010421753, + "learning_rate": 2.5284321689683188e-06, + "loss": 1.6421, + "step": 1508 + }, + { + "epoch": 1.6536986301369863, + "grad_norm": 2.7075212001800537, + "learning_rate": 2.526401299756296e-06, + "loss": 1.6283, + "step": 1509 + }, + { + "epoch": 1.654794520547945, + "grad_norm": 2.6505324840545654, + "learning_rate": 2.524370430544273e-06, + "loss": 1.6028, + "step": 1510 + }, + { + "epoch": 1.654794520547945, + "eval_loss": 1.8556184768676758, + "eval_runtime": 43.731, + "eval_samples_per_second": 22.867, + "eval_steps_per_second": 0.96, + "step": 1510 + }, + { + "epoch": 1.655890410958904, + "grad_norm": 2.597172260284424, + "learning_rate": 2.5223395613322506e-06, + "loss": 1.6631, + "step": 1511 + }, + { + "epoch": 1.6569863013698631, + "grad_norm": 3.1529037952423096, + "learning_rate": 2.5203086921202274e-06, + "loss": 1.6219, + "step": 1512 + }, + { + "epoch": 1.658082191780822, + "grad_norm": 2.5374975204467773, + "learning_rate": 2.518277822908205e-06, + "loss": 1.6476, + "step": 1513 + }, + { + "epoch": 1.6591780821917808, + "grad_norm": 3.0557401180267334, + "learning_rate": 2.516246953696182e-06, + "loss": 1.6627, + "step": 1514 + }, + { + "epoch": 1.6602739726027398, + "grad_norm": 6.759350776672363, + "learning_rate": 2.5142160844841597e-06, + "loss": 1.6532, + "step": 1515 + }, + { + "epoch": 1.6613698630136986, + "grad_norm": 2.670427083969116, + "learning_rate": 2.5121852152721365e-06, + "loss": 1.658, + "step": 1516 + }, + { + "epoch": 1.6624657534246574, + "grad_norm": 4.269686698913574, + "learning_rate": 2.5101543460601142e-06, + "loss": 1.5873, + "step": 1517 + }, + { + "epoch": 1.6635616438356164, + "grad_norm": 13.37558650970459, + "learning_rate": 2.508123476848091e-06, + "loss": 1.6872, + "step": 1518 + }, + { + "epoch": 1.6646575342465755, + "grad_norm": 7.060877323150635, + "learning_rate": 2.506092607636069e-06, + "loss": 1.6406, + "step": 1519 + }, + { + "epoch": 1.6657534246575343, + "grad_norm": 11.259824752807617, + "learning_rate": 2.5040617384240457e-06, + "loss": 1.6421, + "step": 1520 + }, + { + "epoch": 1.6657534246575343, + "eval_loss": 1.861360788345337, + "eval_runtime": 45.4438, + "eval_samples_per_second": 22.005, + "eval_steps_per_second": 0.924, + "step": 1520 + }, + { + "epoch": 1.666849315068493, + "grad_norm": 5.861843585968018, + "learning_rate": 2.5020308692120234e-06, + "loss": 1.701, + "step": 1521 + }, + { + "epoch": 1.6679452054794521, + "grad_norm": 3.9136719703674316, + "learning_rate": 2.5e-06, + "loss": 1.6289, + "step": 1522 + }, + { + "epoch": 1.669041095890411, + "grad_norm": 12.720739364624023, + "learning_rate": 2.4979691307879775e-06, + "loss": 1.6532, + "step": 1523 + }, + { + "epoch": 1.6701369863013698, + "grad_norm": 5.970798492431641, + "learning_rate": 2.4959382615759548e-06, + "loss": 1.7035, + "step": 1524 + }, + { + "epoch": 1.6712328767123288, + "grad_norm": 3.490950584411621, + "learning_rate": 2.493907392363932e-06, + "loss": 1.6565, + "step": 1525 + }, + { + "epoch": 1.6723287671232878, + "grad_norm": 5.826146602630615, + "learning_rate": 2.4918765231519093e-06, + "loss": 1.655, + "step": 1526 + }, + { + "epoch": 1.6734246575342466, + "grad_norm": 9.610876083374023, + "learning_rate": 2.4898456539398866e-06, + "loss": 1.7295, + "step": 1527 + }, + { + "epoch": 1.6745205479452054, + "grad_norm": 3.754833221435547, + "learning_rate": 2.487814784727864e-06, + "loss": 1.6406, + "step": 1528 + }, + { + "epoch": 1.6756164383561645, + "grad_norm": 2.7952232360839844, + "learning_rate": 2.4857839155158407e-06, + "loss": 1.6463, + "step": 1529 + }, + { + "epoch": 1.6767123287671233, + "grad_norm": 6.3178558349609375, + "learning_rate": 2.483753046303818e-06, + "loss": 1.6474, + "step": 1530 + }, + { + "epoch": 1.6767123287671233, + "eval_loss": 1.8672106266021729, + "eval_runtime": 50.0163, + "eval_samples_per_second": 19.993, + "eval_steps_per_second": 0.84, + "step": 1530 + }, + { + "epoch": 1.677808219178082, + "grad_norm": 7.4854960441589355, + "learning_rate": 2.4817221770917953e-06, + "loss": 1.6798, + "step": 1531 + }, + { + "epoch": 1.6789041095890411, + "grad_norm": 2.8903026580810547, + "learning_rate": 2.4796913078797725e-06, + "loss": 1.667, + "step": 1532 + }, + { + "epoch": 1.6800000000000002, + "grad_norm": 2.481947898864746, + "learning_rate": 2.47766043866775e-06, + "loss": 1.6677, + "step": 1533 + }, + { + "epoch": 1.6810958904109587, + "grad_norm": 2.8459818363189697, + "learning_rate": 2.475629569455727e-06, + "loss": 1.6129, + "step": 1534 + }, + { + "epoch": 1.6821917808219178, + "grad_norm": 5.00568962097168, + "learning_rate": 2.4735987002437044e-06, + "loss": 1.6233, + "step": 1535 + }, + { + "epoch": 1.6832876712328768, + "grad_norm": 3.7508153915405273, + "learning_rate": 2.4715678310316816e-06, + "loss": 1.6414, + "step": 1536 + }, + { + "epoch": 1.6843835616438356, + "grad_norm": 2.7895889282226562, + "learning_rate": 2.469536961819659e-06, + "loss": 1.65, + "step": 1537 + }, + { + "epoch": 1.6854794520547944, + "grad_norm": 2.532818555831909, + "learning_rate": 2.467506092607636e-06, + "loss": 1.6174, + "step": 1538 + }, + { + "epoch": 1.6865753424657535, + "grad_norm": 2.699639320373535, + "learning_rate": 2.4654752233956135e-06, + "loss": 1.7128, + "step": 1539 + }, + { + "epoch": 1.6876712328767123, + "grad_norm": 2.5636072158813477, + "learning_rate": 2.4634443541835907e-06, + "loss": 1.6416, + "step": 1540 + }, + { + "epoch": 1.6876712328767123, + "eval_loss": 1.8608473539352417, + "eval_runtime": 43.8315, + "eval_samples_per_second": 22.815, + "eval_steps_per_second": 0.958, + "step": 1540 + }, + { + "epoch": 1.688767123287671, + "grad_norm": 2.7466976642608643, + "learning_rate": 2.461413484971568e-06, + "loss": 1.7246, + "step": 1541 + }, + { + "epoch": 1.6898630136986301, + "grad_norm": 2.690990686416626, + "learning_rate": 2.4593826157595453e-06, + "loss": 1.596, + "step": 1542 + }, + { + "epoch": 1.6909589041095892, + "grad_norm": 2.868340492248535, + "learning_rate": 2.4573517465475226e-06, + "loss": 1.622, + "step": 1543 + }, + { + "epoch": 1.692054794520548, + "grad_norm": 2.4456427097320557, + "learning_rate": 2.4553208773355e-06, + "loss": 1.6446, + "step": 1544 + }, + { + "epoch": 1.6931506849315068, + "grad_norm": 2.5499041080474854, + "learning_rate": 2.453290008123477e-06, + "loss": 1.6587, + "step": 1545 + }, + { + "epoch": 1.6942465753424658, + "grad_norm": 2.68770170211792, + "learning_rate": 2.4512591389114544e-06, + "loss": 1.6612, + "step": 1546 + }, + { + "epoch": 1.6953424657534246, + "grad_norm": 2.589557647705078, + "learning_rate": 2.4492282696994317e-06, + "loss": 1.6428, + "step": 1547 + }, + { + "epoch": 1.6964383561643834, + "grad_norm": 2.7014052867889404, + "learning_rate": 2.447197400487409e-06, + "loss": 1.6605, + "step": 1548 + }, + { + "epoch": 1.6975342465753425, + "grad_norm": 2.6242196559906006, + "learning_rate": 2.4451665312753862e-06, + "loss": 1.6897, + "step": 1549 + }, + { + "epoch": 1.6986301369863015, + "grad_norm": 2.5285110473632812, + "learning_rate": 2.4431356620633635e-06, + "loss": 1.6636, + "step": 1550 + }, + { + "epoch": 1.6986301369863015, + "eval_loss": 1.8544306755065918, + "eval_runtime": 45.6109, + "eval_samples_per_second": 21.925, + "eval_steps_per_second": 0.921, + "step": 1550 + }, + { + "epoch": 1.6997260273972603, + "grad_norm": 2.484952449798584, + "learning_rate": 2.4411047928513408e-06, + "loss": 1.6636, + "step": 1551 + }, + { + "epoch": 1.7008219178082191, + "grad_norm": 2.5793070793151855, + "learning_rate": 2.439073923639318e-06, + "loss": 1.6044, + "step": 1552 + }, + { + "epoch": 1.7019178082191782, + "grad_norm": 2.5558605194091797, + "learning_rate": 2.4370430544272953e-06, + "loss": 1.5885, + "step": 1553 + }, + { + "epoch": 1.703013698630137, + "grad_norm": 2.6365678310394287, + "learning_rate": 2.435012185215272e-06, + "loss": 1.6236, + "step": 1554 + }, + { + "epoch": 1.7041095890410958, + "grad_norm": 2.428422451019287, + "learning_rate": 2.4329813160032495e-06, + "loss": 1.6446, + "step": 1555 + }, + { + "epoch": 1.7052054794520548, + "grad_norm": 2.535776138305664, + "learning_rate": 2.4309504467912267e-06, + "loss": 1.6252, + "step": 1556 + }, + { + "epoch": 1.7063013698630138, + "grad_norm": 2.743934154510498, + "learning_rate": 2.428919577579204e-06, + "loss": 1.729, + "step": 1557 + }, + { + "epoch": 1.7073972602739727, + "grad_norm": 2.4823739528656006, + "learning_rate": 2.4268887083671813e-06, + "loss": 1.6251, + "step": 1558 + }, + { + "epoch": 1.7084931506849315, + "grad_norm": 2.4650003910064697, + "learning_rate": 2.4248578391551586e-06, + "loss": 1.6191, + "step": 1559 + }, + { + "epoch": 1.7095890410958905, + "grad_norm": 2.4338109493255615, + "learning_rate": 2.422826969943136e-06, + "loss": 1.5919, + "step": 1560 + }, + { + "epoch": 1.7095890410958905, + "eval_loss": 1.8501240015029907, + "eval_runtime": 43.8558, + "eval_samples_per_second": 22.802, + "eval_steps_per_second": 0.958, + "step": 1560 + }, + { + "epoch": 1.7106849315068493, + "grad_norm": 2.6054813861846924, + "learning_rate": 2.420796100731113e-06, + "loss": 1.6485, + "step": 1561 + }, + { + "epoch": 1.7117808219178081, + "grad_norm": 2.390143394470215, + "learning_rate": 2.4187652315190904e-06, + "loss": 1.661, + "step": 1562 + }, + { + "epoch": 1.7128767123287671, + "grad_norm": 2.467470645904541, + "learning_rate": 2.4167343623070677e-06, + "loss": 1.6308, + "step": 1563 + }, + { + "epoch": 1.7139726027397262, + "grad_norm": 2.360908269882202, + "learning_rate": 2.414703493095045e-06, + "loss": 1.6394, + "step": 1564 + }, + { + "epoch": 1.7150684931506848, + "grad_norm": 2.5966176986694336, + "learning_rate": 2.4126726238830222e-06, + "loss": 1.6138, + "step": 1565 + }, + { + "epoch": 1.7161643835616438, + "grad_norm": 2.528402328491211, + "learning_rate": 2.410641754670999e-06, + "loss": 1.6817, + "step": 1566 + }, + { + "epoch": 1.7172602739726028, + "grad_norm": 2.5319254398345947, + "learning_rate": 2.4086108854589763e-06, + "loss": 1.6189, + "step": 1567 + }, + { + "epoch": 1.7183561643835616, + "grad_norm": 2.559260129928589, + "learning_rate": 2.4065800162469536e-06, + "loss": 1.6476, + "step": 1568 + }, + { + "epoch": 1.7194520547945205, + "grad_norm": 2.5595016479492188, + "learning_rate": 2.404549147034931e-06, + "loss": 1.6878, + "step": 1569 + }, + { + "epoch": 1.7205479452054795, + "grad_norm": 2.488196849822998, + "learning_rate": 2.402518277822908e-06, + "loss": 1.6095, + "step": 1570 + }, + { + "epoch": 1.7205479452054795, + "eval_loss": 1.8484610319137573, + "eval_runtime": 43.4829, + "eval_samples_per_second": 22.998, + "eval_steps_per_second": 0.966, + "step": 1570 + }, + { + "epoch": 1.7216438356164385, + "grad_norm": 2.6928117275238037, + "learning_rate": 2.4004874086108854e-06, + "loss": 1.6533, + "step": 1571 + }, + { + "epoch": 1.7227397260273971, + "grad_norm": 2.485538959503174, + "learning_rate": 2.3984565393988627e-06, + "loss": 1.6174, + "step": 1572 + }, + { + "epoch": 1.7238356164383561, + "grad_norm": 2.4725208282470703, + "learning_rate": 2.39642567018684e-06, + "loss": 1.6306, + "step": 1573 + }, + { + "epoch": 1.7249315068493152, + "grad_norm": 2.456028461456299, + "learning_rate": 2.3943948009748173e-06, + "loss": 1.6586, + "step": 1574 + }, + { + "epoch": 1.726027397260274, + "grad_norm": 2.3731250762939453, + "learning_rate": 2.3923639317627945e-06, + "loss": 1.6009, + "step": 1575 + }, + { + "epoch": 1.7271232876712328, + "grad_norm": 2.4567086696624756, + "learning_rate": 2.390333062550772e-06, + "loss": 1.5257, + "step": 1576 + }, + { + "epoch": 1.7282191780821918, + "grad_norm": 2.577141284942627, + "learning_rate": 2.388302193338749e-06, + "loss": 1.7127, + "step": 1577 + }, + { + "epoch": 1.7293150684931506, + "grad_norm": 2.4612233638763428, + "learning_rate": 2.3862713241267264e-06, + "loss": 1.638, + "step": 1578 + }, + { + "epoch": 1.7304109589041095, + "grad_norm": 2.463212728500366, + "learning_rate": 2.3842404549147037e-06, + "loss": 1.5933, + "step": 1579 + }, + { + "epoch": 1.7315068493150685, + "grad_norm": 2.5121495723724365, + "learning_rate": 2.382209585702681e-06, + "loss": 1.68, + "step": 1580 + }, + { + "epoch": 1.7315068493150685, + "eval_loss": 1.8499811887741089, + "eval_runtime": 45.066, + "eval_samples_per_second": 22.19, + "eval_steps_per_second": 0.932, + "step": 1580 + }, + { + "epoch": 1.7326027397260275, + "grad_norm": 2.658245801925659, + "learning_rate": 2.380178716490658e-06, + "loss": 1.6354, + "step": 1581 + }, + { + "epoch": 1.7336986301369863, + "grad_norm": 2.6265509128570557, + "learning_rate": 2.3781478472786355e-06, + "loss": 1.6755, + "step": 1582 + }, + { + "epoch": 1.7347945205479451, + "grad_norm": 3.016807794570923, + "learning_rate": 2.3761169780666128e-06, + "loss": 1.6764, + "step": 1583 + }, + { + "epoch": 1.7358904109589042, + "grad_norm": 2.6552720069885254, + "learning_rate": 2.37408610885459e-06, + "loss": 1.6281, + "step": 1584 + }, + { + "epoch": 1.736986301369863, + "grad_norm": 2.520078659057617, + "learning_rate": 2.3720552396425673e-06, + "loss": 1.7077, + "step": 1585 + }, + { + "epoch": 1.7380821917808218, + "grad_norm": 2.528329849243164, + "learning_rate": 2.3700243704305446e-06, + "loss": 1.6494, + "step": 1586 + }, + { + "epoch": 1.7391780821917808, + "grad_norm": 2.623776912689209, + "learning_rate": 2.367993501218522e-06, + "loss": 1.6436, + "step": 1587 + }, + { + "epoch": 1.7402739726027399, + "grad_norm": 2.56669020652771, + "learning_rate": 2.365962632006499e-06, + "loss": 1.6762, + "step": 1588 + }, + { + "epoch": 1.7413698630136987, + "grad_norm": 2.5415968894958496, + "learning_rate": 2.3639317627944764e-06, + "loss": 1.6711, + "step": 1589 + }, + { + "epoch": 1.7424657534246575, + "grad_norm": 2.5164198875427246, + "learning_rate": 2.3619008935824537e-06, + "loss": 1.6725, + "step": 1590 + }, + { + "epoch": 1.7424657534246575, + "eval_loss": 1.8490949869155884, + "eval_runtime": 43.612, + "eval_samples_per_second": 22.929, + "eval_steps_per_second": 0.963, + "step": 1590 + }, + { + "epoch": 1.7435616438356165, + "grad_norm": 2.626594066619873, + "learning_rate": 2.3598700243704305e-06, + "loss": 1.6637, + "step": 1591 + }, + { + "epoch": 1.7446575342465753, + "grad_norm": 2.3480069637298584, + "learning_rate": 2.357839155158408e-06, + "loss": 1.6561, + "step": 1592 + }, + { + "epoch": 1.7457534246575341, + "grad_norm": 2.644103527069092, + "learning_rate": 2.355808285946385e-06, + "loss": 1.6763, + "step": 1593 + }, + { + "epoch": 1.7468493150684932, + "grad_norm": 2.5539023876190186, + "learning_rate": 2.3537774167343624e-06, + "loss": 1.653, + "step": 1594 + }, + { + "epoch": 1.7479452054794522, + "grad_norm": 2.5055317878723145, + "learning_rate": 2.3517465475223396e-06, + "loss": 1.6643, + "step": 1595 + }, + { + "epoch": 1.749041095890411, + "grad_norm": 2.4204092025756836, + "learning_rate": 2.349715678310317e-06, + "loss": 1.6133, + "step": 1596 + }, + { + "epoch": 1.7501369863013698, + "grad_norm": 2.5867843627929688, + "learning_rate": 2.347684809098294e-06, + "loss": 1.6804, + "step": 1597 + }, + { + "epoch": 1.7512328767123289, + "grad_norm": 2.5993685722351074, + "learning_rate": 2.3456539398862715e-06, + "loss": 1.6101, + "step": 1598 + }, + { + "epoch": 1.7523287671232877, + "grad_norm": 2.565439462661743, + "learning_rate": 2.3436230706742487e-06, + "loss": 1.6221, + "step": 1599 + }, + { + "epoch": 1.7534246575342465, + "grad_norm": 2.5296084880828857, + "learning_rate": 2.341592201462226e-06, + "loss": 1.6234, + "step": 1600 + }, + { + "epoch": 1.7534246575342465, + "eval_loss": 1.8482695817947388, + "eval_runtime": 45.1972, + "eval_samples_per_second": 22.125, + "eval_steps_per_second": 0.929, + "step": 1600 + }, + { + "epoch": 1.7545205479452055, + "grad_norm": 2.6142499446868896, + "learning_rate": 2.3395613322502033e-06, + "loss": 1.5886, + "step": 1601 + }, + { + "epoch": 1.7556164383561645, + "grad_norm": 2.486616849899292, + "learning_rate": 2.3375304630381806e-06, + "loss": 1.6097, + "step": 1602 + }, + { + "epoch": 1.7567123287671231, + "grad_norm": 2.5778701305389404, + "learning_rate": 2.335499593826158e-06, + "loss": 1.6585, + "step": 1603 + }, + { + "epoch": 1.7578082191780822, + "grad_norm": 2.6165313720703125, + "learning_rate": 2.333468724614135e-06, + "loss": 1.663, + "step": 1604 + }, + { + "epoch": 1.7589041095890412, + "grad_norm": 2.4976565837860107, + "learning_rate": 2.3314378554021124e-06, + "loss": 1.6083, + "step": 1605 + }, + { + "epoch": 1.76, + "grad_norm": 2.485017776489258, + "learning_rate": 2.3294069861900897e-06, + "loss": 1.6006, + "step": 1606 + }, + { + "epoch": 1.7610958904109588, + "grad_norm": 2.620819330215454, + "learning_rate": 2.327376116978067e-06, + "loss": 1.6641, + "step": 1607 + }, + { + "epoch": 1.7621917808219179, + "grad_norm": 2.528599262237549, + "learning_rate": 2.3253452477660442e-06, + "loss": 1.6128, + "step": 1608 + }, + { + "epoch": 1.7632876712328767, + "grad_norm": 2.541442632675171, + "learning_rate": 2.3233143785540215e-06, + "loss": 1.6545, + "step": 1609 + }, + { + "epoch": 1.7643835616438355, + "grad_norm": 2.4722139835357666, + "learning_rate": 2.3212835093419988e-06, + "loss": 1.6146, + "step": 1610 + }, + { + "epoch": 1.7643835616438355, + "eval_loss": 1.848878264427185, + "eval_runtime": 43.7961, + "eval_samples_per_second": 22.833, + "eval_steps_per_second": 0.959, + "step": 1610 + }, + { + "epoch": 1.7654794520547945, + "grad_norm": 2.6106903553009033, + "learning_rate": 2.319252640129976e-06, + "loss": 1.7086, + "step": 1611 + }, + { + "epoch": 1.7665753424657535, + "grad_norm": 2.5255045890808105, + "learning_rate": 2.3172217709179533e-06, + "loss": 1.6738, + "step": 1612 + }, + { + "epoch": 1.7676712328767124, + "grad_norm": 2.43829083442688, + "learning_rate": 2.3151909017059306e-06, + "loss": 1.5861, + "step": 1613 + }, + { + "epoch": 1.7687671232876712, + "grad_norm": 2.527411460876465, + "learning_rate": 2.313160032493908e-06, + "loss": 1.6634, + "step": 1614 + }, + { + "epoch": 1.7698630136986302, + "grad_norm": 2.4998528957366943, + "learning_rate": 2.311129163281885e-06, + "loss": 1.6187, + "step": 1615 + }, + { + "epoch": 1.770958904109589, + "grad_norm": 2.4838368892669678, + "learning_rate": 2.309098294069862e-06, + "loss": 1.5999, + "step": 1616 + }, + { + "epoch": 1.7720547945205478, + "grad_norm": 2.466900110244751, + "learning_rate": 2.3070674248578393e-06, + "loss": 1.6446, + "step": 1617 + }, + { + "epoch": 1.7731506849315068, + "grad_norm": 2.583103895187378, + "learning_rate": 2.3050365556458166e-06, + "loss": 1.6658, + "step": 1618 + }, + { + "epoch": 1.7742465753424659, + "grad_norm": 2.478386163711548, + "learning_rate": 2.303005686433794e-06, + "loss": 1.6261, + "step": 1619 + }, + { + "epoch": 1.7753424657534247, + "grad_norm": 2.5724942684173584, + "learning_rate": 2.300974817221771e-06, + "loss": 1.5955, + "step": 1620 + }, + { + "epoch": 1.7753424657534247, + "eval_loss": 1.8444501161575317, + "eval_runtime": 50.0263, + "eval_samples_per_second": 19.989, + "eval_steps_per_second": 0.84, + "step": 1620 + }, + { + "epoch": 1.7764383561643835, + "grad_norm": 2.601644277572632, + "learning_rate": 2.2989439480097484e-06, + "loss": 1.6493, + "step": 1621 + }, + { + "epoch": 1.7775342465753425, + "grad_norm": 2.5638768672943115, + "learning_rate": 2.2969130787977257e-06, + "loss": 1.6702, + "step": 1622 + }, + { + "epoch": 1.7786301369863013, + "grad_norm": 2.4802567958831787, + "learning_rate": 2.294882209585703e-06, + "loss": 1.6036, + "step": 1623 + }, + { + "epoch": 1.7797260273972602, + "grad_norm": 2.5114285945892334, + "learning_rate": 2.2928513403736802e-06, + "loss": 1.642, + "step": 1624 + }, + { + "epoch": 1.7808219178082192, + "grad_norm": 2.487563133239746, + "learning_rate": 2.2908204711616575e-06, + "loss": 1.673, + "step": 1625 + }, + { + "epoch": 1.7819178082191782, + "grad_norm": 2.6014463901519775, + "learning_rate": 2.2887896019496348e-06, + "loss": 1.6261, + "step": 1626 + }, + { + "epoch": 1.783013698630137, + "grad_norm": 2.53680157661438, + "learning_rate": 2.286758732737612e-06, + "loss": 1.6364, + "step": 1627 + }, + { + "epoch": 1.7841095890410958, + "grad_norm": 2.764408588409424, + "learning_rate": 2.284727863525589e-06, + "loss": 1.6791, + "step": 1628 + }, + { + "epoch": 1.7852054794520549, + "grad_norm": 2.8998825550079346, + "learning_rate": 2.282696994313566e-06, + "loss": 1.644, + "step": 1629 + }, + { + "epoch": 1.7863013698630137, + "grad_norm": 2.5797383785247803, + "learning_rate": 2.2806661251015434e-06, + "loss": 1.6267, + "step": 1630 + }, + { + "epoch": 1.7863013698630137, + "eval_loss": 1.8467868566513062, + "eval_runtime": 49.4778, + "eval_samples_per_second": 20.211, + "eval_steps_per_second": 0.849, + "step": 1630 + }, + { + "epoch": 1.7873972602739725, + "grad_norm": 2.447178602218628, + "learning_rate": 2.2786352558895207e-06, + "loss": 1.5955, + "step": 1631 + }, + { + "epoch": 1.7884931506849315, + "grad_norm": 2.6275408267974854, + "learning_rate": 2.276604386677498e-06, + "loss": 1.6117, + "step": 1632 + }, + { + "epoch": 1.7895890410958906, + "grad_norm": 2.6952037811279297, + "learning_rate": 2.2745735174654753e-06, + "loss": 1.6894, + "step": 1633 + }, + { + "epoch": 1.7906849315068492, + "grad_norm": 2.5299718379974365, + "learning_rate": 2.2725426482534526e-06, + "loss": 1.6361, + "step": 1634 + }, + { + "epoch": 1.7917808219178082, + "grad_norm": 2.837455987930298, + "learning_rate": 2.27051177904143e-06, + "loss": 1.6269, + "step": 1635 + }, + { + "epoch": 1.7928767123287672, + "grad_norm": 2.5406227111816406, + "learning_rate": 2.268480909829407e-06, + "loss": 1.6538, + "step": 1636 + }, + { + "epoch": 1.793972602739726, + "grad_norm": 2.716038227081299, + "learning_rate": 2.2664500406173844e-06, + "loss": 1.6691, + "step": 1637 + }, + { + "epoch": 1.7950684931506848, + "grad_norm": 2.883439064025879, + "learning_rate": 2.2644191714053617e-06, + "loss": 1.6298, + "step": 1638 + }, + { + "epoch": 1.7961643835616439, + "grad_norm": 2.7631118297576904, + "learning_rate": 2.262388302193339e-06, + "loss": 1.6245, + "step": 1639 + }, + { + "epoch": 1.7972602739726027, + "grad_norm": 2.6777117252349854, + "learning_rate": 2.260357432981316e-06, + "loss": 1.6446, + "step": 1640 + }, + { + "epoch": 1.7972602739726027, + "eval_loss": 1.845224142074585, + "eval_runtime": 46.8476, + "eval_samples_per_second": 21.346, + "eval_steps_per_second": 0.897, + "step": 1640 + }, + { + "epoch": 1.7983561643835615, + "grad_norm": 2.6280243396759033, + "learning_rate": 2.2583265637692935e-06, + "loss": 1.6398, + "step": 1641 + }, + { + "epoch": 1.7994520547945205, + "grad_norm": 2.471925973892212, + "learning_rate": 2.2562956945572708e-06, + "loss": 1.6403, + "step": 1642 + }, + { + "epoch": 1.8005479452054796, + "grad_norm": 2.6293203830718994, + "learning_rate": 2.254264825345248e-06, + "loss": 1.642, + "step": 1643 + }, + { + "epoch": 1.8016438356164384, + "grad_norm": 2.651144027709961, + "learning_rate": 2.2522339561332253e-06, + "loss": 1.6445, + "step": 1644 + }, + { + "epoch": 1.8027397260273972, + "grad_norm": 2.6489877700805664, + "learning_rate": 2.2502030869212026e-06, + "loss": 1.6761, + "step": 1645 + }, + { + "epoch": 1.8038356164383562, + "grad_norm": 2.5869088172912598, + "learning_rate": 2.24817221770918e-06, + "loss": 1.6593, + "step": 1646 + }, + { + "epoch": 1.804931506849315, + "grad_norm": 2.411802053451538, + "learning_rate": 2.246141348497157e-06, + "loss": 1.6232, + "step": 1647 + }, + { + "epoch": 1.8060273972602738, + "grad_norm": 2.522610902786255, + "learning_rate": 2.2441104792851344e-06, + "loss": 1.6358, + "step": 1648 + }, + { + "epoch": 1.8071232876712329, + "grad_norm": 2.5430991649627686, + "learning_rate": 2.2420796100731117e-06, + "loss": 1.6605, + "step": 1649 + }, + { + "epoch": 1.808219178082192, + "grad_norm": 2.4440951347351074, + "learning_rate": 2.240048740861089e-06, + "loss": 1.5354, + "step": 1650 + }, + { + "epoch": 1.808219178082192, + "eval_loss": 1.8440407514572144, + "eval_runtime": 48.0239, + "eval_samples_per_second": 20.823, + "eval_steps_per_second": 0.875, + "step": 1650 + }, + { + "epoch": 1.8093150684931507, + "grad_norm": 2.5492775440216064, + "learning_rate": 2.2380178716490662e-06, + "loss": 1.668, + "step": 1651 + }, + { + "epoch": 1.8104109589041095, + "grad_norm": 2.7511909008026123, + "learning_rate": 2.2359870024370435e-06, + "loss": 1.6424, + "step": 1652 + }, + { + "epoch": 1.8115068493150686, + "grad_norm": 2.5176775455474854, + "learning_rate": 2.2339561332250204e-06, + "loss": 1.6336, + "step": 1653 + }, + { + "epoch": 1.8126027397260274, + "grad_norm": 2.5408577919006348, + "learning_rate": 2.2319252640129976e-06, + "loss": 1.5843, + "step": 1654 + }, + { + "epoch": 1.8136986301369862, + "grad_norm": 2.4791393280029297, + "learning_rate": 2.229894394800975e-06, + "loss": 1.555, + "step": 1655 + }, + { + "epoch": 1.8147945205479452, + "grad_norm": 2.383148670196533, + "learning_rate": 2.227863525588952e-06, + "loss": 1.598, + "step": 1656 + }, + { + "epoch": 1.8158904109589042, + "grad_norm": 2.429348945617676, + "learning_rate": 2.2258326563769295e-06, + "loss": 1.5943, + "step": 1657 + }, + { + "epoch": 1.816986301369863, + "grad_norm": 2.708796501159668, + "learning_rate": 2.2238017871649067e-06, + "loss": 1.6382, + "step": 1658 + }, + { + "epoch": 1.8180821917808219, + "grad_norm": 3.0029661655426025, + "learning_rate": 2.221770917952884e-06, + "loss": 1.6699, + "step": 1659 + }, + { + "epoch": 1.819178082191781, + "grad_norm": 2.5410728454589844, + "learning_rate": 2.2197400487408613e-06, + "loss": 1.5965, + "step": 1660 + }, + { + "epoch": 1.819178082191781, + "eval_loss": 1.8443207740783691, + "eval_runtime": 47.8671, + "eval_samples_per_second": 20.891, + "eval_steps_per_second": 0.877, + "step": 1660 + }, + { + "epoch": 1.8202739726027397, + "grad_norm": 2.5718178749084473, + "learning_rate": 2.2177091795288386e-06, + "loss": 1.6214, + "step": 1661 + }, + { + "epoch": 1.8213698630136985, + "grad_norm": 2.474177837371826, + "learning_rate": 2.215678310316816e-06, + "loss": 1.6498, + "step": 1662 + }, + { + "epoch": 1.8224657534246576, + "grad_norm": 2.5960052013397217, + "learning_rate": 2.213647441104793e-06, + "loss": 1.6156, + "step": 1663 + }, + { + "epoch": 1.8235616438356166, + "grad_norm": 2.6217992305755615, + "learning_rate": 2.2116165718927704e-06, + "loss": 1.6311, + "step": 1664 + }, + { + "epoch": 1.8246575342465754, + "grad_norm": 2.5601158142089844, + "learning_rate": 2.2095857026807473e-06, + "loss": 1.6008, + "step": 1665 + }, + { + "epoch": 1.8257534246575342, + "grad_norm": 2.7786355018615723, + "learning_rate": 2.2075548334687245e-06, + "loss": 1.62, + "step": 1666 + }, + { + "epoch": 1.8268493150684932, + "grad_norm": 2.548356771469116, + "learning_rate": 2.205523964256702e-06, + "loss": 1.6473, + "step": 1667 + }, + { + "epoch": 1.827945205479452, + "grad_norm": 2.6265289783477783, + "learning_rate": 2.203493095044679e-06, + "loss": 1.6565, + "step": 1668 + }, + { + "epoch": 1.8290410958904109, + "grad_norm": 2.522379159927368, + "learning_rate": 2.2014622258326564e-06, + "loss": 1.6764, + "step": 1669 + }, + { + "epoch": 1.83013698630137, + "grad_norm": 2.5479297637939453, + "learning_rate": 2.1994313566206336e-06, + "loss": 1.6194, + "step": 1670 + }, + { + "epoch": 1.83013698630137, + "eval_loss": 1.8434110879898071, + "eval_runtime": 46.6043, + "eval_samples_per_second": 21.457, + "eval_steps_per_second": 0.901, + "step": 1670 + }, + { + "epoch": 1.831232876712329, + "grad_norm": 2.5401806831359863, + "learning_rate": 2.197400487408611e-06, + "loss": 1.5911, + "step": 1671 + }, + { + "epoch": 1.8323287671232875, + "grad_norm": 2.466893434524536, + "learning_rate": 2.195369618196588e-06, + "loss": 1.6145, + "step": 1672 + }, + { + "epoch": 1.8334246575342465, + "grad_norm": 2.566653251647949, + "learning_rate": 2.1933387489845655e-06, + "loss": 1.6558, + "step": 1673 + }, + { + "epoch": 1.8345205479452056, + "grad_norm": 2.664053201675415, + "learning_rate": 2.1913078797725427e-06, + "loss": 1.7089, + "step": 1674 + }, + { + "epoch": 1.8356164383561644, + "grad_norm": 3.6594645977020264, + "learning_rate": 2.18927701056052e-06, + "loss": 1.6549, + "step": 1675 + }, + { + "epoch": 1.8367123287671232, + "grad_norm": 2.68450927734375, + "learning_rate": 2.1872461413484973e-06, + "loss": 1.6306, + "step": 1676 + }, + { + "epoch": 1.8378082191780822, + "grad_norm": 2.7094614505767822, + "learning_rate": 2.1852152721364746e-06, + "loss": 1.6147, + "step": 1677 + }, + { + "epoch": 1.838904109589041, + "grad_norm": 2.481332540512085, + "learning_rate": 2.183184402924452e-06, + "loss": 1.6612, + "step": 1678 + }, + { + "epoch": 1.8399999999999999, + "grad_norm": 2.5651051998138428, + "learning_rate": 2.181153533712429e-06, + "loss": 1.6394, + "step": 1679 + }, + { + "epoch": 1.841095890410959, + "grad_norm": 2.5570919513702393, + "learning_rate": 2.1791226645004064e-06, + "loss": 1.6162, + "step": 1680 + }, + { + "epoch": 1.841095890410959, + "eval_loss": 1.8433003425598145, + "eval_runtime": 46.8673, + "eval_samples_per_second": 21.337, + "eval_steps_per_second": 0.896, + "step": 1680 + }, + { + "epoch": 1.842191780821918, + "grad_norm": 2.6206037998199463, + "learning_rate": 2.1770917952883837e-06, + "loss": 1.6869, + "step": 1681 + }, + { + "epoch": 1.8432876712328767, + "grad_norm": 2.8962907791137695, + "learning_rate": 2.175060926076361e-06, + "loss": 1.7022, + "step": 1682 + }, + { + "epoch": 1.8443835616438355, + "grad_norm": 2.6211588382720947, + "learning_rate": 2.1730300568643382e-06, + "loss": 1.6249, + "step": 1683 + }, + { + "epoch": 1.8454794520547946, + "grad_norm": 2.7770566940307617, + "learning_rate": 2.1709991876523155e-06, + "loss": 1.5821, + "step": 1684 + }, + { + "epoch": 1.8465753424657534, + "grad_norm": 2.543658494949341, + "learning_rate": 2.1689683184402928e-06, + "loss": 1.6587, + "step": 1685 + }, + { + "epoch": 1.8476712328767122, + "grad_norm": 2.508188486099243, + "learning_rate": 2.16693744922827e-06, + "loss": 1.6114, + "step": 1686 + }, + { + "epoch": 1.8487671232876712, + "grad_norm": 2.703335762023926, + "learning_rate": 2.1649065800162473e-06, + "loss": 1.6152, + "step": 1687 + }, + { + "epoch": 1.8498630136986303, + "grad_norm": 2.469688892364502, + "learning_rate": 2.1628757108042246e-06, + "loss": 1.6121, + "step": 1688 + }, + { + "epoch": 1.850958904109589, + "grad_norm": 2.4836199283599854, + "learning_rate": 2.160844841592202e-06, + "loss": 1.5918, + "step": 1689 + }, + { + "epoch": 1.8520547945205479, + "grad_norm": 2.521851062774658, + "learning_rate": 2.1588139723801787e-06, + "loss": 1.6228, + "step": 1690 + }, + { + "epoch": 1.8520547945205479, + "eval_loss": 1.845065712928772, + "eval_runtime": 45.3003, + "eval_samples_per_second": 22.075, + "eval_steps_per_second": 0.927, + "step": 1690 + }, + { + "epoch": 1.853150684931507, + "grad_norm": 2.7808425426483154, + "learning_rate": 2.156783103168156e-06, + "loss": 1.5887, + "step": 1691 + }, + { + "epoch": 1.8542465753424657, + "grad_norm": 2.696708917617798, + "learning_rate": 2.1547522339561333e-06, + "loss": 1.6517, + "step": 1692 + }, + { + "epoch": 1.8553424657534245, + "grad_norm": 2.7222893238067627, + "learning_rate": 2.1527213647441106e-06, + "loss": 1.6817, + "step": 1693 + }, + { + "epoch": 1.8564383561643836, + "grad_norm": 2.504417896270752, + "learning_rate": 2.150690495532088e-06, + "loss": 1.647, + "step": 1694 + }, + { + "epoch": 1.8575342465753426, + "grad_norm": 2.574836015701294, + "learning_rate": 2.148659626320065e-06, + "loss": 1.6554, + "step": 1695 + }, + { + "epoch": 1.8586301369863014, + "grad_norm": 2.5004355907440186, + "learning_rate": 2.1466287571080424e-06, + "loss": 1.6089, + "step": 1696 + }, + { + "epoch": 1.8597260273972602, + "grad_norm": 2.6979293823242188, + "learning_rate": 2.1445978878960197e-06, + "loss": 1.6226, + "step": 1697 + }, + { + "epoch": 1.8608219178082193, + "grad_norm": 2.5239524841308594, + "learning_rate": 2.142567018683997e-06, + "loss": 1.6875, + "step": 1698 + }, + { + "epoch": 1.861917808219178, + "grad_norm": 2.5218467712402344, + "learning_rate": 2.140536149471974e-06, + "loss": 1.6095, + "step": 1699 + }, + { + "epoch": 1.8630136986301369, + "grad_norm": 2.5778310298919678, + "learning_rate": 2.1385052802599515e-06, + "loss": 1.6155, + "step": 1700 + }, + { + "epoch": 1.8630136986301369, + "eval_loss": 1.8443745374679565, + "eval_runtime": 46.6273, + "eval_samples_per_second": 21.447, + "eval_steps_per_second": 0.901, + "step": 1700 + }, + { + "epoch": 1.864109589041096, + "grad_norm": 2.671973705291748, + "learning_rate": 2.1364744110479288e-06, + "loss": 1.6696, + "step": 1701 + }, + { + "epoch": 1.865205479452055, + "grad_norm": 2.5520615577697754, + "learning_rate": 2.134443541835906e-06, + "loss": 1.6291, + "step": 1702 + }, + { + "epoch": 1.8663013698630135, + "grad_norm": 2.39286208152771, + "learning_rate": 2.132412672623883e-06, + "loss": 1.6012, + "step": 1703 + }, + { + "epoch": 1.8673972602739726, + "grad_norm": 2.8277461528778076, + "learning_rate": 2.13038180341186e-06, + "loss": 1.6398, + "step": 1704 + }, + { + "epoch": 1.8684931506849316, + "grad_norm": 2.7147881984710693, + "learning_rate": 2.1283509341998374e-06, + "loss": 1.6685, + "step": 1705 + }, + { + "epoch": 1.8695890410958904, + "grad_norm": 2.499044418334961, + "learning_rate": 2.1263200649878147e-06, + "loss": 1.6128, + "step": 1706 + }, + { + "epoch": 1.8706849315068492, + "grad_norm": 2.514784574508667, + "learning_rate": 2.124289195775792e-06, + "loss": 1.6263, + "step": 1707 + }, + { + "epoch": 1.8717808219178083, + "grad_norm": 2.505342483520508, + "learning_rate": 2.1222583265637693e-06, + "loss": 1.5532, + "step": 1708 + }, + { + "epoch": 1.872876712328767, + "grad_norm": 2.608139753341675, + "learning_rate": 2.1202274573517465e-06, + "loss": 1.5946, + "step": 1709 + }, + { + "epoch": 1.8739726027397259, + "grad_norm": 2.5578057765960693, + "learning_rate": 2.118196588139724e-06, + "loss": 1.6103, + "step": 1710 + }, + { + "epoch": 1.8739726027397259, + "eval_loss": 1.8425958156585693, + "eval_runtime": 44.206, + "eval_samples_per_second": 22.621, + "eval_steps_per_second": 0.95, + "step": 1710 + }, + { + "epoch": 1.875068493150685, + "grad_norm": 2.4881014823913574, + "learning_rate": 2.116165718927701e-06, + "loss": 1.6364, + "step": 1711 + }, + { + "epoch": 1.876164383561644, + "grad_norm": 2.4141550064086914, + "learning_rate": 2.1141348497156784e-06, + "loss": 1.5943, + "step": 1712 + }, + { + "epoch": 1.8772602739726028, + "grad_norm": 2.649827718734741, + "learning_rate": 2.1121039805036556e-06, + "loss": 1.5994, + "step": 1713 + }, + { + "epoch": 1.8783561643835616, + "grad_norm": 2.547238826751709, + "learning_rate": 2.110073111291633e-06, + "loss": 1.6247, + "step": 1714 + }, + { + "epoch": 1.8794520547945206, + "grad_norm": 2.6515395641326904, + "learning_rate": 2.10804224207961e-06, + "loss": 1.6543, + "step": 1715 + }, + { + "epoch": 1.8805479452054794, + "grad_norm": 2.5575809478759766, + "learning_rate": 2.1060113728675875e-06, + "loss": 1.6618, + "step": 1716 + }, + { + "epoch": 1.8816438356164382, + "grad_norm": 2.466344118118286, + "learning_rate": 2.1039805036555647e-06, + "loss": 1.5437, + "step": 1717 + }, + { + "epoch": 1.8827397260273973, + "grad_norm": 2.4897878170013428, + "learning_rate": 2.101949634443542e-06, + "loss": 1.6199, + "step": 1718 + }, + { + "epoch": 1.8838356164383563, + "grad_norm": 2.6578829288482666, + "learning_rate": 2.0999187652315193e-06, + "loss": 1.5989, + "step": 1719 + }, + { + "epoch": 1.884931506849315, + "grad_norm": 2.6173501014709473, + "learning_rate": 2.0978878960194966e-06, + "loss": 1.6392, + "step": 1720 + }, + { + "epoch": 1.884931506849315, + "eval_loss": 1.838721752166748, + "eval_runtime": 49.4357, + "eval_samples_per_second": 20.228, + "eval_steps_per_second": 0.85, + "step": 1720 + }, + { + "epoch": 1.886027397260274, + "grad_norm": 2.5655064582824707, + "learning_rate": 2.095857026807474e-06, + "loss": 1.6276, + "step": 1721 + }, + { + "epoch": 1.887123287671233, + "grad_norm": 2.685790777206421, + "learning_rate": 2.093826157595451e-06, + "loss": 1.6142, + "step": 1722 + }, + { + "epoch": 1.8882191780821918, + "grad_norm": 2.7278990745544434, + "learning_rate": 2.0917952883834284e-06, + "loss": 1.6292, + "step": 1723 + }, + { + "epoch": 1.8893150684931506, + "grad_norm": 2.5769667625427246, + "learning_rate": 2.0897644191714057e-06, + "loss": 1.6774, + "step": 1724 + }, + { + "epoch": 1.8904109589041096, + "grad_norm": 2.5120608806610107, + "learning_rate": 2.087733549959383e-06, + "loss": 1.5972, + "step": 1725 + }, + { + "epoch": 1.8915068493150686, + "grad_norm": 2.4658212661743164, + "learning_rate": 2.0857026807473602e-06, + "loss": 1.5683, + "step": 1726 + }, + { + "epoch": 1.8926027397260274, + "grad_norm": 2.526524782180786, + "learning_rate": 2.0836718115353375e-06, + "loss": 1.6687, + "step": 1727 + }, + { + "epoch": 1.8936986301369862, + "grad_norm": 2.3989250659942627, + "learning_rate": 2.0816409423233144e-06, + "loss": 1.6184, + "step": 1728 + }, + { + "epoch": 1.8947945205479453, + "grad_norm": 2.589115858078003, + "learning_rate": 2.0796100731112916e-06, + "loss": 1.6153, + "step": 1729 + }, + { + "epoch": 1.895890410958904, + "grad_norm": 2.4985275268554688, + "learning_rate": 2.077579203899269e-06, + "loss": 1.5682, + "step": 1730 + }, + { + "epoch": 1.895890410958904, + "eval_loss": 1.8374594449996948, + "eval_runtime": 47.0329, + "eval_samples_per_second": 21.262, + "eval_steps_per_second": 0.893, + "step": 1730 + }, + { + "epoch": 1.896986301369863, + "grad_norm": 2.5770835876464844, + "learning_rate": 2.075548334687246e-06, + "loss": 1.6437, + "step": 1731 + }, + { + "epoch": 1.898082191780822, + "grad_norm": 2.5111560821533203, + "learning_rate": 2.0735174654752235e-06, + "loss": 1.5711, + "step": 1732 + }, + { + "epoch": 1.899178082191781, + "grad_norm": 2.6110589504241943, + "learning_rate": 2.0714865962632007e-06, + "loss": 1.6125, + "step": 1733 + }, + { + "epoch": 1.9002739726027398, + "grad_norm": 2.6229705810546875, + "learning_rate": 2.069455727051178e-06, + "loss": 1.595, + "step": 1734 + }, + { + "epoch": 1.9013698630136986, + "grad_norm": 2.648716926574707, + "learning_rate": 2.0674248578391553e-06, + "loss": 1.5908, + "step": 1735 + }, + { + "epoch": 1.9024657534246576, + "grad_norm": 2.673701763153076, + "learning_rate": 2.0653939886271326e-06, + "loss": 1.6652, + "step": 1736 + }, + { + "epoch": 1.9035616438356164, + "grad_norm": 2.667478561401367, + "learning_rate": 2.06336311941511e-06, + "loss": 1.6088, + "step": 1737 + }, + { + "epoch": 1.9046575342465752, + "grad_norm": 3.5688586235046387, + "learning_rate": 2.061332250203087e-06, + "loss": 1.5852, + "step": 1738 + }, + { + "epoch": 1.9057534246575343, + "grad_norm": 2.512603282928467, + "learning_rate": 2.0593013809910644e-06, + "loss": 1.6218, + "step": 1739 + }, + { + "epoch": 1.9068493150684933, + "grad_norm": 2.5296075344085693, + "learning_rate": 2.0572705117790417e-06, + "loss": 1.6382, + "step": 1740 + }, + { + "epoch": 1.9068493150684933, + "eval_loss": 1.8406264781951904, + "eval_runtime": 43.8351, + "eval_samples_per_second": 22.813, + "eval_steps_per_second": 0.958, + "step": 1740 + }, + { + "epoch": 1.907945205479452, + "grad_norm": 2.598994255065918, + "learning_rate": 2.055239642567019e-06, + "loss": 1.6263, + "step": 1741 + }, + { + "epoch": 1.909041095890411, + "grad_norm": 2.93025279045105, + "learning_rate": 2.0532087733549962e-06, + "loss": 1.6502, + "step": 1742 + }, + { + "epoch": 1.91013698630137, + "grad_norm": 3.3655412197113037, + "learning_rate": 2.0511779041429735e-06, + "loss": 1.5872, + "step": 1743 + }, + { + "epoch": 1.9112328767123288, + "grad_norm": 2.955120325088501, + "learning_rate": 2.0491470349309508e-06, + "loss": 1.6404, + "step": 1744 + }, + { + "epoch": 1.9123287671232876, + "grad_norm": 2.5226638317108154, + "learning_rate": 2.047116165718928e-06, + "loss": 1.6102, + "step": 1745 + }, + { + "epoch": 1.9134246575342466, + "grad_norm": 2.5462093353271484, + "learning_rate": 2.0450852965069053e-06, + "loss": 1.6757, + "step": 1746 + }, + { + "epoch": 1.9145205479452054, + "grad_norm": 4.066190242767334, + "learning_rate": 2.0430544272948826e-06, + "loss": 1.6072, + "step": 1747 + }, + { + "epoch": 1.9156164383561642, + "grad_norm": 2.8548827171325684, + "learning_rate": 2.04102355808286e-06, + "loss": 1.5717, + "step": 1748 + }, + { + "epoch": 1.9167123287671233, + "grad_norm": 2.6206634044647217, + "learning_rate": 2.038992688870837e-06, + "loss": 1.5943, + "step": 1749 + }, + { + "epoch": 1.9178082191780823, + "grad_norm": 2.8793396949768066, + "learning_rate": 2.0369618196588144e-06, + "loss": 1.6121, + "step": 1750 + }, + { + "epoch": 1.9178082191780823, + "eval_loss": 1.83925199508667, + "eval_runtime": 44.747, + "eval_samples_per_second": 22.348, + "eval_steps_per_second": 0.939, + "step": 1750 + }, + { + "epoch": 1.9189041095890411, + "grad_norm": 2.6215157508850098, + "learning_rate": 2.0349309504467917e-06, + "loss": 1.6492, + "step": 1751 + }, + { + "epoch": 1.92, + "grad_norm": 2.6103456020355225, + "learning_rate": 2.032900081234769e-06, + "loss": 1.6531, + "step": 1752 + }, + { + "epoch": 1.921095890410959, + "grad_norm": 2.485775947570801, + "learning_rate": 2.030869212022746e-06, + "loss": 1.6086, + "step": 1753 + }, + { + "epoch": 1.9221917808219178, + "grad_norm": 2.399557113647461, + "learning_rate": 2.028838342810723e-06, + "loss": 1.6126, + "step": 1754 + }, + { + "epoch": 1.9232876712328766, + "grad_norm": 2.4711368083953857, + "learning_rate": 2.0268074735987004e-06, + "loss": 1.5985, + "step": 1755 + }, + { + "epoch": 1.9243835616438356, + "grad_norm": 2.654580593109131, + "learning_rate": 2.0247766043866777e-06, + "loss": 1.6557, + "step": 1756 + }, + { + "epoch": 1.9254794520547946, + "grad_norm": 2.5091657638549805, + "learning_rate": 2.022745735174655e-06, + "loss": 1.6181, + "step": 1757 + }, + { + "epoch": 1.9265753424657535, + "grad_norm": 2.5723490715026855, + "learning_rate": 2.020714865962632e-06, + "loss": 1.6421, + "step": 1758 + }, + { + "epoch": 1.9276712328767123, + "grad_norm": 2.4211535453796387, + "learning_rate": 2.0186839967506095e-06, + "loss": 1.6226, + "step": 1759 + }, + { + "epoch": 1.9287671232876713, + "grad_norm": 2.5550055503845215, + "learning_rate": 2.0166531275385868e-06, + "loss": 1.5729, + "step": 1760 + }, + { + "epoch": 1.9287671232876713, + "eval_loss": 1.8397995233535767, + "eval_runtime": 44.4762, + "eval_samples_per_second": 22.484, + "eval_steps_per_second": 0.944, + "step": 1760 + }, + { + "epoch": 1.9298630136986301, + "grad_norm": 2.618758201599121, + "learning_rate": 2.014622258326564e-06, + "loss": 1.6356, + "step": 1761 + }, + { + "epoch": 1.930958904109589, + "grad_norm": 2.6717913150787354, + "learning_rate": 2.0125913891145413e-06, + "loss": 1.6246, + "step": 1762 + }, + { + "epoch": 1.932054794520548, + "grad_norm": 2.5464835166931152, + "learning_rate": 2.0105605199025186e-06, + "loss": 1.6027, + "step": 1763 + }, + { + "epoch": 1.933150684931507, + "grad_norm": 2.557669162750244, + "learning_rate": 2.008529650690496e-06, + "loss": 1.6531, + "step": 1764 + }, + { + "epoch": 1.9342465753424658, + "grad_norm": 2.581970453262329, + "learning_rate": 2.0064987814784727e-06, + "loss": 1.6076, + "step": 1765 + }, + { + "epoch": 1.9353424657534246, + "grad_norm": 2.609236478805542, + "learning_rate": 2.00446791226645e-06, + "loss": 1.5932, + "step": 1766 + }, + { + "epoch": 1.9364383561643836, + "grad_norm": 2.4249160289764404, + "learning_rate": 2.0024370430544273e-06, + "loss": 1.6372, + "step": 1767 + }, + { + "epoch": 1.9375342465753425, + "grad_norm": 2.5093257427215576, + "learning_rate": 2.0004061738424045e-06, + "loss": 1.6627, + "step": 1768 + }, + { + "epoch": 1.9386301369863013, + "grad_norm": 2.3989412784576416, + "learning_rate": 1.998375304630382e-06, + "loss": 1.6292, + "step": 1769 + }, + { + "epoch": 1.9397260273972603, + "grad_norm": 2.4299519062042236, + "learning_rate": 1.996344435418359e-06, + "loss": 1.6385, + "step": 1770 + }, + { + "epoch": 1.9397260273972603, + "eval_loss": 1.8367400169372559, + "eval_runtime": 45.921, + "eval_samples_per_second": 21.777, + "eval_steps_per_second": 0.915, + "step": 1770 + }, + { + "epoch": 1.9408219178082193, + "grad_norm": 2.509530782699585, + "learning_rate": 1.9943135662063364e-06, + "loss": 1.6441, + "step": 1771 + }, + { + "epoch": 1.941917808219178, + "grad_norm": 2.591426134109497, + "learning_rate": 1.9922826969943136e-06, + "loss": 1.6382, + "step": 1772 + }, + { + "epoch": 1.943013698630137, + "grad_norm": 2.6569297313690186, + "learning_rate": 1.990251827782291e-06, + "loss": 1.6469, + "step": 1773 + }, + { + "epoch": 1.944109589041096, + "grad_norm": 2.6453206539154053, + "learning_rate": 1.988220958570268e-06, + "loss": 1.6348, + "step": 1774 + }, + { + "epoch": 1.9452054794520548, + "grad_norm": 2.5522994995117188, + "learning_rate": 1.9861900893582455e-06, + "loss": 1.6311, + "step": 1775 + }, + { + "epoch": 1.9463013698630136, + "grad_norm": 2.5981009006500244, + "learning_rate": 1.9841592201462227e-06, + "loss": 1.585, + "step": 1776 + }, + { + "epoch": 1.9473972602739726, + "grad_norm": 2.617323637008667, + "learning_rate": 1.9821283509342e-06, + "loss": 1.584, + "step": 1777 + }, + { + "epoch": 1.9484931506849315, + "grad_norm": 2.6743686199188232, + "learning_rate": 1.9800974817221773e-06, + "loss": 1.6334, + "step": 1778 + }, + { + "epoch": 1.9495890410958903, + "grad_norm": 2.418383836746216, + "learning_rate": 1.9780666125101546e-06, + "loss": 1.5934, + "step": 1779 + }, + { + "epoch": 1.9506849315068493, + "grad_norm": 2.7952582836151123, + "learning_rate": 1.976035743298132e-06, + "loss": 1.6779, + "step": 1780 + }, + { + "epoch": 1.9506849315068493, + "eval_loss": 1.8379877805709839, + "eval_runtime": 46.545, + "eval_samples_per_second": 21.485, + "eval_steps_per_second": 0.902, + "step": 1780 + }, + { + "epoch": 1.9517808219178083, + "grad_norm": 2.5364768505096436, + "learning_rate": 1.974004874086109e-06, + "loss": 1.5802, + "step": 1781 + }, + { + "epoch": 1.9528767123287671, + "grad_norm": 2.446056365966797, + "learning_rate": 1.9719740048740864e-06, + "loss": 1.6141, + "step": 1782 + }, + { + "epoch": 1.953972602739726, + "grad_norm": 2.527529239654541, + "learning_rate": 1.9699431356620637e-06, + "loss": 1.6042, + "step": 1783 + }, + { + "epoch": 1.955068493150685, + "grad_norm": 2.5953001976013184, + "learning_rate": 1.967912266450041e-06, + "loss": 1.5959, + "step": 1784 + }, + { + "epoch": 1.9561643835616438, + "grad_norm": 2.545994758605957, + "learning_rate": 1.9658813972380182e-06, + "loss": 1.7091, + "step": 1785 + }, + { + "epoch": 1.9572602739726026, + "grad_norm": 2.3900482654571533, + "learning_rate": 1.9638505280259955e-06, + "loss": 1.5946, + "step": 1786 + }, + { + "epoch": 1.9583561643835616, + "grad_norm": 2.51353120803833, + "learning_rate": 1.9618196588139728e-06, + "loss": 1.6166, + "step": 1787 + }, + { + "epoch": 1.9594520547945207, + "grad_norm": 2.550389528274536, + "learning_rate": 1.95978878960195e-06, + "loss": 1.6817, + "step": 1788 + }, + { + "epoch": 1.9605479452054795, + "grad_norm": 2.6815762519836426, + "learning_rate": 1.9577579203899273e-06, + "loss": 1.6594, + "step": 1789 + }, + { + "epoch": 1.9616438356164383, + "grad_norm": 2.591909885406494, + "learning_rate": 1.955727051177904e-06, + "loss": 1.5991, + "step": 1790 + }, + { + "epoch": 1.9616438356164383, + "eval_loss": 1.8390737771987915, + "eval_runtime": 46.8495, + "eval_samples_per_second": 21.345, + "eval_steps_per_second": 0.896, + "step": 1790 + }, + { + "epoch": 1.9627397260273973, + "grad_norm": 2.5111982822418213, + "learning_rate": 1.9536961819658815e-06, + "loss": 1.5993, + "step": 1791 + }, + { + "epoch": 1.9638356164383561, + "grad_norm": 2.655891180038452, + "learning_rate": 1.9516653127538587e-06, + "loss": 1.5915, + "step": 1792 + }, + { + "epoch": 1.964931506849315, + "grad_norm": 2.471090316772461, + "learning_rate": 1.949634443541836e-06, + "loss": 1.5867, + "step": 1793 + }, + { + "epoch": 1.966027397260274, + "grad_norm": 2.8599209785461426, + "learning_rate": 1.9476035743298133e-06, + "loss": 1.6485, + "step": 1794 + }, + { + "epoch": 1.967123287671233, + "grad_norm": 2.4986093044281006, + "learning_rate": 1.9455727051177906e-06, + "loss": 1.6084, + "step": 1795 + }, + { + "epoch": 1.9682191780821918, + "grad_norm": 2.6345789432525635, + "learning_rate": 1.943541835905768e-06, + "loss": 1.6719, + "step": 1796 + }, + { + "epoch": 1.9693150684931506, + "grad_norm": 2.5636799335479736, + "learning_rate": 1.941510966693745e-06, + "loss": 1.5817, + "step": 1797 + }, + { + "epoch": 1.9704109589041097, + "grad_norm": 2.5867199897766113, + "learning_rate": 1.9394800974817224e-06, + "loss": 1.6498, + "step": 1798 + }, + { + "epoch": 1.9715068493150685, + "grad_norm": 2.6369481086730957, + "learning_rate": 1.9374492282696997e-06, + "loss": 1.6791, + "step": 1799 + }, + { + "epoch": 1.9726027397260273, + "grad_norm": 2.4554147720336914, + "learning_rate": 1.935418359057677e-06, + "loss": 1.6319, + "step": 1800 + }, + { + "epoch": 1.9726027397260273, + "eval_loss": 1.8408559560775757, + "eval_runtime": 49.3515, + "eval_samples_per_second": 20.263, + "eval_steps_per_second": 0.851, + "step": 1800 + }, + { + "epoch": 1.9736986301369863, + "grad_norm": 2.5412509441375732, + "learning_rate": 1.9333874898456542e-06, + "loss": 1.6673, + "step": 1801 + }, + { + "epoch": 1.9747945205479454, + "grad_norm": 2.5755672454833984, + "learning_rate": 1.931356620633631e-06, + "loss": 1.5636, + "step": 1802 + }, + { + "epoch": 1.9758904109589042, + "grad_norm": 2.602530002593994, + "learning_rate": 1.9293257514216083e-06, + "loss": 1.6261, + "step": 1803 + }, + { + "epoch": 1.976986301369863, + "grad_norm": 2.755247116088867, + "learning_rate": 1.9272948822095856e-06, + "loss": 1.5817, + "step": 1804 + }, + { + "epoch": 1.978082191780822, + "grad_norm": 2.528991460800171, + "learning_rate": 1.925264012997563e-06, + "loss": 1.6173, + "step": 1805 + }, + { + "epoch": 1.9791780821917808, + "grad_norm": 2.473057508468628, + "learning_rate": 1.92323314378554e-06, + "loss": 1.648, + "step": 1806 + }, + { + "epoch": 1.9802739726027396, + "grad_norm": 2.62630033493042, + "learning_rate": 1.9212022745735174e-06, + "loss": 1.5864, + "step": 1807 + }, + { + "epoch": 1.9813698630136987, + "grad_norm": 2.7876687049865723, + "learning_rate": 1.9191714053614947e-06, + "loss": 1.5804, + "step": 1808 + }, + { + "epoch": 1.9824657534246577, + "grad_norm": 2.491473436355591, + "learning_rate": 1.917140536149472e-06, + "loss": 1.5716, + "step": 1809 + }, + { + "epoch": 1.9835616438356163, + "grad_norm": 2.656989336013794, + "learning_rate": 1.9151096669374493e-06, + "loss": 1.6595, + "step": 1810 + }, + { + "epoch": 1.9835616438356163, + "eval_loss": 1.839701771736145, + "eval_runtime": 44.5976, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 1810 + }, + { + "epoch": 1.9846575342465753, + "grad_norm": 2.6800942420959473, + "learning_rate": 1.9130787977254266e-06, + "loss": 1.6415, + "step": 1811 + }, + { + "epoch": 1.9857534246575343, + "grad_norm": 2.4725751876831055, + "learning_rate": 1.911047928513404e-06, + "loss": 1.5901, + "step": 1812 + }, + { + "epoch": 1.9868493150684932, + "grad_norm": 2.552624464035034, + "learning_rate": 1.909017059301381e-06, + "loss": 1.5956, + "step": 1813 + }, + { + "epoch": 1.987945205479452, + "grad_norm": 2.6042184829711914, + "learning_rate": 1.9069861900893586e-06, + "loss": 1.6328, + "step": 1814 + }, + { + "epoch": 1.989041095890411, + "grad_norm": 2.5159223079681396, + "learning_rate": 1.9049553208773354e-06, + "loss": 1.604, + "step": 1815 + }, + { + "epoch": 1.9901369863013698, + "grad_norm": 2.488041639328003, + "learning_rate": 1.9029244516653127e-06, + "loss": 1.5539, + "step": 1816 + }, + { + "epoch": 1.9912328767123286, + "grad_norm": 2.659151077270508, + "learning_rate": 1.90089358245329e-06, + "loss": 1.5999, + "step": 1817 + }, + { + "epoch": 1.9923287671232877, + "grad_norm": 2.700378179550171, + "learning_rate": 1.8988627132412673e-06, + "loss": 1.6485, + "step": 1818 + }, + { + "epoch": 1.9934246575342467, + "grad_norm": 2.3943607807159424, + "learning_rate": 1.8968318440292445e-06, + "loss": 1.5736, + "step": 1819 + }, + { + "epoch": 1.9945205479452055, + "grad_norm": 2.8171095848083496, + "learning_rate": 1.8948009748172218e-06, + "loss": 1.6259, + "step": 1820 + }, + { + "epoch": 1.9945205479452055, + "eval_loss": 1.8379307985305786, + "eval_runtime": 48.6918, + "eval_samples_per_second": 20.537, + "eval_steps_per_second": 0.863, + "step": 1820 + }, + { + "epoch": 1.9956164383561643, + "grad_norm": 2.562655210494995, + "learning_rate": 1.892770105605199e-06, + "loss": 1.6439, + "step": 1821 + }, + { + "epoch": 1.9967123287671233, + "grad_norm": 2.5640337467193604, + "learning_rate": 1.8907392363931764e-06, + "loss": 1.6113, + "step": 1822 + }, + { + "epoch": 1.9978082191780822, + "grad_norm": 2.5324809551239014, + "learning_rate": 1.8887083671811536e-06, + "loss": 1.6331, + "step": 1823 + }, + { + "epoch": 1.998904109589041, + "grad_norm": 2.393224000930786, + "learning_rate": 1.886677497969131e-06, + "loss": 1.5228, + "step": 1824 + }, + { + "epoch": 2.0, + "grad_norm": 3.073945999145508, + "learning_rate": 1.8846466287571082e-06, + "loss": 1.632, + "step": 1825 + }, + { + "epoch": 2.001095890410959, + "grad_norm": 2.7066094875335693, + "learning_rate": 1.8826157595450855e-06, + "loss": 1.5508, + "step": 1826 + }, + { + "epoch": 2.0021917808219176, + "grad_norm": 2.981933832168579, + "learning_rate": 1.8805848903330625e-06, + "loss": 1.6973, + "step": 1827 + }, + { + "epoch": 2.0032876712328767, + "grad_norm": 2.903444528579712, + "learning_rate": 1.8785540211210398e-06, + "loss": 1.6036, + "step": 1828 + }, + { + "epoch": 2.0043835616438357, + "grad_norm": 2.8616178035736084, + "learning_rate": 1.876523151909017e-06, + "loss": 1.5723, + "step": 1829 + }, + { + "epoch": 2.0054794520547947, + "grad_norm": 2.8243541717529297, + "learning_rate": 1.8744922826969944e-06, + "loss": 1.555, + "step": 1830 + }, + { + "epoch": 2.0054794520547947, + "eval_loss": 1.8437565565109253, + "eval_runtime": 48.5449, + "eval_samples_per_second": 20.599, + "eval_steps_per_second": 0.865, + "step": 1830 + }, + { + "epoch": 2.0065753424657533, + "grad_norm": 2.803142547607422, + "learning_rate": 1.8724614134849716e-06, + "loss": 1.5861, + "step": 1831 + }, + { + "epoch": 2.0076712328767123, + "grad_norm": 3.146677017211914, + "learning_rate": 1.870430544272949e-06, + "loss": 1.6791, + "step": 1832 + }, + { + "epoch": 2.0087671232876714, + "grad_norm": 3.2554781436920166, + "learning_rate": 1.8683996750609262e-06, + "loss": 1.5626, + "step": 1833 + }, + { + "epoch": 2.00986301369863, + "grad_norm": 2.955395460128784, + "learning_rate": 1.8663688058489035e-06, + "loss": 1.5893, + "step": 1834 + }, + { + "epoch": 2.010958904109589, + "grad_norm": 2.941377878189087, + "learning_rate": 1.8643379366368807e-06, + "loss": 1.5962, + "step": 1835 + }, + { + "epoch": 2.012054794520548, + "grad_norm": 2.765620708465576, + "learning_rate": 1.862307067424858e-06, + "loss": 1.595, + "step": 1836 + }, + { + "epoch": 2.013150684931507, + "grad_norm": 2.8529839515686035, + "learning_rate": 1.8602761982128353e-06, + "loss": 1.6296, + "step": 1837 + }, + { + "epoch": 2.0142465753424657, + "grad_norm": 2.7770590782165527, + "learning_rate": 1.8582453290008126e-06, + "loss": 1.5711, + "step": 1838 + }, + { + "epoch": 2.0153424657534247, + "grad_norm": 2.827406883239746, + "learning_rate": 1.8562144597887898e-06, + "loss": 1.5966, + "step": 1839 + }, + { + "epoch": 2.0164383561643837, + "grad_norm": 2.927119016647339, + "learning_rate": 1.854183590576767e-06, + "loss": 1.6463, + "step": 1840 + }, + { + "epoch": 2.0164383561643837, + "eval_loss": 1.841707468032837, + "eval_runtime": 49.0965, + "eval_samples_per_second": 20.368, + "eval_steps_per_second": 0.855, + "step": 1840 + }, + { + "epoch": 2.0175342465753423, + "grad_norm": 2.8223392963409424, + "learning_rate": 1.8521527213647442e-06, + "loss": 1.559, + "step": 1841 + }, + { + "epoch": 2.0186301369863013, + "grad_norm": 2.862473726272583, + "learning_rate": 1.8501218521527215e-06, + "loss": 1.5189, + "step": 1842 + }, + { + "epoch": 2.0197260273972604, + "grad_norm": 2.9060091972351074, + "learning_rate": 1.8480909829406987e-06, + "loss": 1.5631, + "step": 1843 + }, + { + "epoch": 2.020821917808219, + "grad_norm": 2.8727498054504395, + "learning_rate": 1.846060113728676e-06, + "loss": 1.6013, + "step": 1844 + }, + { + "epoch": 2.021917808219178, + "grad_norm": 2.862298011779785, + "learning_rate": 1.8440292445166533e-06, + "loss": 1.549, + "step": 1845 + }, + { + "epoch": 2.023013698630137, + "grad_norm": 2.8221330642700195, + "learning_rate": 1.8419983753046306e-06, + "loss": 1.6341, + "step": 1846 + }, + { + "epoch": 2.024109589041096, + "grad_norm": 2.938336133956909, + "learning_rate": 1.8399675060926078e-06, + "loss": 1.602, + "step": 1847 + }, + { + "epoch": 2.0252054794520546, + "grad_norm": 2.719489812850952, + "learning_rate": 1.8379366368805851e-06, + "loss": 1.5626, + "step": 1848 + }, + { + "epoch": 2.0263013698630137, + "grad_norm": 2.6615874767303467, + "learning_rate": 1.8359057676685624e-06, + "loss": 1.5535, + "step": 1849 + }, + { + "epoch": 2.0273972602739727, + "grad_norm": 2.835413932800293, + "learning_rate": 1.8338748984565397e-06, + "loss": 1.5742, + "step": 1850 + }, + { + "epoch": 2.0273972602739727, + "eval_loss": 1.8375873565673828, + "eval_runtime": 45.731, + "eval_samples_per_second": 21.867, + "eval_steps_per_second": 0.918, + "step": 1850 + }, + { + "epoch": 2.0284931506849313, + "grad_norm": 2.773707389831543, + "learning_rate": 1.831844029244517e-06, + "loss": 1.5499, + "step": 1851 + }, + { + "epoch": 2.0295890410958903, + "grad_norm": 3.0959930419921875, + "learning_rate": 1.829813160032494e-06, + "loss": 1.5656, + "step": 1852 + }, + { + "epoch": 2.0306849315068494, + "grad_norm": 2.8521792888641357, + "learning_rate": 1.8277822908204713e-06, + "loss": 1.5982, + "step": 1853 + }, + { + "epoch": 2.0317808219178084, + "grad_norm": 2.970618486404419, + "learning_rate": 1.8257514216084486e-06, + "loss": 1.5731, + "step": 1854 + }, + { + "epoch": 2.032876712328767, + "grad_norm": 3.0716512203216553, + "learning_rate": 1.8237205523964258e-06, + "loss": 1.5824, + "step": 1855 + }, + { + "epoch": 2.033972602739726, + "grad_norm": 2.8776299953460693, + "learning_rate": 1.8216896831844031e-06, + "loss": 1.6283, + "step": 1856 + }, + { + "epoch": 2.035068493150685, + "grad_norm": 2.72183895111084, + "learning_rate": 1.8196588139723804e-06, + "loss": 1.5422, + "step": 1857 + }, + { + "epoch": 2.0361643835616436, + "grad_norm": 2.853234052658081, + "learning_rate": 1.8176279447603577e-06, + "loss": 1.5533, + "step": 1858 + }, + { + "epoch": 2.0372602739726027, + "grad_norm": 2.7193424701690674, + "learning_rate": 1.815597075548335e-06, + "loss": 1.5923, + "step": 1859 + }, + { + "epoch": 2.0383561643835617, + "grad_norm": 2.9768450260162354, + "learning_rate": 1.8135662063363122e-06, + "loss": 1.6277, + "step": 1860 + }, + { + "epoch": 2.0383561643835617, + "eval_loss": 1.8401803970336914, + "eval_runtime": 45.8062, + "eval_samples_per_second": 21.831, + "eval_steps_per_second": 0.917, + "step": 1860 + }, + { + "epoch": 2.0394520547945207, + "grad_norm": 2.752772569656372, + "learning_rate": 1.8115353371242895e-06, + "loss": 1.5241, + "step": 1861 + }, + { + "epoch": 2.0405479452054793, + "grad_norm": 2.784980297088623, + "learning_rate": 1.8095044679122668e-06, + "loss": 1.5657, + "step": 1862 + }, + { + "epoch": 2.0416438356164384, + "grad_norm": 2.7485857009887695, + "learning_rate": 1.807473598700244e-06, + "loss": 1.5501, + "step": 1863 + }, + { + "epoch": 2.0427397260273974, + "grad_norm": 2.7191033363342285, + "learning_rate": 1.805442729488221e-06, + "loss": 1.5718, + "step": 1864 + }, + { + "epoch": 2.043835616438356, + "grad_norm": 2.697540044784546, + "learning_rate": 1.8034118602761982e-06, + "loss": 1.5804, + "step": 1865 + }, + { + "epoch": 2.044931506849315, + "grad_norm": 2.7175710201263428, + "learning_rate": 1.8013809910641754e-06, + "loss": 1.5809, + "step": 1866 + }, + { + "epoch": 2.046027397260274, + "grad_norm": 2.764824151992798, + "learning_rate": 1.7993501218521527e-06, + "loss": 1.545, + "step": 1867 + }, + { + "epoch": 2.047123287671233, + "grad_norm": 2.946669816970825, + "learning_rate": 1.79731925264013e-06, + "loss": 1.595, + "step": 1868 + }, + { + "epoch": 2.0482191780821917, + "grad_norm": 2.689326524734497, + "learning_rate": 1.7952883834281073e-06, + "loss": 1.5676, + "step": 1869 + }, + { + "epoch": 2.0493150684931507, + "grad_norm": 2.759997606277466, + "learning_rate": 1.7932575142160846e-06, + "loss": 1.6032, + "step": 1870 + }, + { + "epoch": 2.0493150684931507, + "eval_loss": 1.8379274606704712, + "eval_runtime": 45.5941, + "eval_samples_per_second": 21.933, + "eval_steps_per_second": 0.921, + "step": 1870 + }, + { + "epoch": 2.0504109589041097, + "grad_norm": 2.8656060695648193, + "learning_rate": 1.7912266450040618e-06, + "loss": 1.6086, + "step": 1871 + }, + { + "epoch": 2.0515068493150683, + "grad_norm": 2.7155873775482178, + "learning_rate": 1.789195775792039e-06, + "loss": 1.5218, + "step": 1872 + }, + { + "epoch": 2.0526027397260274, + "grad_norm": 2.8487389087677, + "learning_rate": 1.7871649065800164e-06, + "loss": 1.6285, + "step": 1873 + }, + { + "epoch": 2.0536986301369864, + "grad_norm": 2.8070144653320312, + "learning_rate": 1.7851340373679937e-06, + "loss": 1.5479, + "step": 1874 + }, + { + "epoch": 2.0547945205479454, + "grad_norm": 2.856282949447632, + "learning_rate": 1.783103168155971e-06, + "loss": 1.5689, + "step": 1875 + }, + { + "epoch": 2.055890410958904, + "grad_norm": 2.7465922832489014, + "learning_rate": 1.7810722989439482e-06, + "loss": 1.5733, + "step": 1876 + }, + { + "epoch": 2.056986301369863, + "grad_norm": 2.824997901916504, + "learning_rate": 1.7790414297319253e-06, + "loss": 1.5583, + "step": 1877 + }, + { + "epoch": 2.058082191780822, + "grad_norm": 2.8030786514282227, + "learning_rate": 1.7770105605199025e-06, + "loss": 1.5383, + "step": 1878 + }, + { + "epoch": 2.0591780821917807, + "grad_norm": 2.911377429962158, + "learning_rate": 1.7749796913078798e-06, + "loss": 1.586, + "step": 1879 + }, + { + "epoch": 2.0602739726027397, + "grad_norm": 2.7452991008758545, + "learning_rate": 1.772948822095857e-06, + "loss": 1.6409, + "step": 1880 + }, + { + "epoch": 2.0602739726027397, + "eval_loss": 1.8399248123168945, + "eval_runtime": 46.1241, + "eval_samples_per_second": 21.681, + "eval_steps_per_second": 0.911, + "step": 1880 + }, + { + "epoch": 2.0613698630136987, + "grad_norm": 2.8255481719970703, + "learning_rate": 1.7709179528838344e-06, + "loss": 1.6134, + "step": 1881 + }, + { + "epoch": 2.0624657534246573, + "grad_norm": 2.865100622177124, + "learning_rate": 1.7688870836718116e-06, + "loss": 1.5791, + "step": 1882 + }, + { + "epoch": 2.0635616438356164, + "grad_norm": 2.740133047103882, + "learning_rate": 1.766856214459789e-06, + "loss": 1.5651, + "step": 1883 + }, + { + "epoch": 2.0646575342465754, + "grad_norm": 2.7219033241271973, + "learning_rate": 1.7648253452477662e-06, + "loss": 1.5453, + "step": 1884 + }, + { + "epoch": 2.0657534246575344, + "grad_norm": 2.7266135215759277, + "learning_rate": 1.7627944760357435e-06, + "loss": 1.6375, + "step": 1885 + }, + { + "epoch": 2.066849315068493, + "grad_norm": 2.8476688861846924, + "learning_rate": 1.7607636068237208e-06, + "loss": 1.5877, + "step": 1886 + }, + { + "epoch": 2.067945205479452, + "grad_norm": 2.810105085372925, + "learning_rate": 1.758732737611698e-06, + "loss": 1.5796, + "step": 1887 + }, + { + "epoch": 2.069041095890411, + "grad_norm": 2.78798246383667, + "learning_rate": 1.7567018683996753e-06, + "loss": 1.5708, + "step": 1888 + }, + { + "epoch": 2.0701369863013697, + "grad_norm": 2.750525951385498, + "learning_rate": 1.7546709991876524e-06, + "loss": 1.5567, + "step": 1889 + }, + { + "epoch": 2.0712328767123287, + "grad_norm": 2.818673849105835, + "learning_rate": 1.7526401299756296e-06, + "loss": 1.5661, + "step": 1890 + }, + { + "epoch": 2.0712328767123287, + "eval_loss": 1.8374614715576172, + "eval_runtime": 43.5618, + "eval_samples_per_second": 22.956, + "eval_steps_per_second": 0.964, + "step": 1890 + }, + { + "epoch": 2.0723287671232877, + "grad_norm": 2.7967586517333984, + "learning_rate": 1.750609260763607e-06, + "loss": 1.5846, + "step": 1891 + }, + { + "epoch": 2.0734246575342468, + "grad_norm": 2.766272783279419, + "learning_rate": 1.7485783915515842e-06, + "loss": 1.5721, + "step": 1892 + }, + { + "epoch": 2.0745205479452054, + "grad_norm": 2.8006162643432617, + "learning_rate": 1.7465475223395615e-06, + "loss": 1.5261, + "step": 1893 + }, + { + "epoch": 2.0756164383561644, + "grad_norm": 3.714587688446045, + "learning_rate": 1.7445166531275387e-06, + "loss": 1.6599, + "step": 1894 + }, + { + "epoch": 2.0767123287671234, + "grad_norm": 3.143052339553833, + "learning_rate": 1.742485783915516e-06, + "loss": 1.659, + "step": 1895 + }, + { + "epoch": 2.077808219178082, + "grad_norm": 2.8256232738494873, + "learning_rate": 1.7404549147034933e-06, + "loss": 1.5076, + "step": 1896 + }, + { + "epoch": 2.078904109589041, + "grad_norm": 2.8843302726745605, + "learning_rate": 1.7384240454914706e-06, + "loss": 1.5862, + "step": 1897 + }, + { + "epoch": 2.08, + "grad_norm": 6.692182540893555, + "learning_rate": 1.7363931762794478e-06, + "loss": 1.6144, + "step": 1898 + }, + { + "epoch": 2.081095890410959, + "grad_norm": 2.928053617477417, + "learning_rate": 1.7343623070674251e-06, + "loss": 1.571, + "step": 1899 + }, + { + "epoch": 2.0821917808219177, + "grad_norm": 2.8071024417877197, + "learning_rate": 1.7323314378554024e-06, + "loss": 1.5861, + "step": 1900 + }, + { + "epoch": 2.0821917808219177, + "eval_loss": 1.837381362915039, + "eval_runtime": 44.7089, + "eval_samples_per_second": 22.367, + "eval_steps_per_second": 0.939, + "step": 1900 + }, + { + "epoch": 2.0832876712328767, + "grad_norm": 3.070399522781372, + "learning_rate": 1.7303005686433797e-06, + "loss": 1.5636, + "step": 1901 + }, + { + "epoch": 2.0843835616438358, + "grad_norm": 3.298100709915161, + "learning_rate": 1.7282696994313567e-06, + "loss": 1.5717, + "step": 1902 + }, + { + "epoch": 2.0854794520547943, + "grad_norm": 5.891775131225586, + "learning_rate": 1.726238830219334e-06, + "loss": 1.5757, + "step": 1903 + }, + { + "epoch": 2.0865753424657534, + "grad_norm": 2.838676691055298, + "learning_rate": 1.7242079610073113e-06, + "loss": 1.6369, + "step": 1904 + }, + { + "epoch": 2.0876712328767124, + "grad_norm": 2.8816466331481934, + "learning_rate": 1.7221770917952886e-06, + "loss": 1.5885, + "step": 1905 + }, + { + "epoch": 2.0887671232876714, + "grad_norm": 3.1703031063079834, + "learning_rate": 1.7201462225832658e-06, + "loss": 1.5966, + "step": 1906 + }, + { + "epoch": 2.08986301369863, + "grad_norm": 3.504777669906616, + "learning_rate": 1.7181153533712431e-06, + "loss": 1.4932, + "step": 1907 + }, + { + "epoch": 2.090958904109589, + "grad_norm": 3.1451492309570312, + "learning_rate": 1.7160844841592204e-06, + "loss": 1.5071, + "step": 1908 + }, + { + "epoch": 2.092054794520548, + "grad_norm": 2.7158446311950684, + "learning_rate": 1.7140536149471977e-06, + "loss": 1.5725, + "step": 1909 + }, + { + "epoch": 2.0931506849315067, + "grad_norm": 2.8391170501708984, + "learning_rate": 1.712022745735175e-06, + "loss": 1.5742, + "step": 1910 + }, + { + "epoch": 2.0931506849315067, + "eval_loss": 1.8371202945709229, + "eval_runtime": 43.2464, + "eval_samples_per_second": 23.123, + "eval_steps_per_second": 0.971, + "step": 1910 + }, + { + "epoch": 2.0942465753424657, + "grad_norm": 2.8346052169799805, + "learning_rate": 1.7099918765231522e-06, + "loss": 1.5808, + "step": 1911 + }, + { + "epoch": 2.0953424657534248, + "grad_norm": 2.767813205718994, + "learning_rate": 1.7079610073111295e-06, + "loss": 1.5209, + "step": 1912 + }, + { + "epoch": 2.0964383561643833, + "grad_norm": 2.7955846786499023, + "learning_rate": 1.7059301380991068e-06, + "loss": 1.581, + "step": 1913 + }, + { + "epoch": 2.0975342465753424, + "grad_norm": 2.8616271018981934, + "learning_rate": 1.7038992688870836e-06, + "loss": 1.5758, + "step": 1914 + }, + { + "epoch": 2.0986301369863014, + "grad_norm": 2.7975544929504395, + "learning_rate": 1.701868399675061e-06, + "loss": 1.5912, + "step": 1915 + }, + { + "epoch": 2.0997260273972604, + "grad_norm": 3.095912456512451, + "learning_rate": 1.6998375304630382e-06, + "loss": 1.6522, + "step": 1916 + }, + { + "epoch": 2.100821917808219, + "grad_norm": 3.102691888809204, + "learning_rate": 1.6978066612510155e-06, + "loss": 1.6004, + "step": 1917 + }, + { + "epoch": 2.101917808219178, + "grad_norm": 2.8405165672302246, + "learning_rate": 1.6957757920389927e-06, + "loss": 1.5649, + "step": 1918 + }, + { + "epoch": 2.103013698630137, + "grad_norm": 2.8038063049316406, + "learning_rate": 1.69374492282697e-06, + "loss": 1.545, + "step": 1919 + }, + { + "epoch": 2.1041095890410957, + "grad_norm": 2.8661468029022217, + "learning_rate": 1.6917140536149473e-06, + "loss": 1.6297, + "step": 1920 + }, + { + "epoch": 2.1041095890410957, + "eval_loss": 1.8386919498443604, + "eval_runtime": 46.0334, + "eval_samples_per_second": 21.723, + "eval_steps_per_second": 0.912, + "step": 1920 + }, + { + "epoch": 2.1052054794520547, + "grad_norm": 2.8217391967773438, + "learning_rate": 1.6896831844029246e-06, + "loss": 1.5694, + "step": 1921 + }, + { + "epoch": 2.1063013698630138, + "grad_norm": 3.099724769592285, + "learning_rate": 1.6876523151909018e-06, + "loss": 1.5986, + "step": 1922 + }, + { + "epoch": 2.107397260273973, + "grad_norm": 2.9951272010803223, + "learning_rate": 1.6856214459788791e-06, + "loss": 1.5429, + "step": 1923 + }, + { + "epoch": 2.1084931506849314, + "grad_norm": 2.9366989135742188, + "learning_rate": 1.6835905767668564e-06, + "loss": 1.5666, + "step": 1924 + }, + { + "epoch": 2.1095890410958904, + "grad_norm": 3.0171356201171875, + "learning_rate": 1.6815597075548337e-06, + "loss": 1.6212, + "step": 1925 + }, + { + "epoch": 2.1106849315068494, + "grad_norm": 2.952259063720703, + "learning_rate": 1.679528838342811e-06, + "loss": 1.5611, + "step": 1926 + }, + { + "epoch": 2.111780821917808, + "grad_norm": 2.667905330657959, + "learning_rate": 1.677497969130788e-06, + "loss": 1.5556, + "step": 1927 + }, + { + "epoch": 2.112876712328767, + "grad_norm": 2.8001160621643066, + "learning_rate": 1.6754670999187653e-06, + "loss": 1.5854, + "step": 1928 + }, + { + "epoch": 2.113972602739726, + "grad_norm": 2.6954760551452637, + "learning_rate": 1.6734362307067426e-06, + "loss": 1.5278, + "step": 1929 + }, + { + "epoch": 2.115068493150685, + "grad_norm": 2.8026583194732666, + "learning_rate": 1.6714053614947198e-06, + "loss": 1.5103, + "step": 1930 + }, + { + "epoch": 2.115068493150685, + "eval_loss": 1.8384881019592285, + "eval_runtime": 46.1899, + "eval_samples_per_second": 21.65, + "eval_steps_per_second": 0.909, + "step": 1930 + }, + { + "epoch": 2.1161643835616437, + "grad_norm": 2.898869752883911, + "learning_rate": 1.669374492282697e-06, + "loss": 1.6188, + "step": 1931 + }, + { + "epoch": 2.1172602739726027, + "grad_norm": 3.0432069301605225, + "learning_rate": 1.6673436230706744e-06, + "loss": 1.5945, + "step": 1932 + }, + { + "epoch": 2.118356164383562, + "grad_norm": 2.7744672298431396, + "learning_rate": 1.6653127538586517e-06, + "loss": 1.5642, + "step": 1933 + }, + { + "epoch": 2.1194520547945204, + "grad_norm": 2.8844289779663086, + "learning_rate": 1.663281884646629e-06, + "loss": 1.5657, + "step": 1934 + }, + { + "epoch": 2.1205479452054794, + "grad_norm": 3.0022218227386475, + "learning_rate": 1.6612510154346062e-06, + "loss": 1.6297, + "step": 1935 + }, + { + "epoch": 2.1216438356164384, + "grad_norm": 2.918320417404175, + "learning_rate": 1.6592201462225835e-06, + "loss": 1.5928, + "step": 1936 + }, + { + "epoch": 2.1227397260273975, + "grad_norm": 2.913848876953125, + "learning_rate": 1.6571892770105608e-06, + "loss": 1.6219, + "step": 1937 + }, + { + "epoch": 2.123835616438356, + "grad_norm": 2.954702615737915, + "learning_rate": 1.655158407798538e-06, + "loss": 1.5812, + "step": 1938 + }, + { + "epoch": 2.124931506849315, + "grad_norm": 2.7966420650482178, + "learning_rate": 1.653127538586515e-06, + "loss": 1.6034, + "step": 1939 + }, + { + "epoch": 2.126027397260274, + "grad_norm": 3.1328368186950684, + "learning_rate": 1.6510966693744924e-06, + "loss": 1.6168, + "step": 1940 + }, + { + "epoch": 2.126027397260274, + "eval_loss": 1.836020827293396, + "eval_runtime": 46.1979, + "eval_samples_per_second": 21.646, + "eval_steps_per_second": 0.909, + "step": 1940 + }, + { + "epoch": 2.1271232876712327, + "grad_norm": 2.8109259605407715, + "learning_rate": 1.6490658001624696e-06, + "loss": 1.6091, + "step": 1941 + }, + { + "epoch": 2.1282191780821917, + "grad_norm": 2.823570728302002, + "learning_rate": 1.647034930950447e-06, + "loss": 1.5504, + "step": 1942 + }, + { + "epoch": 2.1293150684931508, + "grad_norm": 2.885016918182373, + "learning_rate": 1.6450040617384242e-06, + "loss": 1.5784, + "step": 1943 + }, + { + "epoch": 2.1304109589041094, + "grad_norm": 3.612942934036255, + "learning_rate": 1.6429731925264015e-06, + "loss": 1.6014, + "step": 1944 + }, + { + "epoch": 2.1315068493150684, + "grad_norm": 2.8981637954711914, + "learning_rate": 1.6409423233143788e-06, + "loss": 1.5552, + "step": 1945 + }, + { + "epoch": 2.1326027397260274, + "grad_norm": 2.6703410148620605, + "learning_rate": 1.638911454102356e-06, + "loss": 1.5567, + "step": 1946 + }, + { + "epoch": 2.1336986301369865, + "grad_norm": 2.7087886333465576, + "learning_rate": 1.6368805848903333e-06, + "loss": 1.613, + "step": 1947 + }, + { + "epoch": 2.134794520547945, + "grad_norm": 2.7711687088012695, + "learning_rate": 1.6348497156783106e-06, + "loss": 1.5586, + "step": 1948 + }, + { + "epoch": 2.135890410958904, + "grad_norm": 2.8473854064941406, + "learning_rate": 1.6328188464662879e-06, + "loss": 1.565, + "step": 1949 + }, + { + "epoch": 2.136986301369863, + "grad_norm": 3.783824920654297, + "learning_rate": 1.6307879772542651e-06, + "loss": 1.5909, + "step": 1950 + }, + { + "epoch": 2.136986301369863, + "eval_loss": 1.8384798765182495, + "eval_runtime": 46.2339, + "eval_samples_per_second": 21.629, + "eval_steps_per_second": 0.908, + "step": 1950 + }, + { + "epoch": 2.138082191780822, + "grad_norm": 2.9971532821655273, + "learning_rate": 1.628757108042242e-06, + "loss": 1.6251, + "step": 1951 + }, + { + "epoch": 2.1391780821917807, + "grad_norm": 2.8581976890563965, + "learning_rate": 1.6267262388302193e-06, + "loss": 1.5632, + "step": 1952 + }, + { + "epoch": 2.1402739726027398, + "grad_norm": 2.9119839668273926, + "learning_rate": 1.6246953696181965e-06, + "loss": 1.5806, + "step": 1953 + }, + { + "epoch": 2.141369863013699, + "grad_norm": 2.840101718902588, + "learning_rate": 1.6226645004061738e-06, + "loss": 1.6706, + "step": 1954 + }, + { + "epoch": 2.1424657534246574, + "grad_norm": 2.8574070930480957, + "learning_rate": 1.620633631194151e-06, + "loss": 1.6138, + "step": 1955 + }, + { + "epoch": 2.1435616438356164, + "grad_norm": 2.820692539215088, + "learning_rate": 1.6186027619821284e-06, + "loss": 1.5534, + "step": 1956 + }, + { + "epoch": 2.1446575342465755, + "grad_norm": 2.7690017223358154, + "learning_rate": 1.6165718927701056e-06, + "loss": 1.5922, + "step": 1957 + }, + { + "epoch": 2.145753424657534, + "grad_norm": 3.062229871749878, + "learning_rate": 1.614541023558083e-06, + "loss": 1.5893, + "step": 1958 + }, + { + "epoch": 2.146849315068493, + "grad_norm": 2.727062225341797, + "learning_rate": 1.6125101543460602e-06, + "loss": 1.5851, + "step": 1959 + }, + { + "epoch": 2.147945205479452, + "grad_norm": 2.7693302631378174, + "learning_rate": 1.6104792851340375e-06, + "loss": 1.6031, + "step": 1960 + }, + { + "epoch": 2.147945205479452, + "eval_loss": 1.8332058191299438, + "eval_runtime": 45.0301, + "eval_samples_per_second": 22.207, + "eval_steps_per_second": 0.933, + "step": 1960 + }, + { + "epoch": 2.149041095890411, + "grad_norm": 2.803758382797241, + "learning_rate": 1.6084484159220147e-06, + "loss": 1.5504, + "step": 1961 + }, + { + "epoch": 2.1501369863013697, + "grad_norm": 2.775770902633667, + "learning_rate": 1.606417546709992e-06, + "loss": 1.5707, + "step": 1962 + }, + { + "epoch": 2.1512328767123288, + "grad_norm": 2.7698636054992676, + "learning_rate": 1.6043866774979693e-06, + "loss": 1.5763, + "step": 1963 + }, + { + "epoch": 2.152328767123288, + "grad_norm": 2.739940881729126, + "learning_rate": 1.6023558082859464e-06, + "loss": 1.5386, + "step": 1964 + }, + { + "epoch": 2.1534246575342464, + "grad_norm": 2.7594099044799805, + "learning_rate": 1.6003249390739236e-06, + "loss": 1.5805, + "step": 1965 + }, + { + "epoch": 2.1545205479452054, + "grad_norm": 2.853273391723633, + "learning_rate": 1.598294069861901e-06, + "loss": 1.5548, + "step": 1966 + }, + { + "epoch": 2.1556164383561645, + "grad_norm": 2.9859204292297363, + "learning_rate": 1.5962632006498782e-06, + "loss": 1.6278, + "step": 1967 + }, + { + "epoch": 2.1567123287671235, + "grad_norm": 2.742898941040039, + "learning_rate": 1.5942323314378555e-06, + "loss": 1.555, + "step": 1968 + }, + { + "epoch": 2.157808219178082, + "grad_norm": 2.769071340560913, + "learning_rate": 1.5922014622258327e-06, + "loss": 1.5504, + "step": 1969 + }, + { + "epoch": 2.158904109589041, + "grad_norm": 2.7191832065582275, + "learning_rate": 1.59017059301381e-06, + "loss": 1.6014, + "step": 1970 + }, + { + "epoch": 2.158904109589041, + "eval_loss": 1.8336049318313599, + "eval_runtime": 47.0078, + "eval_samples_per_second": 21.273, + "eval_steps_per_second": 0.893, + "step": 1970 + }, + { + "epoch": 2.16, + "grad_norm": 2.842529773712158, + "learning_rate": 1.5881397238017873e-06, + "loss": 1.5723, + "step": 1971 + }, + { + "epoch": 2.1610958904109587, + "grad_norm": 2.768066883087158, + "learning_rate": 1.5861088545897646e-06, + "loss": 1.5272, + "step": 1972 + }, + { + "epoch": 2.1621917808219178, + "grad_norm": 2.7481656074523926, + "learning_rate": 1.5840779853777418e-06, + "loss": 1.5843, + "step": 1973 + }, + { + "epoch": 2.163287671232877, + "grad_norm": 2.806687593460083, + "learning_rate": 1.5820471161657191e-06, + "loss": 1.5598, + "step": 1974 + }, + { + "epoch": 2.1643835616438354, + "grad_norm": 2.8214614391326904, + "learning_rate": 1.5800162469536964e-06, + "loss": 1.5972, + "step": 1975 + }, + { + "epoch": 2.1654794520547944, + "grad_norm": 2.8541247844696045, + "learning_rate": 1.5779853777416735e-06, + "loss": 1.5988, + "step": 1976 + }, + { + "epoch": 2.1665753424657535, + "grad_norm": 3.000246286392212, + "learning_rate": 1.5759545085296507e-06, + "loss": 1.583, + "step": 1977 + }, + { + "epoch": 2.1676712328767125, + "grad_norm": 2.9164586067199707, + "learning_rate": 1.573923639317628e-06, + "loss": 1.5949, + "step": 1978 + }, + { + "epoch": 2.168767123287671, + "grad_norm": 2.8099935054779053, + "learning_rate": 1.5718927701056053e-06, + "loss": 1.5959, + "step": 1979 + }, + { + "epoch": 2.16986301369863, + "grad_norm": 2.88322377204895, + "learning_rate": 1.5698619008935826e-06, + "loss": 1.6062, + "step": 1980 + }, + { + "epoch": 2.16986301369863, + "eval_loss": 1.832599401473999, + "eval_runtime": 45.9576, + "eval_samples_per_second": 21.759, + "eval_steps_per_second": 0.914, + "step": 1980 + }, + { + "epoch": 2.170958904109589, + "grad_norm": 2.8019371032714844, + "learning_rate": 1.5678310316815598e-06, + "loss": 1.5445, + "step": 1981 + }, + { + "epoch": 2.172054794520548, + "grad_norm": 2.8585665225982666, + "learning_rate": 1.5658001624695371e-06, + "loss": 1.6079, + "step": 1982 + }, + { + "epoch": 2.1731506849315068, + "grad_norm": 2.7423956394195557, + "learning_rate": 1.5637692932575144e-06, + "loss": 1.5537, + "step": 1983 + }, + { + "epoch": 2.174246575342466, + "grad_norm": 2.7420766353607178, + "learning_rate": 1.5617384240454917e-06, + "loss": 1.5846, + "step": 1984 + }, + { + "epoch": 2.175342465753425, + "grad_norm": 2.8864388465881348, + "learning_rate": 1.559707554833469e-06, + "loss": 1.6213, + "step": 1985 + }, + { + "epoch": 2.1764383561643834, + "grad_norm": 2.7483463287353516, + "learning_rate": 1.5576766856214462e-06, + "loss": 1.5012, + "step": 1986 + }, + { + "epoch": 2.1775342465753424, + "grad_norm": 2.7340080738067627, + "learning_rate": 1.5556458164094235e-06, + "loss": 1.555, + "step": 1987 + }, + { + "epoch": 2.1786301369863015, + "grad_norm": 2.8000667095184326, + "learning_rate": 1.5536149471974008e-06, + "loss": 1.5259, + "step": 1988 + }, + { + "epoch": 2.17972602739726, + "grad_norm": 2.9176089763641357, + "learning_rate": 1.5515840779853778e-06, + "loss": 1.5726, + "step": 1989 + }, + { + "epoch": 2.180821917808219, + "grad_norm": 2.9147658348083496, + "learning_rate": 1.549553208773355e-06, + "loss": 1.5685, + "step": 1990 + }, + { + "epoch": 2.180821917808219, + "eval_loss": 1.834657907485962, + "eval_runtime": 45.8709, + "eval_samples_per_second": 21.8, + "eval_steps_per_second": 0.916, + "step": 1990 + }, + { + "epoch": 2.181917808219178, + "grad_norm": 2.9109392166137695, + "learning_rate": 1.5475223395613324e-06, + "loss": 1.5533, + "step": 1991 + }, + { + "epoch": 2.183013698630137, + "grad_norm": 3.0263078212738037, + "learning_rate": 1.5454914703493097e-06, + "loss": 1.6277, + "step": 1992 + }, + { + "epoch": 2.1841095890410958, + "grad_norm": 2.9347317218780518, + "learning_rate": 1.543460601137287e-06, + "loss": 1.5738, + "step": 1993 + }, + { + "epoch": 2.185205479452055, + "grad_norm": 2.8229548931121826, + "learning_rate": 1.5414297319252642e-06, + "loss": 1.5969, + "step": 1994 + }, + { + "epoch": 2.186301369863014, + "grad_norm": 2.906954288482666, + "learning_rate": 1.5393988627132415e-06, + "loss": 1.5772, + "step": 1995 + }, + { + "epoch": 2.1873972602739724, + "grad_norm": 2.9214975833892822, + "learning_rate": 1.5373679935012188e-06, + "loss": 1.5147, + "step": 1996 + }, + { + "epoch": 2.1884931506849314, + "grad_norm": 2.844081163406372, + "learning_rate": 1.535337124289196e-06, + "loss": 1.5635, + "step": 1997 + }, + { + "epoch": 2.1895890410958905, + "grad_norm": 3.0214555263519287, + "learning_rate": 1.5333062550771733e-06, + "loss": 1.6466, + "step": 1998 + }, + { + "epoch": 2.1906849315068495, + "grad_norm": 2.9254050254821777, + "learning_rate": 1.5312753858651506e-06, + "loss": 1.5837, + "step": 1999 + }, + { + "epoch": 2.191780821917808, + "grad_norm": 2.8986024856567383, + "learning_rate": 1.5292445166531279e-06, + "loss": 1.5997, + "step": 2000 + }, + { + "epoch": 2.191780821917808, + "eval_loss": 1.8318188190460205, + "eval_runtime": 46.7695, + "eval_samples_per_second": 21.381, + "eval_steps_per_second": 0.898, + "step": 2000 + }, + { + "epoch": 2.192876712328767, + "grad_norm": 2.974231719970703, + "learning_rate": 1.5272136474411047e-06, + "loss": 1.6143, + "step": 2001 + }, + { + "epoch": 2.193972602739726, + "grad_norm": 2.7890946865081787, + "learning_rate": 1.525182778229082e-06, + "loss": 1.5956, + "step": 2002 + }, + { + "epoch": 2.1950684931506848, + "grad_norm": 2.820800304412842, + "learning_rate": 1.5231519090170593e-06, + "loss": 1.5626, + "step": 2003 + }, + { + "epoch": 2.196164383561644, + "grad_norm": 2.84753680229187, + "learning_rate": 1.5211210398050365e-06, + "loss": 1.5473, + "step": 2004 + }, + { + "epoch": 2.197260273972603, + "grad_norm": 2.829115152359009, + "learning_rate": 1.5190901705930138e-06, + "loss": 1.591, + "step": 2005 + }, + { + "epoch": 2.198356164383562, + "grad_norm": 2.786787509918213, + "learning_rate": 1.517059301380991e-06, + "loss": 1.565, + "step": 2006 + }, + { + "epoch": 2.1994520547945204, + "grad_norm": 2.7778282165527344, + "learning_rate": 1.5150284321689684e-06, + "loss": 1.5666, + "step": 2007 + }, + { + "epoch": 2.2005479452054795, + "grad_norm": 2.9727330207824707, + "learning_rate": 1.5129975629569456e-06, + "loss": 1.577, + "step": 2008 + }, + { + "epoch": 2.2016438356164385, + "grad_norm": 2.876600980758667, + "learning_rate": 1.510966693744923e-06, + "loss": 1.5742, + "step": 2009 + }, + { + "epoch": 2.202739726027397, + "grad_norm": 2.7339680194854736, + "learning_rate": 1.5089358245329002e-06, + "loss": 1.5378, + "step": 2010 + }, + { + "epoch": 2.202739726027397, + "eval_loss": 1.834323763847351, + "eval_runtime": 44.2017, + "eval_samples_per_second": 22.624, + "eval_steps_per_second": 0.95, + "step": 2010 + }, + { + "epoch": 2.203835616438356, + "grad_norm": 2.9201254844665527, + "learning_rate": 1.5069049553208775e-06, + "loss": 1.5962, + "step": 2011 + }, + { + "epoch": 2.204931506849315, + "grad_norm": 2.8632285594940186, + "learning_rate": 1.5048740861088547e-06, + "loss": 1.589, + "step": 2012 + }, + { + "epoch": 2.206027397260274, + "grad_norm": 2.9003729820251465, + "learning_rate": 1.502843216896832e-06, + "loss": 1.5125, + "step": 2013 + }, + { + "epoch": 2.207123287671233, + "grad_norm": 2.81120228767395, + "learning_rate": 1.500812347684809e-06, + "loss": 1.5809, + "step": 2014 + }, + { + "epoch": 2.208219178082192, + "grad_norm": 2.747528553009033, + "learning_rate": 1.4987814784727864e-06, + "loss": 1.5611, + "step": 2015 + }, + { + "epoch": 2.209315068493151, + "grad_norm": 2.6891186237335205, + "learning_rate": 1.4967506092607636e-06, + "loss": 1.5387, + "step": 2016 + }, + { + "epoch": 2.2104109589041094, + "grad_norm": 2.9105594158172607, + "learning_rate": 1.494719740048741e-06, + "loss": 1.5921, + "step": 2017 + }, + { + "epoch": 2.2115068493150685, + "grad_norm": 2.8186421394348145, + "learning_rate": 1.4926888708367182e-06, + "loss": 1.6075, + "step": 2018 + }, + { + "epoch": 2.2126027397260275, + "grad_norm": 3.0198328495025635, + "learning_rate": 1.4906580016246955e-06, + "loss": 1.611, + "step": 2019 + }, + { + "epoch": 2.213698630136986, + "grad_norm": 3.0839414596557617, + "learning_rate": 1.4886271324126727e-06, + "loss": 1.6298, + "step": 2020 + }, + { + "epoch": 2.213698630136986, + "eval_loss": 1.8325375318527222, + "eval_runtime": 53.1022, + "eval_samples_per_second": 18.832, + "eval_steps_per_second": 0.791, + "step": 2020 + }, + { + "epoch": 2.214794520547945, + "grad_norm": 2.8375303745269775, + "learning_rate": 1.48659626320065e-06, + "loss": 1.4942, + "step": 2021 + }, + { + "epoch": 2.215890410958904, + "grad_norm": 2.857344627380371, + "learning_rate": 1.4845653939886273e-06, + "loss": 1.5182, + "step": 2022 + }, + { + "epoch": 2.216986301369863, + "grad_norm": 2.7296979427337646, + "learning_rate": 1.4825345247766046e-06, + "loss": 1.5081, + "step": 2023 + }, + { + "epoch": 2.2180821917808218, + "grad_norm": 2.9989287853240967, + "learning_rate": 1.4805036555645818e-06, + "loss": 1.5577, + "step": 2024 + }, + { + "epoch": 2.219178082191781, + "grad_norm": 2.9123456478118896, + "learning_rate": 1.4784727863525591e-06, + "loss": 1.5339, + "step": 2025 + }, + { + "epoch": 2.22027397260274, + "grad_norm": 2.845898151397705, + "learning_rate": 1.4764419171405362e-06, + "loss": 1.6084, + "step": 2026 + }, + { + "epoch": 2.2213698630136984, + "grad_norm": 3.089202404022217, + "learning_rate": 1.4744110479285135e-06, + "loss": 1.6133, + "step": 2027 + }, + { + "epoch": 2.2224657534246575, + "grad_norm": 2.9908018112182617, + "learning_rate": 1.4723801787164907e-06, + "loss": 1.5984, + "step": 2028 + }, + { + "epoch": 2.2235616438356165, + "grad_norm": 2.838611125946045, + "learning_rate": 1.470349309504468e-06, + "loss": 1.5971, + "step": 2029 + }, + { + "epoch": 2.2246575342465755, + "grad_norm": 2.902926445007324, + "learning_rate": 1.4683184402924453e-06, + "loss": 1.6029, + "step": 2030 + }, + { + "epoch": 2.2246575342465755, + "eval_loss": 1.8312546014785767, + "eval_runtime": 46.3457, + "eval_samples_per_second": 21.577, + "eval_steps_per_second": 0.906, + "step": 2030 + }, + { + "epoch": 2.225753424657534, + "grad_norm": 2.8279829025268555, + "learning_rate": 1.4662875710804226e-06, + "loss": 1.65, + "step": 2031 + }, + { + "epoch": 2.226849315068493, + "grad_norm": 2.7380146980285645, + "learning_rate": 1.4642567018683998e-06, + "loss": 1.5673, + "step": 2032 + }, + { + "epoch": 2.227945205479452, + "grad_norm": 2.723890781402588, + "learning_rate": 1.4622258326563771e-06, + "loss": 1.5573, + "step": 2033 + }, + { + "epoch": 2.2290410958904108, + "grad_norm": 2.8826022148132324, + "learning_rate": 1.4601949634443544e-06, + "loss": 1.5648, + "step": 2034 + }, + { + "epoch": 2.23013698630137, + "grad_norm": 2.948598861694336, + "learning_rate": 1.4581640942323317e-06, + "loss": 1.6603, + "step": 2035 + }, + { + "epoch": 2.231232876712329, + "grad_norm": 2.8282432556152344, + "learning_rate": 1.456133225020309e-06, + "loss": 1.5875, + "step": 2036 + }, + { + "epoch": 2.232328767123288, + "grad_norm": 2.874748468399048, + "learning_rate": 1.4541023558082862e-06, + "loss": 1.5713, + "step": 2037 + }, + { + "epoch": 2.2334246575342465, + "grad_norm": 2.894249677658081, + "learning_rate": 1.4520714865962635e-06, + "loss": 1.6063, + "step": 2038 + }, + { + "epoch": 2.2345205479452055, + "grad_norm": 2.907095193862915, + "learning_rate": 1.4500406173842406e-06, + "loss": 1.5744, + "step": 2039 + }, + { + "epoch": 2.2356164383561645, + "grad_norm": 2.892090082168579, + "learning_rate": 1.4480097481722178e-06, + "loss": 1.5877, + "step": 2040 + }, + { + "epoch": 2.2356164383561645, + "eval_loss": 1.8305796384811401, + "eval_runtime": 48.3583, + "eval_samples_per_second": 20.679, + "eval_steps_per_second": 0.869, + "step": 2040 + }, + { + "epoch": 2.236712328767123, + "grad_norm": 2.7991275787353516, + "learning_rate": 1.4459788789601951e-06, + "loss": 1.5475, + "step": 2041 + }, + { + "epoch": 2.237808219178082, + "grad_norm": 2.9188716411590576, + "learning_rate": 1.4439480097481724e-06, + "loss": 1.5385, + "step": 2042 + }, + { + "epoch": 2.238904109589041, + "grad_norm": 2.8575944900512695, + "learning_rate": 1.4419171405361497e-06, + "loss": 1.5855, + "step": 2043 + }, + { + "epoch": 2.24, + "grad_norm": 2.814990520477295, + "learning_rate": 1.439886271324127e-06, + "loss": 1.5923, + "step": 2044 + }, + { + "epoch": 2.241095890410959, + "grad_norm": 2.7953567504882812, + "learning_rate": 1.4378554021121042e-06, + "loss": 1.5724, + "step": 2045 + }, + { + "epoch": 2.242191780821918, + "grad_norm": 2.882002115249634, + "learning_rate": 1.4358245329000815e-06, + "loss": 1.6086, + "step": 2046 + }, + { + "epoch": 2.243287671232877, + "grad_norm": 2.910799503326416, + "learning_rate": 1.4337936636880588e-06, + "loss": 1.5933, + "step": 2047 + }, + { + "epoch": 2.2443835616438355, + "grad_norm": 2.911015748977661, + "learning_rate": 1.431762794476036e-06, + "loss": 1.5356, + "step": 2048 + }, + { + "epoch": 2.2454794520547945, + "grad_norm": 2.9788129329681396, + "learning_rate": 1.4297319252640133e-06, + "loss": 1.5546, + "step": 2049 + }, + { + "epoch": 2.2465753424657535, + "grad_norm": 2.8637843132019043, + "learning_rate": 1.4277010560519906e-06, + "loss": 1.5423, + "step": 2050 + }, + { + "epoch": 2.2465753424657535, + "eval_loss": 1.8307991027832031, + "eval_runtime": 45.4486, + "eval_samples_per_second": 22.003, + "eval_steps_per_second": 0.924, + "step": 2050 + }, + { + "epoch": 2.247671232876712, + "grad_norm": 3.007390260696411, + "learning_rate": 1.4256701868399674e-06, + "loss": 1.6244, + "step": 2051 + }, + { + "epoch": 2.248767123287671, + "grad_norm": 2.908496856689453, + "learning_rate": 1.4236393176279447e-06, + "loss": 1.5767, + "step": 2052 + }, + { + "epoch": 2.24986301369863, + "grad_norm": 2.9650135040283203, + "learning_rate": 1.421608448415922e-06, + "loss": 1.6639, + "step": 2053 + }, + { + "epoch": 2.250958904109589, + "grad_norm": 2.8085806369781494, + "learning_rate": 1.4195775792038993e-06, + "loss": 1.5486, + "step": 2054 + }, + { + "epoch": 2.252054794520548, + "grad_norm": 2.7759034633636475, + "learning_rate": 1.4175467099918765e-06, + "loss": 1.5847, + "step": 2055 + }, + { + "epoch": 2.253150684931507, + "grad_norm": 2.9309468269348145, + "learning_rate": 1.4155158407798538e-06, + "loss": 1.6014, + "step": 2056 + }, + { + "epoch": 2.254246575342466, + "grad_norm": 2.8542673587799072, + "learning_rate": 1.413484971567831e-06, + "loss": 1.5183, + "step": 2057 + }, + { + "epoch": 2.255342465753425, + "grad_norm": 3.0022106170654297, + "learning_rate": 1.4114541023558084e-06, + "loss": 1.5511, + "step": 2058 + }, + { + "epoch": 2.2564383561643835, + "grad_norm": 2.826852321624756, + "learning_rate": 1.4094232331437856e-06, + "loss": 1.6079, + "step": 2059 + }, + { + "epoch": 2.2575342465753425, + "grad_norm": 2.8682327270507812, + "learning_rate": 1.407392363931763e-06, + "loss": 1.6579, + "step": 2060 + }, + { + "epoch": 2.2575342465753425, + "eval_loss": 1.8299224376678467, + "eval_runtime": 44.6401, + "eval_samples_per_second": 22.401, + "eval_steps_per_second": 0.941, + "step": 2060 + }, + { + "epoch": 2.2586301369863016, + "grad_norm": 2.8816471099853516, + "learning_rate": 1.4053614947197402e-06, + "loss": 1.5825, + "step": 2061 + }, + { + "epoch": 2.25972602739726, + "grad_norm": 2.696047306060791, + "learning_rate": 1.4033306255077175e-06, + "loss": 1.5068, + "step": 2062 + }, + { + "epoch": 2.260821917808219, + "grad_norm": 2.9243481159210205, + "learning_rate": 1.4012997562956945e-06, + "loss": 1.6172, + "step": 2063 + }, + { + "epoch": 2.261917808219178, + "grad_norm": 2.810744285583496, + "learning_rate": 1.3992688870836718e-06, + "loss": 1.5976, + "step": 2064 + }, + { + "epoch": 2.263013698630137, + "grad_norm": 2.7787344455718994, + "learning_rate": 1.397238017871649e-06, + "loss": 1.5829, + "step": 2065 + }, + { + "epoch": 2.264109589041096, + "grad_norm": 2.817336082458496, + "learning_rate": 1.3952071486596264e-06, + "loss": 1.5482, + "step": 2066 + }, + { + "epoch": 2.265205479452055, + "grad_norm": 2.7691237926483154, + "learning_rate": 1.3931762794476036e-06, + "loss": 1.6098, + "step": 2067 + }, + { + "epoch": 2.266301369863014, + "grad_norm": 2.7974281311035156, + "learning_rate": 1.391145410235581e-06, + "loss": 1.5591, + "step": 2068 + }, + { + "epoch": 2.2673972602739725, + "grad_norm": 2.8233275413513184, + "learning_rate": 1.3891145410235582e-06, + "loss": 1.4975, + "step": 2069 + }, + { + "epoch": 2.2684931506849315, + "grad_norm": 2.7465834617614746, + "learning_rate": 1.3870836718115355e-06, + "loss": 1.5682, + "step": 2070 + }, + { + "epoch": 2.2684931506849315, + "eval_loss": 1.8317815065383911, + "eval_runtime": 44.8326, + "eval_samples_per_second": 22.305, + "eval_steps_per_second": 0.937, + "step": 2070 + }, + { + "epoch": 2.2695890410958905, + "grad_norm": 2.9384610652923584, + "learning_rate": 1.3850528025995127e-06, + "loss": 1.5627, + "step": 2071 + }, + { + "epoch": 2.270684931506849, + "grad_norm": 2.6946020126342773, + "learning_rate": 1.38302193338749e-06, + "loss": 1.4851, + "step": 2072 + }, + { + "epoch": 2.271780821917808, + "grad_norm": 2.875818967819214, + "learning_rate": 1.3809910641754673e-06, + "loss": 1.595, + "step": 2073 + }, + { + "epoch": 2.272876712328767, + "grad_norm": 2.871393918991089, + "learning_rate": 1.3789601949634446e-06, + "loss": 1.5297, + "step": 2074 + }, + { + "epoch": 2.2739726027397262, + "grad_norm": 2.8451645374298096, + "learning_rate": 1.3769293257514218e-06, + "loss": 1.5912, + "step": 2075 + }, + { + "epoch": 2.275068493150685, + "grad_norm": 2.808422803878784, + "learning_rate": 1.374898456539399e-06, + "loss": 1.5316, + "step": 2076 + }, + { + "epoch": 2.276164383561644, + "grad_norm": 2.866929292678833, + "learning_rate": 1.3728675873273762e-06, + "loss": 1.6223, + "step": 2077 + }, + { + "epoch": 2.277260273972603, + "grad_norm": 2.969078302383423, + "learning_rate": 1.3708367181153535e-06, + "loss": 1.5855, + "step": 2078 + }, + { + "epoch": 2.2783561643835615, + "grad_norm": 2.739272117614746, + "learning_rate": 1.3688058489033307e-06, + "loss": 1.5915, + "step": 2079 + }, + { + "epoch": 2.2794520547945205, + "grad_norm": 2.9254791736602783, + "learning_rate": 1.366774979691308e-06, + "loss": 1.5989, + "step": 2080 + }, + { + "epoch": 2.2794520547945205, + "eval_loss": 1.8284611701965332, + "eval_runtime": 45.0516, + "eval_samples_per_second": 22.197, + "eval_steps_per_second": 0.932, + "step": 2080 + }, + { + "epoch": 2.2805479452054795, + "grad_norm": 2.7639904022216797, + "learning_rate": 1.3647441104792853e-06, + "loss": 1.5735, + "step": 2081 + }, + { + "epoch": 2.281643835616438, + "grad_norm": 2.782290458679199, + "learning_rate": 1.3627132412672626e-06, + "loss": 1.5365, + "step": 2082 + }, + { + "epoch": 2.282739726027397, + "grad_norm": 2.8718478679656982, + "learning_rate": 1.3606823720552398e-06, + "loss": 1.5776, + "step": 2083 + }, + { + "epoch": 2.283835616438356, + "grad_norm": 2.770900249481201, + "learning_rate": 1.3586515028432171e-06, + "loss": 1.5242, + "step": 2084 + }, + { + "epoch": 2.2849315068493152, + "grad_norm": 2.7527379989624023, + "learning_rate": 1.3566206336311944e-06, + "loss": 1.5783, + "step": 2085 + }, + { + "epoch": 2.286027397260274, + "grad_norm": 2.724205732345581, + "learning_rate": 1.3545897644191717e-06, + "loss": 1.5556, + "step": 2086 + }, + { + "epoch": 2.287123287671233, + "grad_norm": 2.7595510482788086, + "learning_rate": 1.352558895207149e-06, + "loss": 1.6084, + "step": 2087 + }, + { + "epoch": 2.288219178082192, + "grad_norm": 2.832383394241333, + "learning_rate": 1.3505280259951258e-06, + "loss": 1.5755, + "step": 2088 + }, + { + "epoch": 2.289315068493151, + "grad_norm": 2.8154635429382324, + "learning_rate": 1.348497156783103e-06, + "loss": 1.6091, + "step": 2089 + }, + { + "epoch": 2.2904109589041095, + "grad_norm": 2.808480739593506, + "learning_rate": 1.3464662875710804e-06, + "loss": 1.5403, + "step": 2090 + }, + { + "epoch": 2.2904109589041095, + "eval_loss": 1.8285717964172363, + "eval_runtime": 44.0301, + "eval_samples_per_second": 22.712, + "eval_steps_per_second": 0.954, + "step": 2090 + }, + { + "epoch": 2.2915068493150685, + "grad_norm": 2.7877376079559326, + "learning_rate": 1.3444354183590576e-06, + "loss": 1.5985, + "step": 2091 + }, + { + "epoch": 2.2926027397260276, + "grad_norm": 2.854957103729248, + "learning_rate": 1.342404549147035e-06, + "loss": 1.5576, + "step": 2092 + }, + { + "epoch": 2.293698630136986, + "grad_norm": 2.8522772789001465, + "learning_rate": 1.3403736799350122e-06, + "loss": 1.5914, + "step": 2093 + }, + { + "epoch": 2.294794520547945, + "grad_norm": 2.866199493408203, + "learning_rate": 1.3383428107229895e-06, + "loss": 1.6094, + "step": 2094 + }, + { + "epoch": 2.2958904109589042, + "grad_norm": 2.9277379512786865, + "learning_rate": 1.3363119415109667e-06, + "loss": 1.6065, + "step": 2095 + }, + { + "epoch": 2.296986301369863, + "grad_norm": 2.8854689598083496, + "learning_rate": 1.334281072298944e-06, + "loss": 1.5892, + "step": 2096 + }, + { + "epoch": 2.298082191780822, + "grad_norm": 2.7135169506073, + "learning_rate": 1.3322502030869213e-06, + "loss": 1.5733, + "step": 2097 + }, + { + "epoch": 2.299178082191781, + "grad_norm": 2.8702995777130127, + "learning_rate": 1.3302193338748986e-06, + "loss": 1.6054, + "step": 2098 + }, + { + "epoch": 2.30027397260274, + "grad_norm": 2.8330469131469727, + "learning_rate": 1.3281884646628758e-06, + "loss": 1.5902, + "step": 2099 + }, + { + "epoch": 2.3013698630136985, + "grad_norm": 3.016582727432251, + "learning_rate": 1.3261575954508531e-06, + "loss": 1.5931, + "step": 2100 + }, + { + "epoch": 2.3013698630136985, + "eval_loss": 1.82985520362854, + "eval_runtime": 45.8601, + "eval_samples_per_second": 21.805, + "eval_steps_per_second": 0.916, + "step": 2100 + }, + { + "epoch": 2.3024657534246575, + "grad_norm": 2.8396756649017334, + "learning_rate": 1.3241267262388302e-06, + "loss": 1.5882, + "step": 2101 + }, + { + "epoch": 2.3035616438356166, + "grad_norm": 2.7104387283325195, + "learning_rate": 1.3220958570268074e-06, + "loss": 1.5223, + "step": 2102 + }, + { + "epoch": 2.304657534246575, + "grad_norm": 2.787862777709961, + "learning_rate": 1.3200649878147847e-06, + "loss": 1.5549, + "step": 2103 + }, + { + "epoch": 2.305753424657534, + "grad_norm": 2.9957401752471924, + "learning_rate": 1.318034118602762e-06, + "loss": 1.6036, + "step": 2104 + }, + { + "epoch": 2.3068493150684932, + "grad_norm": 2.916990280151367, + "learning_rate": 1.3160032493907393e-06, + "loss": 1.5947, + "step": 2105 + }, + { + "epoch": 2.3079452054794523, + "grad_norm": 2.972477912902832, + "learning_rate": 1.3139723801787166e-06, + "loss": 1.624, + "step": 2106 + }, + { + "epoch": 2.309041095890411, + "grad_norm": 2.84188175201416, + "learning_rate": 1.3119415109666938e-06, + "loss": 1.5708, + "step": 2107 + }, + { + "epoch": 2.31013698630137, + "grad_norm": 2.908295154571533, + "learning_rate": 1.309910641754671e-06, + "loss": 1.5412, + "step": 2108 + }, + { + "epoch": 2.311232876712329, + "grad_norm": 2.8915605545043945, + "learning_rate": 1.3078797725426484e-06, + "loss": 1.5463, + "step": 2109 + }, + { + "epoch": 2.3123287671232875, + "grad_norm": 2.7805707454681396, + "learning_rate": 1.3058489033306257e-06, + "loss": 1.5555, + "step": 2110 + }, + { + "epoch": 2.3123287671232875, + "eval_loss": 1.8281073570251465, + "eval_runtime": 44.7855, + "eval_samples_per_second": 22.329, + "eval_steps_per_second": 0.938, + "step": 2110 + }, + { + "epoch": 2.3134246575342465, + "grad_norm": 2.933260679244995, + "learning_rate": 1.303818034118603e-06, + "loss": 1.578, + "step": 2111 + }, + { + "epoch": 2.3145205479452056, + "grad_norm": 2.970050096511841, + "learning_rate": 1.3017871649065802e-06, + "loss": 1.5606, + "step": 2112 + }, + { + "epoch": 2.315616438356164, + "grad_norm": 2.717027187347412, + "learning_rate": 1.2997562956945573e-06, + "loss": 1.5446, + "step": 2113 + }, + { + "epoch": 2.316712328767123, + "grad_norm": 2.966905355453491, + "learning_rate": 1.2977254264825345e-06, + "loss": 1.549, + "step": 2114 + }, + { + "epoch": 2.317808219178082, + "grad_norm": 2.847358465194702, + "learning_rate": 1.2956945572705118e-06, + "loss": 1.581, + "step": 2115 + }, + { + "epoch": 2.3189041095890413, + "grad_norm": 2.814897060394287, + "learning_rate": 1.293663688058489e-06, + "loss": 1.5416, + "step": 2116 + }, + { + "epoch": 2.32, + "grad_norm": 2.788320302963257, + "learning_rate": 1.2916328188464664e-06, + "loss": 1.5018, + "step": 2117 + }, + { + "epoch": 2.321095890410959, + "grad_norm": 2.944441556930542, + "learning_rate": 1.2896019496344436e-06, + "loss": 1.5318, + "step": 2118 + }, + { + "epoch": 2.322191780821918, + "grad_norm": 2.8705525398254395, + "learning_rate": 1.287571080422421e-06, + "loss": 1.5953, + "step": 2119 + }, + { + "epoch": 2.323287671232877, + "grad_norm": 2.7703897953033447, + "learning_rate": 1.2855402112103982e-06, + "loss": 1.5287, + "step": 2120 + }, + { + "epoch": 2.323287671232877, + "eval_loss": 1.8304927349090576, + "eval_runtime": 47.1736, + "eval_samples_per_second": 21.198, + "eval_steps_per_second": 0.89, + "step": 2120 + }, + { + "epoch": 2.3243835616438355, + "grad_norm": 2.862663745880127, + "learning_rate": 1.2835093419983755e-06, + "loss": 1.541, + "step": 2121 + }, + { + "epoch": 2.3254794520547946, + "grad_norm": 2.8288817405700684, + "learning_rate": 1.2814784727863528e-06, + "loss": 1.5466, + "step": 2122 + }, + { + "epoch": 2.3265753424657536, + "grad_norm": 3.0720043182373047, + "learning_rate": 1.27944760357433e-06, + "loss": 1.6106, + "step": 2123 + }, + { + "epoch": 2.327671232876712, + "grad_norm": 2.9464635848999023, + "learning_rate": 1.2774167343623073e-06, + "loss": 1.6051, + "step": 2124 + }, + { + "epoch": 2.328767123287671, + "grad_norm": 2.9039576053619385, + "learning_rate": 1.2753858651502846e-06, + "loss": 1.5525, + "step": 2125 + }, + { + "epoch": 2.3298630136986302, + "grad_norm": 2.8839735984802246, + "learning_rate": 1.2733549959382616e-06, + "loss": 1.5551, + "step": 2126 + }, + { + "epoch": 2.330958904109589, + "grad_norm": 2.915053129196167, + "learning_rate": 1.271324126726239e-06, + "loss": 1.5853, + "step": 2127 + }, + { + "epoch": 2.332054794520548, + "grad_norm": 2.862844467163086, + "learning_rate": 1.2692932575142162e-06, + "loss": 1.5233, + "step": 2128 + }, + { + "epoch": 2.333150684931507, + "grad_norm": 2.717557430267334, + "learning_rate": 1.2672623883021935e-06, + "loss": 1.5138, + "step": 2129 + }, + { + "epoch": 2.334246575342466, + "grad_norm": 3.1381704807281494, + "learning_rate": 1.2652315190901707e-06, + "loss": 1.6073, + "step": 2130 + }, + { + "epoch": 2.334246575342466, + "eval_loss": 1.82700514793396, + "eval_runtime": 51.3349, + "eval_samples_per_second": 19.48, + "eval_steps_per_second": 0.818, + "step": 2130 + }, + { + "epoch": 2.3353424657534245, + "grad_norm": 2.8603286743164062, + "learning_rate": 1.263200649878148e-06, + "loss": 1.5713, + "step": 2131 + }, + { + "epoch": 2.3364383561643836, + "grad_norm": 2.9590084552764893, + "learning_rate": 1.2611697806661253e-06, + "loss": 1.5952, + "step": 2132 + }, + { + "epoch": 2.3375342465753426, + "grad_norm": 2.8419439792633057, + "learning_rate": 1.2591389114541026e-06, + "loss": 1.5729, + "step": 2133 + }, + { + "epoch": 2.338630136986301, + "grad_norm": 2.7028636932373047, + "learning_rate": 1.2571080422420798e-06, + "loss": 1.5427, + "step": 2134 + }, + { + "epoch": 2.33972602739726, + "grad_norm": 2.819112539291382, + "learning_rate": 1.2550771730300571e-06, + "loss": 1.6221, + "step": 2135 + }, + { + "epoch": 2.3408219178082192, + "grad_norm": 2.821469783782959, + "learning_rate": 1.2530463038180344e-06, + "loss": 1.5756, + "step": 2136 + }, + { + "epoch": 2.3419178082191783, + "grad_norm": 2.979747772216797, + "learning_rate": 1.2510154346060117e-06, + "loss": 1.5662, + "step": 2137 + }, + { + "epoch": 2.343013698630137, + "grad_norm": 3.176698684692383, + "learning_rate": 1.2489845653939887e-06, + "loss": 1.591, + "step": 2138 + }, + { + "epoch": 2.344109589041096, + "grad_norm": 2.8962442874908447, + "learning_rate": 1.246953696181966e-06, + "loss": 1.6549, + "step": 2139 + }, + { + "epoch": 2.345205479452055, + "grad_norm": 2.840876817703247, + "learning_rate": 1.2449228269699433e-06, + "loss": 1.5636, + "step": 2140 + }, + { + "epoch": 2.345205479452055, + "eval_loss": 1.8279746770858765, + "eval_runtime": 45.3966, + "eval_samples_per_second": 22.028, + "eval_steps_per_second": 0.925, + "step": 2140 + }, + { + "epoch": 2.3463013698630135, + "grad_norm": 2.8487303256988525, + "learning_rate": 1.2428919577579204e-06, + "loss": 1.5345, + "step": 2141 + }, + { + "epoch": 2.3473972602739726, + "grad_norm": 2.8009743690490723, + "learning_rate": 1.2408610885458976e-06, + "loss": 1.5673, + "step": 2142 + }, + { + "epoch": 2.3484931506849316, + "grad_norm": 2.7863199710845947, + "learning_rate": 1.238830219333875e-06, + "loss": 1.5351, + "step": 2143 + }, + { + "epoch": 2.34958904109589, + "grad_norm": 3.4834156036376953, + "learning_rate": 1.2367993501218522e-06, + "loss": 1.6171, + "step": 2144 + }, + { + "epoch": 2.350684931506849, + "grad_norm": 2.9101381301879883, + "learning_rate": 1.2347684809098295e-06, + "loss": 1.6026, + "step": 2145 + }, + { + "epoch": 2.3517808219178082, + "grad_norm": 2.750434160232544, + "learning_rate": 1.2327376116978067e-06, + "loss": 1.5199, + "step": 2146 + }, + { + "epoch": 2.3528767123287673, + "grad_norm": 2.8388402462005615, + "learning_rate": 1.230706742485784e-06, + "loss": 1.5253, + "step": 2147 + }, + { + "epoch": 2.353972602739726, + "grad_norm": 3.000380754470825, + "learning_rate": 1.2286758732737613e-06, + "loss": 1.6463, + "step": 2148 + }, + { + "epoch": 2.355068493150685, + "grad_norm": 3.0284345149993896, + "learning_rate": 1.2266450040617386e-06, + "loss": 1.637, + "step": 2149 + }, + { + "epoch": 2.356164383561644, + "grad_norm": 2.9016034603118896, + "learning_rate": 1.2246141348497158e-06, + "loss": 1.5794, + "step": 2150 + }, + { + "epoch": 2.356164383561644, + "eval_loss": 1.8288570642471313, + "eval_runtime": 47.0365, + "eval_samples_per_second": 21.26, + "eval_steps_per_second": 0.893, + "step": 2150 + }, + { + "epoch": 2.357260273972603, + "grad_norm": 3.27822208404541, + "learning_rate": 1.2225832656376931e-06, + "loss": 1.5734, + "step": 2151 + }, + { + "epoch": 2.3583561643835615, + "grad_norm": 2.8195855617523193, + "learning_rate": 1.2205523964256704e-06, + "loss": 1.5512, + "step": 2152 + }, + { + "epoch": 2.3594520547945206, + "grad_norm": 2.7829315662384033, + "learning_rate": 1.2185215272136477e-06, + "loss": 1.5804, + "step": 2153 + }, + { + "epoch": 2.3605479452054796, + "grad_norm": 2.8760147094726562, + "learning_rate": 1.2164906580016247e-06, + "loss": 1.5858, + "step": 2154 + }, + { + "epoch": 2.361643835616438, + "grad_norm": 2.8194808959960938, + "learning_rate": 1.214459788789602e-06, + "loss": 1.485, + "step": 2155 + }, + { + "epoch": 2.3627397260273972, + "grad_norm": 2.901355266571045, + "learning_rate": 1.2124289195775793e-06, + "loss": 1.5287, + "step": 2156 + }, + { + "epoch": 2.3638356164383563, + "grad_norm": 3.0580921173095703, + "learning_rate": 1.2103980503655566e-06, + "loss": 1.5596, + "step": 2157 + }, + { + "epoch": 2.364931506849315, + "grad_norm": 3.0313403606414795, + "learning_rate": 1.2083671811535338e-06, + "loss": 1.5536, + "step": 2158 + }, + { + "epoch": 2.366027397260274, + "grad_norm": 2.7663581371307373, + "learning_rate": 1.2063363119415111e-06, + "loss": 1.5479, + "step": 2159 + }, + { + "epoch": 2.367123287671233, + "grad_norm": 2.7729651927948, + "learning_rate": 1.2043054427294882e-06, + "loss": 1.56, + "step": 2160 + }, + { + "epoch": 2.367123287671233, + "eval_loss": 1.8264360427856445, + "eval_runtime": 45.8995, + "eval_samples_per_second": 21.787, + "eval_steps_per_second": 0.915, + "step": 2160 + }, + { + "epoch": 2.368219178082192, + "grad_norm": 3.0862157344818115, + "learning_rate": 1.2022745735174654e-06, + "loss": 1.5869, + "step": 2161 + }, + { + "epoch": 2.3693150684931505, + "grad_norm": 2.9891650676727295, + "learning_rate": 1.2002437043054427e-06, + "loss": 1.6249, + "step": 2162 + }, + { + "epoch": 2.3704109589041096, + "grad_norm": 2.8348777294158936, + "learning_rate": 1.19821283509342e-06, + "loss": 1.6116, + "step": 2163 + }, + { + "epoch": 2.3715068493150686, + "grad_norm": 2.915382146835327, + "learning_rate": 1.1961819658813973e-06, + "loss": 1.5783, + "step": 2164 + }, + { + "epoch": 2.3726027397260276, + "grad_norm": 3.086989402770996, + "learning_rate": 1.1941510966693746e-06, + "loss": 1.5473, + "step": 2165 + }, + { + "epoch": 2.3736986301369862, + "grad_norm": 2.8165829181671143, + "learning_rate": 1.1921202274573518e-06, + "loss": 1.5617, + "step": 2166 + }, + { + "epoch": 2.3747945205479453, + "grad_norm": 3.121267795562744, + "learning_rate": 1.190089358245329e-06, + "loss": 1.6125, + "step": 2167 + }, + { + "epoch": 2.3758904109589043, + "grad_norm": 2.889802932739258, + "learning_rate": 1.1880584890333064e-06, + "loss": 1.5637, + "step": 2168 + }, + { + "epoch": 2.376986301369863, + "grad_norm": 2.8893845081329346, + "learning_rate": 1.1860276198212837e-06, + "loss": 1.5929, + "step": 2169 + }, + { + "epoch": 2.378082191780822, + "grad_norm": 2.900737762451172, + "learning_rate": 1.183996750609261e-06, + "loss": 1.5798, + "step": 2170 + }, + { + "epoch": 2.378082191780822, + "eval_loss": 1.826514720916748, + "eval_runtime": 46.0672, + "eval_samples_per_second": 21.707, + "eval_steps_per_second": 0.912, + "step": 2170 + }, + { + "epoch": 2.379178082191781, + "grad_norm": 2.8097946643829346, + "learning_rate": 1.1819658813972382e-06, + "loss": 1.5653, + "step": 2171 + }, + { + "epoch": 2.3802739726027395, + "grad_norm": 2.82616925239563, + "learning_rate": 1.1799350121852153e-06, + "loss": 1.612, + "step": 2172 + }, + { + "epoch": 2.3813698630136986, + "grad_norm": 3.3667945861816406, + "learning_rate": 1.1779041429731925e-06, + "loss": 1.6107, + "step": 2173 + }, + { + "epoch": 2.3824657534246576, + "grad_norm": 2.9167160987854004, + "learning_rate": 1.1758732737611698e-06, + "loss": 1.5424, + "step": 2174 + }, + { + "epoch": 2.383561643835616, + "grad_norm": 2.9599924087524414, + "learning_rate": 1.173842404549147e-06, + "loss": 1.6195, + "step": 2175 + }, + { + "epoch": 2.3846575342465752, + "grad_norm": 2.8982062339782715, + "learning_rate": 1.1718115353371244e-06, + "loss": 1.5579, + "step": 2176 + }, + { + "epoch": 2.3857534246575343, + "grad_norm": 2.915224075317383, + "learning_rate": 1.1697806661251016e-06, + "loss": 1.6204, + "step": 2177 + }, + { + "epoch": 2.3868493150684933, + "grad_norm": 2.9213180541992188, + "learning_rate": 1.167749796913079e-06, + "loss": 1.5988, + "step": 2178 + }, + { + "epoch": 2.387945205479452, + "grad_norm": 2.9090816974639893, + "learning_rate": 1.1657189277010562e-06, + "loss": 1.5773, + "step": 2179 + }, + { + "epoch": 2.389041095890411, + "grad_norm": 2.8555307388305664, + "learning_rate": 1.1636880584890335e-06, + "loss": 1.5687, + "step": 2180 + }, + { + "epoch": 2.389041095890411, + "eval_loss": 1.8246150016784668, + "eval_runtime": 45.9157, + "eval_samples_per_second": 21.779, + "eval_steps_per_second": 0.915, + "step": 2180 + }, + { + "epoch": 2.39013698630137, + "grad_norm": 2.8956985473632812, + "learning_rate": 1.1616571892770108e-06, + "loss": 1.5878, + "step": 2181 + }, + { + "epoch": 2.391232876712329, + "grad_norm": 4.177687644958496, + "learning_rate": 1.159626320064988e-06, + "loss": 1.579, + "step": 2182 + }, + { + "epoch": 2.3923287671232876, + "grad_norm": 2.9070162773132324, + "learning_rate": 1.1575954508529653e-06, + "loss": 1.5926, + "step": 2183 + }, + { + "epoch": 2.3934246575342466, + "grad_norm": 2.7478721141815186, + "learning_rate": 1.1555645816409426e-06, + "loss": 1.5898, + "step": 2184 + }, + { + "epoch": 2.3945205479452056, + "grad_norm": 2.6728436946868896, + "learning_rate": 1.1535337124289196e-06, + "loss": 1.5345, + "step": 2185 + }, + { + "epoch": 2.3956164383561642, + "grad_norm": 2.809030532836914, + "learning_rate": 1.151502843216897e-06, + "loss": 1.5174, + "step": 2186 + }, + { + "epoch": 2.3967123287671233, + "grad_norm": 3.078012466430664, + "learning_rate": 1.1494719740048742e-06, + "loss": 1.5914, + "step": 2187 + }, + { + "epoch": 2.3978082191780823, + "grad_norm": 4.21104621887207, + "learning_rate": 1.1474411047928515e-06, + "loss": 1.5694, + "step": 2188 + }, + { + "epoch": 2.398904109589041, + "grad_norm": 3.0975046157836914, + "learning_rate": 1.1454102355808287e-06, + "loss": 1.545, + "step": 2189 + }, + { + "epoch": 2.4, + "grad_norm": 2.925477981567383, + "learning_rate": 1.143379366368806e-06, + "loss": 1.5396, + "step": 2190 + }, + { + "epoch": 2.4, + "eval_loss": 1.8294752836227417, + "eval_runtime": 47.0872, + "eval_samples_per_second": 21.237, + "eval_steps_per_second": 0.892, + "step": 2190 + }, + { + "epoch": 2.401095890410959, + "grad_norm": 3.391608238220215, + "learning_rate": 1.141348497156783e-06, + "loss": 1.6501, + "step": 2191 + }, + { + "epoch": 2.402191780821918, + "grad_norm": 3.053389549255371, + "learning_rate": 1.1393176279447604e-06, + "loss": 1.6376, + "step": 2192 + }, + { + "epoch": 2.4032876712328766, + "grad_norm": 2.8790252208709717, + "learning_rate": 1.1372867587327376e-06, + "loss": 1.5002, + "step": 2193 + }, + { + "epoch": 2.4043835616438356, + "grad_norm": 3.0458412170410156, + "learning_rate": 1.135255889520715e-06, + "loss": 1.5955, + "step": 2194 + }, + { + "epoch": 2.4054794520547946, + "grad_norm": 2.8105313777923584, + "learning_rate": 1.1332250203086922e-06, + "loss": 1.5849, + "step": 2195 + }, + { + "epoch": 2.4065753424657537, + "grad_norm": 2.893143892288208, + "learning_rate": 1.1311941510966695e-06, + "loss": 1.5554, + "step": 2196 + }, + { + "epoch": 2.4076712328767123, + "grad_norm": 2.86181640625, + "learning_rate": 1.1291632818846467e-06, + "loss": 1.5409, + "step": 2197 + }, + { + "epoch": 2.4087671232876713, + "grad_norm": 2.749263286590576, + "learning_rate": 1.127132412672624e-06, + "loss": 1.544, + "step": 2198 + }, + { + "epoch": 2.4098630136986303, + "grad_norm": 2.759432554244995, + "learning_rate": 1.1251015434606013e-06, + "loss": 1.5787, + "step": 2199 + }, + { + "epoch": 2.410958904109589, + "grad_norm": 2.8302371501922607, + "learning_rate": 1.1230706742485786e-06, + "loss": 1.6204, + "step": 2200 + }, + { + "epoch": 2.410958904109589, + "eval_loss": 1.8242546319961548, + "eval_runtime": 45.7883, + "eval_samples_per_second": 21.84, + "eval_steps_per_second": 0.917, + "step": 2200 + }, + { + "epoch": 2.412054794520548, + "grad_norm": 2.8081696033477783, + "learning_rate": 1.1210398050365558e-06, + "loss": 1.5497, + "step": 2201 + }, + { + "epoch": 2.413150684931507, + "grad_norm": 2.777566432952881, + "learning_rate": 1.1190089358245331e-06, + "loss": 1.5249, + "step": 2202 + }, + { + "epoch": 2.4142465753424656, + "grad_norm": 2.851879835128784, + "learning_rate": 1.1169780666125102e-06, + "loss": 1.5885, + "step": 2203 + }, + { + "epoch": 2.4153424657534246, + "grad_norm": 2.859072685241699, + "learning_rate": 1.1149471974004875e-06, + "loss": 1.5438, + "step": 2204 + }, + { + "epoch": 2.4164383561643836, + "grad_norm": 2.908538818359375, + "learning_rate": 1.1129163281884647e-06, + "loss": 1.6276, + "step": 2205 + }, + { + "epoch": 2.417534246575342, + "grad_norm": 2.7648227214813232, + "learning_rate": 1.110885458976442e-06, + "loss": 1.5367, + "step": 2206 + }, + { + "epoch": 2.4186301369863012, + "grad_norm": 2.9282727241516113, + "learning_rate": 1.1088545897644193e-06, + "loss": 1.6055, + "step": 2207 + }, + { + "epoch": 2.4197260273972603, + "grad_norm": 2.7551443576812744, + "learning_rate": 1.1068237205523966e-06, + "loss": 1.5984, + "step": 2208 + }, + { + "epoch": 2.4208219178082193, + "grad_norm": 2.88720440864563, + "learning_rate": 1.1047928513403736e-06, + "loss": 1.5988, + "step": 2209 + }, + { + "epoch": 2.421917808219178, + "grad_norm": 2.75961971282959, + "learning_rate": 1.102761982128351e-06, + "loss": 1.5291, + "step": 2210 + }, + { + "epoch": 2.421917808219178, + "eval_loss": 1.8230210542678833, + "eval_runtime": 45.9261, + "eval_samples_per_second": 21.774, + "eval_steps_per_second": 0.915, + "step": 2210 + }, + { + "epoch": 2.423013698630137, + "grad_norm": 2.7809903621673584, + "learning_rate": 1.1007311129163282e-06, + "loss": 1.5563, + "step": 2211 + }, + { + "epoch": 2.424109589041096, + "grad_norm": 2.7741317749023438, + "learning_rate": 1.0987002437043055e-06, + "loss": 1.5594, + "step": 2212 + }, + { + "epoch": 2.425205479452055, + "grad_norm": 2.840890884399414, + "learning_rate": 1.0966693744922827e-06, + "loss": 1.52, + "step": 2213 + }, + { + "epoch": 2.4263013698630136, + "grad_norm": 2.9647538661956787, + "learning_rate": 1.09463850528026e-06, + "loss": 1.5447, + "step": 2214 + }, + { + "epoch": 2.4273972602739726, + "grad_norm": 2.9884531497955322, + "learning_rate": 1.0926076360682373e-06, + "loss": 1.5426, + "step": 2215 + }, + { + "epoch": 2.4284931506849317, + "grad_norm": 2.8884270191192627, + "learning_rate": 1.0905767668562146e-06, + "loss": 1.5546, + "step": 2216 + }, + { + "epoch": 2.4295890410958902, + "grad_norm": 2.97591495513916, + "learning_rate": 1.0885458976441918e-06, + "loss": 1.632, + "step": 2217 + }, + { + "epoch": 2.4306849315068493, + "grad_norm": 2.944126605987549, + "learning_rate": 1.0865150284321691e-06, + "loss": 1.5845, + "step": 2218 + }, + { + "epoch": 2.4317808219178083, + "grad_norm": 2.9660258293151855, + "learning_rate": 1.0844841592201464e-06, + "loss": 1.5939, + "step": 2219 + }, + { + "epoch": 2.432876712328767, + "grad_norm": 2.8848304748535156, + "learning_rate": 1.0824532900081237e-06, + "loss": 1.5848, + "step": 2220 + }, + { + "epoch": 2.432876712328767, + "eval_loss": 1.8266901969909668, + "eval_runtime": 51.9703, + "eval_samples_per_second": 19.242, + "eval_steps_per_second": 0.808, + "step": 2220 + }, + { + "epoch": 2.433972602739726, + "grad_norm": 2.9656991958618164, + "learning_rate": 1.080422420796101e-06, + "loss": 1.568, + "step": 2221 + }, + { + "epoch": 2.435068493150685, + "grad_norm": 2.9461898803710938, + "learning_rate": 1.078391551584078e-06, + "loss": 1.5922, + "step": 2222 + }, + { + "epoch": 2.436164383561644, + "grad_norm": 2.937289237976074, + "learning_rate": 1.0763606823720553e-06, + "loss": 1.6414, + "step": 2223 + }, + { + "epoch": 2.4372602739726026, + "grad_norm": 2.8320538997650146, + "learning_rate": 1.0743298131600326e-06, + "loss": 1.5995, + "step": 2224 + }, + { + "epoch": 2.4383561643835616, + "grad_norm": 2.8221585750579834, + "learning_rate": 1.0722989439480098e-06, + "loss": 1.549, + "step": 2225 + }, + { + "epoch": 2.4394520547945207, + "grad_norm": 2.854454517364502, + "learning_rate": 1.070268074735987e-06, + "loss": 1.5474, + "step": 2226 + }, + { + "epoch": 2.4405479452054797, + "grad_norm": 2.7813878059387207, + "learning_rate": 1.0682372055239644e-06, + "loss": 1.5638, + "step": 2227 + }, + { + "epoch": 2.4416438356164383, + "grad_norm": 2.985661029815674, + "learning_rate": 1.0662063363119414e-06, + "loss": 1.5457, + "step": 2228 + }, + { + "epoch": 2.4427397260273973, + "grad_norm": 2.863631010055542, + "learning_rate": 1.0641754670999187e-06, + "loss": 1.5824, + "step": 2229 + }, + { + "epoch": 2.4438356164383563, + "grad_norm": 2.8328969478607178, + "learning_rate": 1.062144597887896e-06, + "loss": 1.5708, + "step": 2230 + }, + { + "epoch": 2.4438356164383563, + "eval_loss": 1.8261159658432007, + "eval_runtime": 50.0095, + "eval_samples_per_second": 19.996, + "eval_steps_per_second": 0.84, + "step": 2230 + }, + { + "epoch": 2.444931506849315, + "grad_norm": 2.9791696071624756, + "learning_rate": 1.0601137286758733e-06, + "loss": 1.6242, + "step": 2231 + }, + { + "epoch": 2.446027397260274, + "grad_norm": 2.7748308181762695, + "learning_rate": 1.0580828594638505e-06, + "loss": 1.499, + "step": 2232 + }, + { + "epoch": 2.447123287671233, + "grad_norm": 2.8511860370635986, + "learning_rate": 1.0560519902518278e-06, + "loss": 1.5787, + "step": 2233 + }, + { + "epoch": 2.4482191780821916, + "grad_norm": 2.802671194076538, + "learning_rate": 1.054021121039805e-06, + "loss": 1.4995, + "step": 2234 + }, + { + "epoch": 2.4493150684931506, + "grad_norm": 2.8005664348602295, + "learning_rate": 1.0519902518277824e-06, + "loss": 1.5083, + "step": 2235 + }, + { + "epoch": 2.4504109589041096, + "grad_norm": 2.92136287689209, + "learning_rate": 1.0499593826157596e-06, + "loss": 1.5807, + "step": 2236 + }, + { + "epoch": 2.4515068493150687, + "grad_norm": 2.850313186645508, + "learning_rate": 1.047928513403737e-06, + "loss": 1.5592, + "step": 2237 + }, + { + "epoch": 2.4526027397260273, + "grad_norm": 2.745370864868164, + "learning_rate": 1.0458976441917142e-06, + "loss": 1.495, + "step": 2238 + }, + { + "epoch": 2.4536986301369863, + "grad_norm": 2.794156789779663, + "learning_rate": 1.0438667749796915e-06, + "loss": 1.5024, + "step": 2239 + }, + { + "epoch": 2.4547945205479453, + "grad_norm": 2.8183324337005615, + "learning_rate": 1.0418359057676688e-06, + "loss": 1.5461, + "step": 2240 + }, + { + "epoch": 2.4547945205479453, + "eval_loss": 1.8265395164489746, + "eval_runtime": 47.1962, + "eval_samples_per_second": 21.188, + "eval_steps_per_second": 0.89, + "step": 2240 + }, + { + "epoch": 2.455890410958904, + "grad_norm": 2.863630771636963, + "learning_rate": 1.0398050365556458e-06, + "loss": 1.5977, + "step": 2241 + }, + { + "epoch": 2.456986301369863, + "grad_norm": 2.7688169479370117, + "learning_rate": 1.037774167343623e-06, + "loss": 1.5399, + "step": 2242 + }, + { + "epoch": 2.458082191780822, + "grad_norm": 2.8812880516052246, + "learning_rate": 1.0357432981316004e-06, + "loss": 1.5722, + "step": 2243 + }, + { + "epoch": 2.459178082191781, + "grad_norm": 3.2114691734313965, + "learning_rate": 1.0337124289195776e-06, + "loss": 1.6012, + "step": 2244 + }, + { + "epoch": 2.4602739726027396, + "grad_norm": 2.961271047592163, + "learning_rate": 1.031681559707555e-06, + "loss": 1.606, + "step": 2245 + }, + { + "epoch": 2.4613698630136986, + "grad_norm": 2.835495710372925, + "learning_rate": 1.0296506904955322e-06, + "loss": 1.5518, + "step": 2246 + }, + { + "epoch": 2.4624657534246577, + "grad_norm": 2.8796403408050537, + "learning_rate": 1.0276198212835095e-06, + "loss": 1.573, + "step": 2247 + }, + { + "epoch": 2.4635616438356163, + "grad_norm": 2.774409770965576, + "learning_rate": 1.0255889520714867e-06, + "loss": 1.5485, + "step": 2248 + }, + { + "epoch": 2.4646575342465753, + "grad_norm": 2.8847763538360596, + "learning_rate": 1.023558082859464e-06, + "loss": 1.5168, + "step": 2249 + }, + { + "epoch": 2.4657534246575343, + "grad_norm": 2.9533803462982178, + "learning_rate": 1.0215272136474413e-06, + "loss": 1.6179, + "step": 2250 + }, + { + "epoch": 2.4657534246575343, + "eval_loss": 1.8261420726776123, + "eval_runtime": 46.1356, + "eval_samples_per_second": 21.675, + "eval_steps_per_second": 0.91, + "step": 2250 + }, + { + "epoch": 2.466849315068493, + "grad_norm": 2.8762729167938232, + "learning_rate": 1.0194963444354186e-06, + "loss": 1.5109, + "step": 2251 + }, + { + "epoch": 2.467945205479452, + "grad_norm": 2.9228789806365967, + "learning_rate": 1.0174654752233958e-06, + "loss": 1.5861, + "step": 2252 + }, + { + "epoch": 2.469041095890411, + "grad_norm": 2.9467368125915527, + "learning_rate": 1.015434606011373e-06, + "loss": 1.5533, + "step": 2253 + }, + { + "epoch": 2.47013698630137, + "grad_norm": 2.9693222045898438, + "learning_rate": 1.0134037367993502e-06, + "loss": 1.5741, + "step": 2254 + }, + { + "epoch": 2.4712328767123286, + "grad_norm": 2.933368682861328, + "learning_rate": 1.0113728675873275e-06, + "loss": 1.5555, + "step": 2255 + }, + { + "epoch": 2.4723287671232876, + "grad_norm": 2.9675047397613525, + "learning_rate": 1.0093419983753047e-06, + "loss": 1.5803, + "step": 2256 + }, + { + "epoch": 2.4734246575342467, + "grad_norm": 2.931260108947754, + "learning_rate": 1.007311129163282e-06, + "loss": 1.5263, + "step": 2257 + }, + { + "epoch": 2.4745205479452057, + "grad_norm": 2.8160653114318848, + "learning_rate": 1.0052802599512593e-06, + "loss": 1.5281, + "step": 2258 + }, + { + "epoch": 2.4756164383561643, + "grad_norm": 3.0168352127075195, + "learning_rate": 1.0032493907392364e-06, + "loss": 1.5746, + "step": 2259 + }, + { + "epoch": 2.4767123287671233, + "grad_norm": 2.7623541355133057, + "learning_rate": 1.0012185215272136e-06, + "loss": 1.5181, + "step": 2260 + }, + { + "epoch": 2.4767123287671233, + "eval_loss": 1.8249253034591675, + "eval_runtime": 47.2027, + "eval_samples_per_second": 21.185, + "eval_steps_per_second": 0.89, + "step": 2260 + }, + { + "epoch": 2.4778082191780824, + "grad_norm": 2.8098902702331543, + "learning_rate": 9.99187652315191e-07, + "loss": 1.6099, + "step": 2261 + }, + { + "epoch": 2.478904109589041, + "grad_norm": 2.8562352657318115, + "learning_rate": 9.971567831031682e-07, + "loss": 1.5795, + "step": 2262 + }, + { + "epoch": 2.48, + "grad_norm": 2.7867422103881836, + "learning_rate": 9.951259138911455e-07, + "loss": 1.5104, + "step": 2263 + }, + { + "epoch": 2.481095890410959, + "grad_norm": 2.9658195972442627, + "learning_rate": 9.930950446791227e-07, + "loss": 1.5891, + "step": 2264 + }, + { + "epoch": 2.4821917808219176, + "grad_norm": 2.824355363845825, + "learning_rate": 9.910641754671e-07, + "loss": 1.6385, + "step": 2265 + }, + { + "epoch": 2.4832876712328766, + "grad_norm": 2.833772897720337, + "learning_rate": 9.890333062550773e-07, + "loss": 1.5759, + "step": 2266 + }, + { + "epoch": 2.4843835616438357, + "grad_norm": 2.9420840740203857, + "learning_rate": 9.870024370430546e-07, + "loss": 1.573, + "step": 2267 + }, + { + "epoch": 2.4854794520547947, + "grad_norm": 2.84629225730896, + "learning_rate": 9.849715678310318e-07, + "loss": 1.5606, + "step": 2268 + }, + { + "epoch": 2.4865753424657533, + "grad_norm": 2.7983481884002686, + "learning_rate": 9.829406986190091e-07, + "loss": 1.5718, + "step": 2269 + }, + { + "epoch": 2.4876712328767123, + "grad_norm": 2.8418283462524414, + "learning_rate": 9.809098294069864e-07, + "loss": 1.5052, + "step": 2270 + }, + { + "epoch": 2.4876712328767123, + "eval_loss": 1.8237050771713257, + "eval_runtime": 53.2157, + "eval_samples_per_second": 18.791, + "eval_steps_per_second": 0.789, + "step": 2270 + }, + { + "epoch": 2.4887671232876714, + "grad_norm": 2.8386685848236084, + "learning_rate": 9.788789601949637e-07, + "loss": 1.5638, + "step": 2271 + }, + { + "epoch": 2.48986301369863, + "grad_norm": 2.8796494007110596, + "learning_rate": 9.768480909829407e-07, + "loss": 1.6258, + "step": 2272 + }, + { + "epoch": 2.490958904109589, + "grad_norm": 2.8450164794921875, + "learning_rate": 9.74817221770918e-07, + "loss": 1.5567, + "step": 2273 + }, + { + "epoch": 2.492054794520548, + "grad_norm": 2.8501436710357666, + "learning_rate": 9.727863525588953e-07, + "loss": 1.5731, + "step": 2274 + }, + { + "epoch": 2.493150684931507, + "grad_norm": 2.8499205112457275, + "learning_rate": 9.707554833468726e-07, + "loss": 1.5383, + "step": 2275 + }, + { + "epoch": 2.4942465753424656, + "grad_norm": 2.904446601867676, + "learning_rate": 9.687246141348498e-07, + "loss": 1.5423, + "step": 2276 + }, + { + "epoch": 2.4953424657534247, + "grad_norm": 2.7885360717773438, + "learning_rate": 9.666937449228271e-07, + "loss": 1.5583, + "step": 2277 + }, + { + "epoch": 2.4964383561643837, + "grad_norm": 2.8230857849121094, + "learning_rate": 9.646628757108042e-07, + "loss": 1.531, + "step": 2278 + }, + { + "epoch": 2.4975342465753423, + "grad_norm": 2.9093189239501953, + "learning_rate": 9.626320064987814e-07, + "loss": 1.5901, + "step": 2279 + }, + { + "epoch": 2.4986301369863013, + "grad_norm": 2.853430986404419, + "learning_rate": 9.606011372867587e-07, + "loss": 1.5425, + "step": 2280 + }, + { + "epoch": 2.4986301369863013, + "eval_loss": 1.824331283569336, + "eval_runtime": 51.6196, + "eval_samples_per_second": 19.372, + "eval_steps_per_second": 0.814, + "step": 2280 + }, + { + "epoch": 2.4997260273972604, + "grad_norm": 2.857558488845825, + "learning_rate": 9.58570268074736e-07, + "loss": 1.555, + "step": 2281 + }, + { + "epoch": 2.500821917808219, + "grad_norm": 2.758384943008423, + "learning_rate": 9.565393988627133e-07, + "loss": 1.5546, + "step": 2282 + }, + { + "epoch": 2.501917808219178, + "grad_norm": 2.8295326232910156, + "learning_rate": 9.545085296506906e-07, + "loss": 1.5842, + "step": 2283 + }, + { + "epoch": 2.503013698630137, + "grad_norm": 2.855534315109253, + "learning_rate": 9.524776604386677e-07, + "loss": 1.5636, + "step": 2284 + }, + { + "epoch": 2.504109589041096, + "grad_norm": 2.7737905979156494, + "learning_rate": 9.50446791226645e-07, + "loss": 1.5101, + "step": 2285 + }, + { + "epoch": 2.5052054794520546, + "grad_norm": 2.7556989192962646, + "learning_rate": 9.484159220146223e-07, + "loss": 1.6216, + "step": 2286 + }, + { + "epoch": 2.5063013698630137, + "grad_norm": 2.8598365783691406, + "learning_rate": 9.463850528025995e-07, + "loss": 1.5818, + "step": 2287 + }, + { + "epoch": 2.5073972602739727, + "grad_norm": 3.0465996265411377, + "learning_rate": 9.443541835905768e-07, + "loss": 1.6237, + "step": 2288 + }, + { + "epoch": 2.5084931506849317, + "grad_norm": 3.067974328994751, + "learning_rate": 9.423233143785541e-07, + "loss": 1.6346, + "step": 2289 + }, + { + "epoch": 2.5095890410958903, + "grad_norm": 2.9208505153656006, + "learning_rate": 9.402924451665313e-07, + "loss": 1.543, + "step": 2290 + }, + { + "epoch": 2.5095890410958903, + "eval_loss": 1.8239408731460571, + "eval_runtime": 52.1634, + "eval_samples_per_second": 19.171, + "eval_steps_per_second": 0.805, + "step": 2290 + }, + { + "epoch": 2.5106849315068493, + "grad_norm": 3.0587332248687744, + "learning_rate": 9.382615759545085e-07, + "loss": 1.5814, + "step": 2291 + }, + { + "epoch": 2.5117808219178084, + "grad_norm": 2.9887471199035645, + "learning_rate": 9.362307067424858e-07, + "loss": 1.5481, + "step": 2292 + }, + { + "epoch": 2.512876712328767, + "grad_norm": 3.127073049545288, + "learning_rate": 9.341998375304631e-07, + "loss": 1.6121, + "step": 2293 + }, + { + "epoch": 2.513972602739726, + "grad_norm": 2.7613704204559326, + "learning_rate": 9.321689683184404e-07, + "loss": 1.5337, + "step": 2294 + }, + { + "epoch": 2.515068493150685, + "grad_norm": 2.9512479305267334, + "learning_rate": 9.301380991064176e-07, + "loss": 1.5161, + "step": 2295 + }, + { + "epoch": 2.5161643835616436, + "grad_norm": 2.9375245571136475, + "learning_rate": 9.281072298943949e-07, + "loss": 1.5538, + "step": 2296 + }, + { + "epoch": 2.5172602739726027, + "grad_norm": 3.8778791427612305, + "learning_rate": 9.260763606823721e-07, + "loss": 1.5736, + "step": 2297 + }, + { + "epoch": 2.5183561643835617, + "grad_norm": 3.176246404647827, + "learning_rate": 9.240454914703494e-07, + "loss": 1.5855, + "step": 2298 + }, + { + "epoch": 2.5194520547945203, + "grad_norm": 2.826871395111084, + "learning_rate": 9.220146222583266e-07, + "loss": 1.526, + "step": 2299 + }, + { + "epoch": 2.5205479452054793, + "grad_norm": 2.7983620166778564, + "learning_rate": 9.199837530463039e-07, + "loss": 1.5656, + "step": 2300 + }, + { + "epoch": 2.5205479452054793, + "eval_loss": 1.8233634233474731, + "eval_runtime": 51.1613, + "eval_samples_per_second": 19.546, + "eval_steps_per_second": 0.821, + "step": 2300 + }, + { + "epoch": 2.5216438356164383, + "grad_norm": 2.8449342250823975, + "learning_rate": 9.179528838342812e-07, + "loss": 1.5574, + "step": 2301 + }, + { + "epoch": 2.5227397260273974, + "grad_norm": 2.8809049129486084, + "learning_rate": 9.159220146222585e-07, + "loss": 1.5828, + "step": 2302 + }, + { + "epoch": 2.5238356164383564, + "grad_norm": 2.7337660789489746, + "learning_rate": 9.138911454102356e-07, + "loss": 1.5006, + "step": 2303 + }, + { + "epoch": 2.524931506849315, + "grad_norm": 2.7337660789489746, + "learning_rate": 9.138911454102356e-07, + "loss": 1.5592, + "step": 2304 + }, + { + "epoch": 2.526027397260274, + "grad_norm": 3.1104655265808105, + "learning_rate": 9.118602761982129e-07, + "loss": 1.5642, + "step": 2305 + }, + { + "epoch": 2.527123287671233, + "grad_norm": 4.62161922454834, + "learning_rate": 9.098294069861902e-07, + "loss": 1.5333, + "step": 2306 + }, + { + "epoch": 2.5282191780821917, + "grad_norm": 2.879892110824585, + "learning_rate": 9.077985377741675e-07, + "loss": 1.504, + "step": 2307 + }, + { + "epoch": 2.5293150684931507, + "grad_norm": 2.9683990478515625, + "learning_rate": 9.057676685621447e-07, + "loss": 1.6184, + "step": 2308 + }, + { + "epoch": 2.5304109589041097, + "grad_norm": 2.919980049133301, + "learning_rate": 9.03736799350122e-07, + "loss": 1.5908, + "step": 2309 + }, + { + "epoch": 2.5315068493150683, + "grad_norm": 3.0819506645202637, + "learning_rate": 9.017059301380991e-07, + "loss": 1.6252, + "step": 2310 + }, + { + "epoch": 2.5315068493150683, + "eval_loss": 1.8249709606170654, + "eval_runtime": 44.4295, + "eval_samples_per_second": 22.508, + "eval_steps_per_second": 0.945, + "step": 2310 + }, + { + "epoch": 2.5326027397260273, + "grad_norm": 3.0078251361846924, + "learning_rate": 8.996750609260764e-07, + "loss": 1.5653, + "step": 2311 + }, + { + "epoch": 2.5336986301369864, + "grad_norm": 3.0950827598571777, + "learning_rate": 8.976441917140536e-07, + "loss": 1.6341, + "step": 2312 + }, + { + "epoch": 2.534794520547945, + "grad_norm": 2.9090089797973633, + "learning_rate": 8.956133225020309e-07, + "loss": 1.5441, + "step": 2313 + }, + { + "epoch": 2.535890410958904, + "grad_norm": 2.8889381885528564, + "learning_rate": 8.935824532900082e-07, + "loss": 1.5371, + "step": 2314 + }, + { + "epoch": 2.536986301369863, + "grad_norm": 3.3039140701293945, + "learning_rate": 8.915515840779855e-07, + "loss": 1.5936, + "step": 2315 + }, + { + "epoch": 2.538082191780822, + "grad_norm": 2.9188146591186523, + "learning_rate": 8.895207148659626e-07, + "loss": 1.5787, + "step": 2316 + }, + { + "epoch": 2.539178082191781, + "grad_norm": 2.8546876907348633, + "learning_rate": 8.874898456539399e-07, + "loss": 1.58, + "step": 2317 + }, + { + "epoch": 2.5402739726027397, + "grad_norm": 2.8007283210754395, + "learning_rate": 8.854589764419172e-07, + "loss": 1.518, + "step": 2318 + }, + { + "epoch": 2.5413698630136987, + "grad_norm": 2.8198978900909424, + "learning_rate": 8.834281072298945e-07, + "loss": 1.5695, + "step": 2319 + }, + { + "epoch": 2.5424657534246577, + "grad_norm": 3.3483028411865234, + "learning_rate": 8.813972380178717e-07, + "loss": 1.5814, + "step": 2320 + }, + { + "epoch": 2.5424657534246577, + "eval_loss": 1.8233474493026733, + "eval_runtime": 49.2421, + "eval_samples_per_second": 20.308, + "eval_steps_per_second": 0.853, + "step": 2320 + }, + { + "epoch": 2.5435616438356163, + "grad_norm": 2.838481903076172, + "learning_rate": 8.79366368805849e-07, + "loss": 1.5311, + "step": 2321 + }, + { + "epoch": 2.5446575342465754, + "grad_norm": 2.8281280994415283, + "learning_rate": 8.773354995938262e-07, + "loss": 1.5235, + "step": 2322 + }, + { + "epoch": 2.5457534246575344, + "grad_norm": 2.8211021423339844, + "learning_rate": 8.753046303818035e-07, + "loss": 1.5484, + "step": 2323 + }, + { + "epoch": 2.546849315068493, + "grad_norm": 2.865307331085205, + "learning_rate": 8.732737611697807e-07, + "loss": 1.5892, + "step": 2324 + }, + { + "epoch": 2.547945205479452, + "grad_norm": 2.87202787399292, + "learning_rate": 8.71242891957758e-07, + "loss": 1.5961, + "step": 2325 + }, + { + "epoch": 2.549041095890411, + "grad_norm": 2.9286715984344482, + "learning_rate": 8.692120227457353e-07, + "loss": 1.5597, + "step": 2326 + }, + { + "epoch": 2.5501369863013696, + "grad_norm": 2.965538740158081, + "learning_rate": 8.671811535337126e-07, + "loss": 1.6278, + "step": 2327 + }, + { + "epoch": 2.5512328767123287, + "grad_norm": 2.991199254989624, + "learning_rate": 8.651502843216898e-07, + "loss": 1.5904, + "step": 2328 + }, + { + "epoch": 2.5523287671232877, + "grad_norm": 2.902418613433838, + "learning_rate": 8.63119415109667e-07, + "loss": 1.584, + "step": 2329 + }, + { + "epoch": 2.5534246575342463, + "grad_norm": 2.8904943466186523, + "learning_rate": 8.610885458976443e-07, + "loss": 1.5178, + "step": 2330 + }, + { + "epoch": 2.5534246575342463, + "eval_loss": 1.8210104703903198, + "eval_runtime": 49.456, + "eval_samples_per_second": 20.22, + "eval_steps_per_second": 0.849, + "step": 2330 + }, + { + "epoch": 2.5545205479452053, + "grad_norm": 2.778411865234375, + "learning_rate": 8.590576766856216e-07, + "loss": 1.5318, + "step": 2331 + }, + { + "epoch": 2.5556164383561644, + "grad_norm": 2.8281290531158447, + "learning_rate": 8.570268074735988e-07, + "loss": 1.5399, + "step": 2332 + }, + { + "epoch": 2.5567123287671234, + "grad_norm": 2.8783960342407227, + "learning_rate": 8.549959382615761e-07, + "loss": 1.5653, + "step": 2333 + }, + { + "epoch": 2.5578082191780824, + "grad_norm": 2.7188220024108887, + "learning_rate": 8.529650690495534e-07, + "loss": 1.5113, + "step": 2334 + }, + { + "epoch": 2.558904109589041, + "grad_norm": 2.705050468444824, + "learning_rate": 8.509341998375305e-07, + "loss": 1.4955, + "step": 2335 + }, + { + "epoch": 2.56, + "grad_norm": 2.798614740371704, + "learning_rate": 8.489033306255077e-07, + "loss": 1.529, + "step": 2336 + }, + { + "epoch": 2.561095890410959, + "grad_norm": 2.795304298400879, + "learning_rate": 8.46872461413485e-07, + "loss": 1.5431, + "step": 2337 + }, + { + "epoch": 2.5621917808219177, + "grad_norm": 2.838188648223877, + "learning_rate": 8.448415922014623e-07, + "loss": 1.5772, + "step": 2338 + }, + { + "epoch": 2.5632876712328767, + "grad_norm": 2.9334065914154053, + "learning_rate": 8.428107229894396e-07, + "loss": 1.5621, + "step": 2339 + }, + { + "epoch": 2.5643835616438357, + "grad_norm": 2.9324002265930176, + "learning_rate": 8.407798537774168e-07, + "loss": 1.5422, + "step": 2340 + }, + { + "epoch": 2.5643835616438357, + "eval_loss": 1.8223708868026733, + "eval_runtime": 50.7517, + "eval_samples_per_second": 19.704, + "eval_steps_per_second": 0.828, + "step": 2340 + }, + { + "epoch": 2.5654794520547943, + "grad_norm": 2.7721550464630127, + "learning_rate": 8.38748984565394e-07, + "loss": 1.5507, + "step": 2341 + }, + { + "epoch": 2.5665753424657534, + "grad_norm": 2.918529510498047, + "learning_rate": 8.367181153533713e-07, + "loss": 1.5703, + "step": 2342 + }, + { + "epoch": 2.5676712328767124, + "grad_norm": 2.926311492919922, + "learning_rate": 8.346872461413486e-07, + "loss": 1.5611, + "step": 2343 + }, + { + "epoch": 2.568767123287671, + "grad_norm": 2.9318113327026367, + "learning_rate": 8.326563769293258e-07, + "loss": 1.6348, + "step": 2344 + }, + { + "epoch": 2.56986301369863, + "grad_norm": 2.8003222942352295, + "learning_rate": 8.306255077173031e-07, + "loss": 1.5064, + "step": 2345 + }, + { + "epoch": 2.570958904109589, + "grad_norm": 2.797013998031616, + "learning_rate": 8.285946385052804e-07, + "loss": 1.5907, + "step": 2346 + }, + { + "epoch": 2.572054794520548, + "grad_norm": 2.7971713542938232, + "learning_rate": 8.265637692932575e-07, + "loss": 1.5844, + "step": 2347 + }, + { + "epoch": 2.573150684931507, + "grad_norm": 2.8959572315216064, + "learning_rate": 8.245329000812348e-07, + "loss": 1.5173, + "step": 2348 + }, + { + "epoch": 2.5742465753424657, + "grad_norm": 2.8146274089813232, + "learning_rate": 8.225020308692121e-07, + "loss": 1.5615, + "step": 2349 + }, + { + "epoch": 2.5753424657534247, + "grad_norm": 2.817338705062866, + "learning_rate": 8.204711616571894e-07, + "loss": 1.5076, + "step": 2350 + }, + { + "epoch": 2.5753424657534247, + "eval_loss": 1.8196357488632202, + "eval_runtime": 48.5561, + "eval_samples_per_second": 20.595, + "eval_steps_per_second": 0.865, + "step": 2350 + }, + { + "epoch": 2.5764383561643838, + "grad_norm": 2.9276442527770996, + "learning_rate": 8.184402924451667e-07, + "loss": 1.5776, + "step": 2351 + }, + { + "epoch": 2.5775342465753424, + "grad_norm": 2.860095739364624, + "learning_rate": 8.164094232331439e-07, + "loss": 1.5677, + "step": 2352 + }, + { + "epoch": 2.5786301369863014, + "grad_norm": 2.917368173599243, + "learning_rate": 8.14378554021121e-07, + "loss": 1.5489, + "step": 2353 + }, + { + "epoch": 2.5797260273972604, + "grad_norm": 2.8872427940368652, + "learning_rate": 8.123476848090983e-07, + "loss": 1.5458, + "step": 2354 + }, + { + "epoch": 2.580821917808219, + "grad_norm": 2.8507397174835205, + "learning_rate": 8.103168155970755e-07, + "loss": 1.5662, + "step": 2355 + }, + { + "epoch": 2.581917808219178, + "grad_norm": 2.8826792240142822, + "learning_rate": 8.082859463850528e-07, + "loss": 1.5574, + "step": 2356 + }, + { + "epoch": 2.583013698630137, + "grad_norm": 2.8728246688842773, + "learning_rate": 8.062550771730301e-07, + "loss": 1.5516, + "step": 2357 + }, + { + "epoch": 2.5841095890410957, + "grad_norm": 2.9101784229278564, + "learning_rate": 8.042242079610074e-07, + "loss": 1.5472, + "step": 2358 + }, + { + "epoch": 2.5852054794520547, + "grad_norm": 2.791224718093872, + "learning_rate": 8.021933387489846e-07, + "loss": 1.5663, + "step": 2359 + }, + { + "epoch": 2.5863013698630137, + "grad_norm": 2.983250856399536, + "learning_rate": 8.001624695369618e-07, + "loss": 1.5908, + "step": 2360 + }, + { + "epoch": 2.5863013698630137, + "eval_loss": 1.8226059675216675, + "eval_runtime": 48.2046, + "eval_samples_per_second": 20.745, + "eval_steps_per_second": 0.871, + "step": 2360 + }, + { + "epoch": 2.5873972602739728, + "grad_norm": 2.920762538909912, + "learning_rate": 7.981316003249391e-07, + "loss": 1.5111, + "step": 2361 + }, + { + "epoch": 2.5884931506849314, + "grad_norm": 2.983764171600342, + "learning_rate": 7.961007311129164e-07, + "loss": 1.566, + "step": 2362 + }, + { + "epoch": 2.5895890410958904, + "grad_norm": 2.976851463317871, + "learning_rate": 7.940698619008936e-07, + "loss": 1.6571, + "step": 2363 + }, + { + "epoch": 2.5906849315068494, + "grad_norm": 2.856072187423706, + "learning_rate": 7.920389926888709e-07, + "loss": 1.5176, + "step": 2364 + }, + { + "epoch": 2.5917808219178085, + "grad_norm": 2.8299734592437744, + "learning_rate": 7.900081234768482e-07, + "loss": 1.587, + "step": 2365 + }, + { + "epoch": 2.592876712328767, + "grad_norm": 2.937560558319092, + "learning_rate": 7.879772542648254e-07, + "loss": 1.5769, + "step": 2366 + }, + { + "epoch": 2.593972602739726, + "grad_norm": 2.8437840938568115, + "learning_rate": 7.859463850528026e-07, + "loss": 1.5665, + "step": 2367 + }, + { + "epoch": 2.595068493150685, + "grad_norm": 3.0254647731781006, + "learning_rate": 7.839155158407799e-07, + "loss": 1.5783, + "step": 2368 + }, + { + "epoch": 2.5961643835616437, + "grad_norm": 2.895967960357666, + "learning_rate": 7.818846466287572e-07, + "loss": 1.5783, + "step": 2369 + }, + { + "epoch": 2.5972602739726027, + "grad_norm": 2.8868582248687744, + "learning_rate": 7.798537774167345e-07, + "loss": 1.527, + "step": 2370 + }, + { + "epoch": 2.5972602739726027, + "eval_loss": 1.8213152885437012, + "eval_runtime": 51.0508, + "eval_samples_per_second": 19.588, + "eval_steps_per_second": 0.823, + "step": 2370 + }, + { + "epoch": 2.5983561643835618, + "grad_norm": 2.906033992767334, + "learning_rate": 7.778229082047117e-07, + "loss": 1.5688, + "step": 2371 + }, + { + "epoch": 2.5994520547945204, + "grad_norm": 2.846966505050659, + "learning_rate": 7.757920389926889e-07, + "loss": 1.5184, + "step": 2372 + }, + { + "epoch": 2.6005479452054794, + "grad_norm": 3.124784231185913, + "learning_rate": 7.737611697806662e-07, + "loss": 1.5953, + "step": 2373 + }, + { + "epoch": 2.6016438356164384, + "grad_norm": 3.014098644256592, + "learning_rate": 7.717303005686435e-07, + "loss": 1.6318, + "step": 2374 + }, + { + "epoch": 2.602739726027397, + "grad_norm": 2.82450532913208, + "learning_rate": 7.696994313566207e-07, + "loss": 1.5106, + "step": 2375 + }, + { + "epoch": 2.603835616438356, + "grad_norm": 2.8513712882995605, + "learning_rate": 7.67668562144598e-07, + "loss": 1.5053, + "step": 2376 + }, + { + "epoch": 2.604931506849315, + "grad_norm": 2.99210786819458, + "learning_rate": 7.656376929325753e-07, + "loss": 1.5844, + "step": 2377 + }, + { + "epoch": 2.606027397260274, + "grad_norm": 2.858902931213379, + "learning_rate": 7.636068237205524e-07, + "loss": 1.5156, + "step": 2378 + }, + { + "epoch": 2.607123287671233, + "grad_norm": 2.9171338081359863, + "learning_rate": 7.615759545085296e-07, + "loss": 1.5474, + "step": 2379 + }, + { + "epoch": 2.6082191780821917, + "grad_norm": 2.868882417678833, + "learning_rate": 7.595450852965069e-07, + "loss": 1.5826, + "step": 2380 + }, + { + "epoch": 2.6082191780821917, + "eval_loss": 1.8230814933776855, + "eval_runtime": 46.2884, + "eval_samples_per_second": 21.604, + "eval_steps_per_second": 0.907, + "step": 2380 + }, + { + "epoch": 2.6093150684931508, + "grad_norm": 2.8466811180114746, + "learning_rate": 7.575142160844842e-07, + "loss": 1.5272, + "step": 2381 + }, + { + "epoch": 2.61041095890411, + "grad_norm": 2.8146281242370605, + "learning_rate": 7.554833468724615e-07, + "loss": 1.5303, + "step": 2382 + }, + { + "epoch": 2.6115068493150684, + "grad_norm": 2.8843190670013428, + "learning_rate": 7.534524776604387e-07, + "loss": 1.5767, + "step": 2383 + }, + { + "epoch": 2.6126027397260274, + "grad_norm": 2.988900899887085, + "learning_rate": 7.51421608448416e-07, + "loss": 1.5335, + "step": 2384 + }, + { + "epoch": 2.6136986301369864, + "grad_norm": 3.4128847122192383, + "learning_rate": 7.493907392363932e-07, + "loss": 1.485, + "step": 2385 + }, + { + "epoch": 2.614794520547945, + "grad_norm": 2.8596994876861572, + "learning_rate": 7.473598700243705e-07, + "loss": 1.5597, + "step": 2386 + }, + { + "epoch": 2.615890410958904, + "grad_norm": 2.923269748687744, + "learning_rate": 7.453290008123477e-07, + "loss": 1.5736, + "step": 2387 + }, + { + "epoch": 2.616986301369863, + "grad_norm": 2.8989574909210205, + "learning_rate": 7.43298131600325e-07, + "loss": 1.5458, + "step": 2388 + }, + { + "epoch": 2.6180821917808217, + "grad_norm": 2.935607671737671, + "learning_rate": 7.412672623883023e-07, + "loss": 1.4983, + "step": 2389 + }, + { + "epoch": 2.6191780821917807, + "grad_norm": 3.215050220489502, + "learning_rate": 7.392363931762796e-07, + "loss": 1.5571, + "step": 2390 + }, + { + "epoch": 2.6191780821917807, + "eval_loss": 1.8221906423568726, + "eval_runtime": 47.9911, + "eval_samples_per_second": 20.837, + "eval_steps_per_second": 0.875, + "step": 2390 + }, + { + "epoch": 2.6202739726027398, + "grad_norm": 3.0844898223876953, + "learning_rate": 7.372055239642567e-07, + "loss": 1.5767, + "step": 2391 + }, + { + "epoch": 2.621369863013699, + "grad_norm": 3.113297700881958, + "learning_rate": 7.35174654752234e-07, + "loss": 1.5895, + "step": 2392 + }, + { + "epoch": 2.6224657534246574, + "grad_norm": 2.904573917388916, + "learning_rate": 7.331437855402113e-07, + "loss": 1.5162, + "step": 2393 + }, + { + "epoch": 2.6235616438356164, + "grad_norm": 2.8379499912261963, + "learning_rate": 7.311129163281886e-07, + "loss": 1.5901, + "step": 2394 + }, + { + "epoch": 2.6246575342465754, + "grad_norm": 2.9121007919311523, + "learning_rate": 7.290820471161658e-07, + "loss": 1.5908, + "step": 2395 + }, + { + "epoch": 2.6257534246575345, + "grad_norm": 2.8151888847351074, + "learning_rate": 7.270511779041431e-07, + "loss": 1.5163, + "step": 2396 + }, + { + "epoch": 2.626849315068493, + "grad_norm": 2.9227914810180664, + "learning_rate": 7.250203086921203e-07, + "loss": 1.5466, + "step": 2397 + }, + { + "epoch": 2.627945205479452, + "grad_norm": 2.834669589996338, + "learning_rate": 7.229894394800976e-07, + "loss": 1.5724, + "step": 2398 + }, + { + "epoch": 2.629041095890411, + "grad_norm": 2.7271687984466553, + "learning_rate": 7.209585702680748e-07, + "loss": 1.5099, + "step": 2399 + }, + { + "epoch": 2.6301369863013697, + "grad_norm": 2.7215335369110107, + "learning_rate": 7.189277010560521e-07, + "loss": 1.5419, + "step": 2400 + }, + { + "epoch": 2.6301369863013697, + "eval_loss": 1.8201849460601807, + "eval_runtime": 50.7567, + "eval_samples_per_second": 19.702, + "eval_steps_per_second": 0.827, + "step": 2400 + }, + { + "epoch": 2.6312328767123287, + "grad_norm": 2.840486764907837, + "learning_rate": 7.168968318440294e-07, + "loss": 1.5203, + "step": 2401 + }, + { + "epoch": 2.632328767123288, + "grad_norm": 2.885180711746216, + "learning_rate": 7.148659626320067e-07, + "loss": 1.6037, + "step": 2402 + }, + { + "epoch": 2.6334246575342464, + "grad_norm": 2.9077260494232178, + "learning_rate": 7.128350934199837e-07, + "loss": 1.587, + "step": 2403 + }, + { + "epoch": 2.6345205479452054, + "grad_norm": 2.781118154525757, + "learning_rate": 7.10804224207961e-07, + "loss": 1.545, + "step": 2404 + }, + { + "epoch": 2.6356164383561644, + "grad_norm": 2.815735101699829, + "learning_rate": 7.087733549959383e-07, + "loss": 1.5327, + "step": 2405 + }, + { + "epoch": 2.636712328767123, + "grad_norm": 2.9639029502868652, + "learning_rate": 7.067424857839155e-07, + "loss": 1.5591, + "step": 2406 + }, + { + "epoch": 2.637808219178082, + "grad_norm": 2.9713387489318848, + "learning_rate": 7.047116165718928e-07, + "loss": 1.5416, + "step": 2407 + }, + { + "epoch": 2.638904109589041, + "grad_norm": 3.1526248455047607, + "learning_rate": 7.026807473598701e-07, + "loss": 1.607, + "step": 2408 + }, + { + "epoch": 2.64, + "grad_norm": 3.009585380554199, + "learning_rate": 7.006498781478473e-07, + "loss": 1.5223, + "step": 2409 + }, + { + "epoch": 2.641095890410959, + "grad_norm": 2.7750561237335205, + "learning_rate": 6.986190089358245e-07, + "loss": 1.5663, + "step": 2410 + }, + { + "epoch": 2.641095890410959, + "eval_loss": 1.8215185403823853, + "eval_runtime": 44.6832, + "eval_samples_per_second": 22.38, + "eval_steps_per_second": 0.94, + "step": 2410 + }, + { + "epoch": 2.6421917808219177, + "grad_norm": 3.1443002223968506, + "learning_rate": 6.965881397238018e-07, + "loss": 1.5878, + "step": 2411 + }, + { + "epoch": 2.643287671232877, + "grad_norm": 2.8203115463256836, + "learning_rate": 6.945572705117791e-07, + "loss": 1.5749, + "step": 2412 + }, + { + "epoch": 2.644383561643836, + "grad_norm": 2.878829002380371, + "learning_rate": 6.925264012997564e-07, + "loss": 1.5788, + "step": 2413 + }, + { + "epoch": 2.6454794520547944, + "grad_norm": 2.9213924407958984, + "learning_rate": 6.904955320877336e-07, + "loss": 1.57, + "step": 2414 + }, + { + "epoch": 2.6465753424657534, + "grad_norm": 3.0402517318725586, + "learning_rate": 6.884646628757109e-07, + "loss": 1.6245, + "step": 2415 + }, + { + "epoch": 2.6476712328767125, + "grad_norm": 2.874117612838745, + "learning_rate": 6.864337936636881e-07, + "loss": 1.6213, + "step": 2416 + }, + { + "epoch": 2.648767123287671, + "grad_norm": 2.841010570526123, + "learning_rate": 6.844029244516654e-07, + "loss": 1.5244, + "step": 2417 + }, + { + "epoch": 2.64986301369863, + "grad_norm": 2.8916752338409424, + "learning_rate": 6.823720552396426e-07, + "loss": 1.5824, + "step": 2418 + }, + { + "epoch": 2.650958904109589, + "grad_norm": 2.8366785049438477, + "learning_rate": 6.803411860276199e-07, + "loss": 1.5663, + "step": 2419 + }, + { + "epoch": 2.6520547945205477, + "grad_norm": 2.881715774536133, + "learning_rate": 6.783103168155972e-07, + "loss": 1.5599, + "step": 2420 + }, + { + "epoch": 2.6520547945205477, + "eval_loss": 1.8196642398834229, + "eval_runtime": 50.2214, + "eval_samples_per_second": 19.912, + "eval_steps_per_second": 0.836, + "step": 2420 + }, + { + "epoch": 2.6531506849315067, + "grad_norm": 2.83880615234375, + "learning_rate": 6.762794476035745e-07, + "loss": 1.5839, + "step": 2421 + }, + { + "epoch": 2.6542465753424658, + "grad_norm": 2.8980135917663574, + "learning_rate": 6.742485783915515e-07, + "loss": 1.5898, + "step": 2422 + }, + { + "epoch": 2.655342465753425, + "grad_norm": 2.82328724861145, + "learning_rate": 6.722177091795288e-07, + "loss": 1.5778, + "step": 2423 + }, + { + "epoch": 2.6564383561643834, + "grad_norm": 2.7781100273132324, + "learning_rate": 6.701868399675061e-07, + "loss": 1.5381, + "step": 2424 + }, + { + "epoch": 2.6575342465753424, + "grad_norm": 2.94514536857605, + "learning_rate": 6.681559707554834e-07, + "loss": 1.5305, + "step": 2425 + }, + { + "epoch": 2.6586301369863015, + "grad_norm": 2.9485371112823486, + "learning_rate": 6.661251015434606e-07, + "loss": 1.5348, + "step": 2426 + }, + { + "epoch": 2.6597260273972605, + "grad_norm": 2.8444228172302246, + "learning_rate": 6.640942323314379e-07, + "loss": 1.5651, + "step": 2427 + }, + { + "epoch": 2.660821917808219, + "grad_norm": 2.923813819885254, + "learning_rate": 6.620633631194151e-07, + "loss": 1.5827, + "step": 2428 + }, + { + "epoch": 2.661917808219178, + "grad_norm": 2.8742222785949707, + "learning_rate": 6.600324939073924e-07, + "loss": 1.5464, + "step": 2429 + }, + { + "epoch": 2.663013698630137, + "grad_norm": 2.963705539703369, + "learning_rate": 6.580016246953696e-07, + "loss": 1.615, + "step": 2430 + }, + { + "epoch": 2.663013698630137, + "eval_loss": 1.8202346563339233, + "eval_runtime": 53.0891, + "eval_samples_per_second": 18.836, + "eval_steps_per_second": 0.791, + "step": 2430 + }, + { + "epoch": 2.6641095890410957, + "grad_norm": 2.9104220867156982, + "learning_rate": 6.559707554833469e-07, + "loss": 1.5432, + "step": 2431 + }, + { + "epoch": 2.6652054794520548, + "grad_norm": 2.9200713634490967, + "learning_rate": 6.539398862713242e-07, + "loss": 1.5988, + "step": 2432 + }, + { + "epoch": 2.666301369863014, + "grad_norm": 2.861328363418579, + "learning_rate": 6.519090170593015e-07, + "loss": 1.5721, + "step": 2433 + }, + { + "epoch": 2.6673972602739724, + "grad_norm": 2.9656898975372314, + "learning_rate": 6.498781478472786e-07, + "loss": 1.598, + "step": 2434 + }, + { + "epoch": 2.6684931506849314, + "grad_norm": 2.9453625679016113, + "learning_rate": 6.478472786352559e-07, + "loss": 1.564, + "step": 2435 + }, + { + "epoch": 2.6695890410958905, + "grad_norm": 2.8961451053619385, + "learning_rate": 6.458164094232332e-07, + "loss": 1.5021, + "step": 2436 + }, + { + "epoch": 2.670684931506849, + "grad_norm": 2.8768153190612793, + "learning_rate": 6.437855402112105e-07, + "loss": 1.5226, + "step": 2437 + }, + { + "epoch": 2.671780821917808, + "grad_norm": 2.942457914352417, + "learning_rate": 6.417546709991877e-07, + "loss": 1.5903, + "step": 2438 + }, + { + "epoch": 2.672876712328767, + "grad_norm": 2.82218074798584, + "learning_rate": 6.39723801787165e-07, + "loss": 1.5596, + "step": 2439 + }, + { + "epoch": 2.673972602739726, + "grad_norm": 2.828130006790161, + "learning_rate": 6.376929325751423e-07, + "loss": 1.5491, + "step": 2440 + }, + { + "epoch": 2.673972602739726, + "eval_loss": 1.820644736289978, + "eval_runtime": 48.3133, + "eval_samples_per_second": 20.698, + "eval_steps_per_second": 0.869, + "step": 2440 + }, + { + "epoch": 2.675068493150685, + "grad_norm": 2.8218741416931152, + "learning_rate": 6.356620633631195e-07, + "loss": 1.5274, + "step": 2441 + }, + { + "epoch": 2.6761643835616438, + "grad_norm": 3.038788080215454, + "learning_rate": 6.336311941510967e-07, + "loss": 1.5708, + "step": 2442 + }, + { + "epoch": 2.677260273972603, + "grad_norm": 3.0761072635650635, + "learning_rate": 6.31600324939074e-07, + "loss": 1.6051, + "step": 2443 + }, + { + "epoch": 2.678356164383562, + "grad_norm": 2.881915807723999, + "learning_rate": 6.295694557270513e-07, + "loss": 1.5661, + "step": 2444 + }, + { + "epoch": 2.6794520547945204, + "grad_norm": 2.8945391178131104, + "learning_rate": 6.275385865150286e-07, + "loss": 1.5515, + "step": 2445 + }, + { + "epoch": 2.6805479452054795, + "grad_norm": 2.8491926193237305, + "learning_rate": 6.255077173030058e-07, + "loss": 1.5675, + "step": 2446 + }, + { + "epoch": 2.6816438356164385, + "grad_norm": 2.860481023788452, + "learning_rate": 6.23476848090983e-07, + "loss": 1.5485, + "step": 2447 + }, + { + "epoch": 2.682739726027397, + "grad_norm": 2.8958401679992676, + "learning_rate": 6.214459788789602e-07, + "loss": 1.5927, + "step": 2448 + }, + { + "epoch": 2.683835616438356, + "grad_norm": 2.8343098163604736, + "learning_rate": 6.194151096669375e-07, + "loss": 1.5443, + "step": 2449 + }, + { + "epoch": 2.684931506849315, + "grad_norm": 2.8697097301483154, + "learning_rate": 6.173842404549147e-07, + "loss": 1.5521, + "step": 2450 + }, + { + "epoch": 2.684931506849315, + "eval_loss": 1.8198002576828003, + "eval_runtime": 50.7198, + "eval_samples_per_second": 19.716, + "eval_steps_per_second": 0.828, + "step": 2450 + }, + { + "epoch": 2.6860273972602737, + "grad_norm": 2.8190839290618896, + "learning_rate": 6.15353371242892e-07, + "loss": 1.4845, + "step": 2451 + }, + { + "epoch": 2.6871232876712328, + "grad_norm": 2.7570931911468506, + "learning_rate": 6.133225020308693e-07, + "loss": 1.5452, + "step": 2452 + }, + { + "epoch": 2.688219178082192, + "grad_norm": 2.960784673690796, + "learning_rate": 6.112916328188466e-07, + "loss": 1.6249, + "step": 2453 + }, + { + "epoch": 2.689315068493151, + "grad_norm": 2.7679786682128906, + "learning_rate": 6.092607636068238e-07, + "loss": 1.5327, + "step": 2454 + }, + { + "epoch": 2.69041095890411, + "grad_norm": 2.8044891357421875, + "learning_rate": 6.07229894394801e-07, + "loss": 1.5435, + "step": 2455 + }, + { + "epoch": 2.6915068493150685, + "grad_norm": 2.8200135231018066, + "learning_rate": 6.051990251827783e-07, + "loss": 1.5759, + "step": 2456 + }, + { + "epoch": 2.6926027397260275, + "grad_norm": 2.904179096221924, + "learning_rate": 6.031681559707556e-07, + "loss": 1.5597, + "step": 2457 + }, + { + "epoch": 2.6936986301369865, + "grad_norm": 2.8314926624298096, + "learning_rate": 6.011372867587327e-07, + "loss": 1.5394, + "step": 2458 + }, + { + "epoch": 2.694794520547945, + "grad_norm": 2.8240511417388916, + "learning_rate": 5.9910641754671e-07, + "loss": 1.5517, + "step": 2459 + }, + { + "epoch": 2.695890410958904, + "grad_norm": 2.844630718231201, + "learning_rate": 5.970755483346873e-07, + "loss": 1.5426, + "step": 2460 + }, + { + "epoch": 2.695890410958904, + "eval_loss": 1.8208848237991333, + "eval_runtime": 52.7964, + "eval_samples_per_second": 18.941, + "eval_steps_per_second": 0.796, + "step": 2460 + }, + { + "epoch": 2.696986301369863, + "grad_norm": 2.926901340484619, + "learning_rate": 5.950446791226646e-07, + "loss": 1.5457, + "step": 2461 + }, + { + "epoch": 2.6980821917808218, + "grad_norm": 2.845421314239502, + "learning_rate": 5.930138099106418e-07, + "loss": 1.5397, + "step": 2462 + }, + { + "epoch": 2.699178082191781, + "grad_norm": 2.984661817550659, + "learning_rate": 5.909829406986191e-07, + "loss": 1.5649, + "step": 2463 + }, + { + "epoch": 2.70027397260274, + "grad_norm": 2.8373823165893555, + "learning_rate": 5.889520714865963e-07, + "loss": 1.5255, + "step": 2464 + }, + { + "epoch": 2.7013698630136984, + "grad_norm": 2.9435815811157227, + "learning_rate": 5.869212022745735e-07, + "loss": 1.5705, + "step": 2465 + }, + { + "epoch": 2.7024657534246574, + "grad_norm": 2.8383870124816895, + "learning_rate": 5.848903330625508e-07, + "loss": 1.5457, + "step": 2466 + }, + { + "epoch": 2.7035616438356165, + "grad_norm": 2.899071216583252, + "learning_rate": 5.828594638505281e-07, + "loss": 1.5474, + "step": 2467 + }, + { + "epoch": 2.704657534246575, + "grad_norm": 2.9021337032318115, + "learning_rate": 5.808285946385054e-07, + "loss": 1.5549, + "step": 2468 + }, + { + "epoch": 2.705753424657534, + "grad_norm": 2.9504072666168213, + "learning_rate": 5.787977254264827e-07, + "loss": 1.6014, + "step": 2469 + }, + { + "epoch": 2.706849315068493, + "grad_norm": 2.9223573207855225, + "learning_rate": 5.767668562144598e-07, + "loss": 1.5574, + "step": 2470 + }, + { + "epoch": 2.706849315068493, + "eval_loss": 1.8203133344650269, + "eval_runtime": 49.5424, + "eval_samples_per_second": 20.185, + "eval_steps_per_second": 0.848, + "step": 2470 + }, + { + "epoch": 2.707945205479452, + "grad_norm": 2.8095791339874268, + "learning_rate": 5.747359870024371e-07, + "loss": 1.5462, + "step": 2471 + }, + { + "epoch": 2.709041095890411, + "grad_norm": 2.8600597381591797, + "learning_rate": 5.727051177904144e-07, + "loss": 1.5872, + "step": 2472 + }, + { + "epoch": 2.71013698630137, + "grad_norm": 2.8269565105438232, + "learning_rate": 5.706742485783915e-07, + "loss": 1.5218, + "step": 2473 + }, + { + "epoch": 2.711232876712329, + "grad_norm": 2.7480826377868652, + "learning_rate": 5.686433793663688e-07, + "loss": 1.5049, + "step": 2474 + }, + { + "epoch": 2.712328767123288, + "grad_norm": 2.9086508750915527, + "learning_rate": 5.666125101543461e-07, + "loss": 1.604, + "step": 2475 + }, + { + "epoch": 2.7134246575342464, + "grad_norm": 2.791057825088501, + "learning_rate": 5.645816409423234e-07, + "loss": 1.5647, + "step": 2476 + }, + { + "epoch": 2.7145205479452055, + "grad_norm": 2.8584253787994385, + "learning_rate": 5.625507717303006e-07, + "loss": 1.5477, + "step": 2477 + }, + { + "epoch": 2.7156164383561645, + "grad_norm": 2.8320658206939697, + "learning_rate": 5.605199025182779e-07, + "loss": 1.5245, + "step": 2478 + }, + { + "epoch": 2.716712328767123, + "grad_norm": 2.867785930633545, + "learning_rate": 5.584890333062551e-07, + "loss": 1.6039, + "step": 2479 + }, + { + "epoch": 2.717808219178082, + "grad_norm": 2.942857027053833, + "learning_rate": 5.564581640942324e-07, + "loss": 1.5313, + "step": 2480 + }, + { + "epoch": 2.717808219178082, + "eval_loss": 1.821783423423767, + "eval_runtime": 51.124, + "eval_samples_per_second": 19.56, + "eval_steps_per_second": 0.822, + "step": 2480 + }, + { + "epoch": 2.718904109589041, + "grad_norm": 2.8085427284240723, + "learning_rate": 5.544272948822096e-07, + "loss": 1.5575, + "step": 2481 + }, + { + "epoch": 2.7199999999999998, + "grad_norm": 2.8479437828063965, + "learning_rate": 5.523964256701868e-07, + "loss": 1.5816, + "step": 2482 + }, + { + "epoch": 2.721095890410959, + "grad_norm": 2.8424737453460693, + "learning_rate": 5.503655564581641e-07, + "loss": 1.5434, + "step": 2483 + }, + { + "epoch": 2.722191780821918, + "grad_norm": 2.8704795837402344, + "learning_rate": 5.483346872461414e-07, + "loss": 1.5565, + "step": 2484 + }, + { + "epoch": 2.723287671232877, + "grad_norm": 2.868283987045288, + "learning_rate": 5.463038180341186e-07, + "loss": 1.5173, + "step": 2485 + }, + { + "epoch": 2.724383561643836, + "grad_norm": 2.8845019340515137, + "learning_rate": 5.442729488220959e-07, + "loss": 1.5581, + "step": 2486 + }, + { + "epoch": 2.7254794520547945, + "grad_norm": 2.83331036567688, + "learning_rate": 5.422420796100732e-07, + "loss": 1.5502, + "step": 2487 + }, + { + "epoch": 2.7265753424657535, + "grad_norm": 2.8796298503875732, + "learning_rate": 5.402112103980505e-07, + "loss": 1.5737, + "step": 2488 + }, + { + "epoch": 2.7276712328767125, + "grad_norm": 2.939396381378174, + "learning_rate": 5.381803411860276e-07, + "loss": 1.6211, + "step": 2489 + }, + { + "epoch": 2.728767123287671, + "grad_norm": 2.8413333892822266, + "learning_rate": 5.361494719740049e-07, + "loss": 1.5263, + "step": 2490 + }, + { + "epoch": 2.728767123287671, + "eval_loss": 1.819288969039917, + "eval_runtime": 50.1547, + "eval_samples_per_second": 19.938, + "eval_steps_per_second": 0.837, + "step": 2490 + }, + { + "epoch": 2.72986301369863, + "grad_norm": 2.9400246143341064, + "learning_rate": 5.341186027619822e-07, + "loss": 1.5439, + "step": 2491 + }, + { + "epoch": 2.730958904109589, + "grad_norm": 2.8259570598602295, + "learning_rate": 5.320877335499594e-07, + "loss": 1.6046, + "step": 2492 + }, + { + "epoch": 2.732054794520548, + "grad_norm": 2.9316084384918213, + "learning_rate": 5.300568643379366e-07, + "loss": 1.569, + "step": 2493 + }, + { + "epoch": 2.733150684931507, + "grad_norm": 2.8864996433258057, + "learning_rate": 5.280259951259139e-07, + "loss": 1.5384, + "step": 2494 + }, + { + "epoch": 2.734246575342466, + "grad_norm": 3.0581302642822266, + "learning_rate": 5.259951259138912e-07, + "loss": 1.5318, + "step": 2495 + }, + { + "epoch": 2.7353424657534244, + "grad_norm": 2.8676114082336426, + "learning_rate": 5.239642567018685e-07, + "loss": 1.543, + "step": 2496 + }, + { + "epoch": 2.7364383561643835, + "grad_norm": 2.935861110687256, + "learning_rate": 5.219333874898457e-07, + "loss": 1.5603, + "step": 2497 + }, + { + "epoch": 2.7375342465753425, + "grad_norm": 2.963350296020508, + "learning_rate": 5.199025182778229e-07, + "loss": 1.5479, + "step": 2498 + }, + { + "epoch": 2.7386301369863015, + "grad_norm": 2.9598515033721924, + "learning_rate": 5.178716490658002e-07, + "loss": 1.5116, + "step": 2499 + }, + { + "epoch": 2.73972602739726, + "grad_norm": 2.7877373695373535, + "learning_rate": 5.158407798537775e-07, + "loss": 1.4805, + "step": 2500 + }, + { + "epoch": 2.73972602739726, + "eval_loss": 1.8200916051864624, + "eval_runtime": 50.4013, + "eval_samples_per_second": 19.841, + "eval_steps_per_second": 0.833, + "step": 2500 + }, + { + "epoch": 2.740821917808219, + "grad_norm": 2.9831666946411133, + "learning_rate": 5.138099106417547e-07, + "loss": 1.5455, + "step": 2501 + }, + { + "epoch": 2.741917808219178, + "grad_norm": 2.9333431720733643, + "learning_rate": 5.11779041429732e-07, + "loss": 1.5281, + "step": 2502 + }, + { + "epoch": 2.743013698630137, + "grad_norm": 2.907478094100952, + "learning_rate": 5.097481722177093e-07, + "loss": 1.6127, + "step": 2503 + }, + { + "epoch": 2.744109589041096, + "grad_norm": 2.958024024963379, + "learning_rate": 5.077173030056865e-07, + "loss": 1.601, + "step": 2504 + }, + { + "epoch": 2.745205479452055, + "grad_norm": 3.0101583003997803, + "learning_rate": 5.056864337936637e-07, + "loss": 1.5734, + "step": 2505 + }, + { + "epoch": 2.746301369863014, + "grad_norm": 2.9242193698883057, + "learning_rate": 5.03655564581641e-07, + "loss": 1.6734, + "step": 2506 + }, + { + "epoch": 2.7473972602739725, + "grad_norm": 3.003774404525757, + "learning_rate": 5.016246953696182e-07, + "loss": 1.5234, + "step": 2507 + }, + { + "epoch": 2.7484931506849315, + "grad_norm": 2.88041353225708, + "learning_rate": 4.995938261575955e-07, + "loss": 1.5705, + "step": 2508 + }, + { + "epoch": 2.7495890410958905, + "grad_norm": 2.9543464183807373, + "learning_rate": 4.975629569455727e-07, + "loss": 1.6155, + "step": 2509 + }, + { + "epoch": 2.750684931506849, + "grad_norm": 2.866711139678955, + "learning_rate": 4.9553208773355e-07, + "loss": 1.5469, + "step": 2510 + }, + { + "epoch": 2.750684931506849, + "eval_loss": 1.817280888557434, + "eval_runtime": 45.7924, + "eval_samples_per_second": 21.838, + "eval_steps_per_second": 0.917, + "step": 2510 + }, + { + "epoch": 2.751780821917808, + "grad_norm": 2.94659161567688, + "learning_rate": 4.935012185215273e-07, + "loss": 1.5198, + "step": 2511 + }, + { + "epoch": 2.752876712328767, + "grad_norm": 2.782384157180786, + "learning_rate": 4.914703493095046e-07, + "loss": 1.5286, + "step": 2512 + }, + { + "epoch": 2.7539726027397258, + "grad_norm": 2.86202335357666, + "learning_rate": 4.894394800974818e-07, + "loss": 1.6437, + "step": 2513 + }, + { + "epoch": 2.755068493150685, + "grad_norm": 2.9339122772216797, + "learning_rate": 4.87408610885459e-07, + "loss": 1.5821, + "step": 2514 + }, + { + "epoch": 2.756164383561644, + "grad_norm": 2.883612871170044, + "learning_rate": 4.853777416734363e-07, + "loss": 1.5605, + "step": 2515 + }, + { + "epoch": 2.757260273972603, + "grad_norm": 2.861711025238037, + "learning_rate": 4.833468724614136e-07, + "loss": 1.5416, + "step": 2516 + }, + { + "epoch": 2.758356164383562, + "grad_norm": 2.764655590057373, + "learning_rate": 4.813160032493907e-07, + "loss": 1.5159, + "step": 2517 + }, + { + "epoch": 2.7594520547945205, + "grad_norm": 2.880544900894165, + "learning_rate": 4.79285134037368e-07, + "loss": 1.5336, + "step": 2518 + }, + { + "epoch": 2.7605479452054795, + "grad_norm": 2.851407766342163, + "learning_rate": 4.772542648253453e-07, + "loss": 1.5445, + "step": 2519 + }, + { + "epoch": 2.7616438356164386, + "grad_norm": 2.9547126293182373, + "learning_rate": 4.752233956133225e-07, + "loss": 1.6028, + "step": 2520 + }, + { + "epoch": 2.7616438356164386, + "eval_loss": 1.8189756870269775, + "eval_runtime": 50.0014, + "eval_samples_per_second": 19.999, + "eval_steps_per_second": 0.84, + "step": 2520 + }, + { + "epoch": 2.762739726027397, + "grad_norm": 2.963991641998291, + "learning_rate": 4.731925264012998e-07, + "loss": 1.565, + "step": 2521 + }, + { + "epoch": 2.763835616438356, + "grad_norm": 2.953455686569214, + "learning_rate": 4.7116165718927705e-07, + "loss": 1.5705, + "step": 2522 + }, + { + "epoch": 2.764931506849315, + "grad_norm": 2.846900701522827, + "learning_rate": 4.6913078797725427e-07, + "loss": 1.5404, + "step": 2523 + }, + { + "epoch": 2.766027397260274, + "grad_norm": 3.0525879859924316, + "learning_rate": 4.6709991876523155e-07, + "loss": 1.622, + "step": 2524 + }, + { + "epoch": 2.767123287671233, + "grad_norm": 2.948841094970703, + "learning_rate": 4.650690495532088e-07, + "loss": 1.5752, + "step": 2525 + }, + { + "epoch": 2.768219178082192, + "grad_norm": 2.8494491577148438, + "learning_rate": 4.6303818034118605e-07, + "loss": 1.503, + "step": 2526 + }, + { + "epoch": 2.7693150684931505, + "grad_norm": 2.8788719177246094, + "learning_rate": 4.610073111291633e-07, + "loss": 1.5483, + "step": 2527 + }, + { + "epoch": 2.7704109589041095, + "grad_norm": 2.86061954498291, + "learning_rate": 4.589764419171406e-07, + "loss": 1.5142, + "step": 2528 + }, + { + "epoch": 2.7715068493150685, + "grad_norm": 2.995512008666992, + "learning_rate": 4.569455727051178e-07, + "loss": 1.598, + "step": 2529 + }, + { + "epoch": 2.7726027397260276, + "grad_norm": 2.810875177383423, + "learning_rate": 4.549147034930951e-07, + "loss": 1.5676, + "step": 2530 + }, + { + "epoch": 2.7726027397260276, + "eval_loss": 1.8178433179855347, + "eval_runtime": 50.3071, + "eval_samples_per_second": 19.878, + "eval_steps_per_second": 0.835, + "step": 2530 + }, + { + "epoch": 2.773698630136986, + "grad_norm": 2.827171564102173, + "learning_rate": 4.5288383428107237e-07, + "loss": 1.5384, + "step": 2531 + }, + { + "epoch": 2.774794520547945, + "grad_norm": 2.802733898162842, + "learning_rate": 4.5085296506904954e-07, + "loss": 1.5294, + "step": 2532 + }, + { + "epoch": 2.775890410958904, + "grad_norm": 2.9027161598205566, + "learning_rate": 4.488220958570268e-07, + "loss": 1.5219, + "step": 2533 + }, + { + "epoch": 2.7769863013698632, + "grad_norm": 2.9201040267944336, + "learning_rate": 4.467912266450041e-07, + "loss": 1.5847, + "step": 2534 + }, + { + "epoch": 2.778082191780822, + "grad_norm": 2.793375253677368, + "learning_rate": 4.447603574329813e-07, + "loss": 1.5232, + "step": 2535 + }, + { + "epoch": 2.779178082191781, + "grad_norm": 2.9842793941497803, + "learning_rate": 4.427294882209586e-07, + "loss": 1.5679, + "step": 2536 + }, + { + "epoch": 2.78027397260274, + "grad_norm": 2.942744255065918, + "learning_rate": 4.4069861900893587e-07, + "loss": 1.6284, + "step": 2537 + }, + { + "epoch": 2.7813698630136985, + "grad_norm": 2.8464622497558594, + "learning_rate": 4.386677497969131e-07, + "loss": 1.5694, + "step": 2538 + }, + { + "epoch": 2.7824657534246575, + "grad_norm": 2.9041919708251953, + "learning_rate": 4.3663688058489037e-07, + "loss": 1.5758, + "step": 2539 + }, + { + "epoch": 2.7835616438356166, + "grad_norm": 2.962247133255005, + "learning_rate": 4.3460601137286764e-07, + "loss": 1.5625, + "step": 2540 + }, + { + "epoch": 2.7835616438356166, + "eval_loss": 1.8173555135726929, + "eval_runtime": 49.7157, + "eval_samples_per_second": 20.114, + "eval_steps_per_second": 0.845, + "step": 2540 + }, + { + "epoch": 2.784657534246575, + "grad_norm": 2.886990547180176, + "learning_rate": 4.325751421608449e-07, + "loss": 1.5927, + "step": 2541 + }, + { + "epoch": 2.785753424657534, + "grad_norm": 2.888997793197632, + "learning_rate": 4.3054427294882214e-07, + "loss": 1.5434, + "step": 2542 + }, + { + "epoch": 2.786849315068493, + "grad_norm": 2.7936785221099854, + "learning_rate": 4.285134037367994e-07, + "loss": 1.4914, + "step": 2543 + }, + { + "epoch": 2.787945205479452, + "grad_norm": 2.8689370155334473, + "learning_rate": 4.264825345247767e-07, + "loss": 1.5206, + "step": 2544 + }, + { + "epoch": 2.789041095890411, + "grad_norm": 2.666140079498291, + "learning_rate": 4.2445166531275386e-07, + "loss": 1.5108, + "step": 2545 + }, + { + "epoch": 2.79013698630137, + "grad_norm": 2.999361515045166, + "learning_rate": 4.2242079610073114e-07, + "loss": 1.6066, + "step": 2546 + }, + { + "epoch": 2.791232876712329, + "grad_norm": 2.919015407562256, + "learning_rate": 4.203899268887084e-07, + "loss": 1.6208, + "step": 2547 + }, + { + "epoch": 2.792328767123288, + "grad_norm": 2.774463415145874, + "learning_rate": 4.1835905767668564e-07, + "loss": 1.5628, + "step": 2548 + }, + { + "epoch": 2.7934246575342465, + "grad_norm": 2.7784156799316406, + "learning_rate": 4.163281884646629e-07, + "loss": 1.5711, + "step": 2549 + }, + { + "epoch": 2.7945205479452055, + "grad_norm": 2.8377902507781982, + "learning_rate": 4.142973192526402e-07, + "loss": 1.5766, + "step": 2550 + }, + { + "epoch": 2.7945205479452055, + "eval_loss": 1.8163820505142212, + "eval_runtime": 51.6657, + "eval_samples_per_second": 19.355, + "eval_steps_per_second": 0.813, + "step": 2550 + }, + { + "epoch": 2.7956164383561646, + "grad_norm": 3.265052318572998, + "learning_rate": 4.122664500406174e-07, + "loss": 1.5373, + "step": 2551 + }, + { + "epoch": 2.796712328767123, + "grad_norm": 2.8727524280548096, + "learning_rate": 4.102355808285947e-07, + "loss": 1.569, + "step": 2552 + }, + { + "epoch": 2.797808219178082, + "grad_norm": 2.904752016067505, + "learning_rate": 4.0820471161657196e-07, + "loss": 1.5814, + "step": 2553 + }, + { + "epoch": 2.7989041095890412, + "grad_norm": 2.7889554500579834, + "learning_rate": 4.0617384240454913e-07, + "loss": 1.4965, + "step": 2554 + }, + { + "epoch": 2.8, + "grad_norm": 2.8267621994018555, + "learning_rate": 4.041429731925264e-07, + "loss": 1.5322, + "step": 2555 + }, + { + "epoch": 2.801095890410959, + "grad_norm": 2.907101631164551, + "learning_rate": 4.021121039805037e-07, + "loss": 1.5455, + "step": 2556 + }, + { + "epoch": 2.802191780821918, + "grad_norm": 2.9485068321228027, + "learning_rate": 4.000812347684809e-07, + "loss": 1.5586, + "step": 2557 + }, + { + "epoch": 2.8032876712328765, + "grad_norm": 2.8086092472076416, + "learning_rate": 3.980503655564582e-07, + "loss": 1.5677, + "step": 2558 + }, + { + "epoch": 2.8043835616438355, + "grad_norm": 2.834923505783081, + "learning_rate": 3.9601949634443546e-07, + "loss": 1.5367, + "step": 2559 + }, + { + "epoch": 2.8054794520547945, + "grad_norm": 2.8668291568756104, + "learning_rate": 3.939886271324127e-07, + "loss": 1.5903, + "step": 2560 + }, + { + "epoch": 2.8054794520547945, + "eval_loss": 1.8173569440841675, + "eval_runtime": 48.5732, + "eval_samples_per_second": 20.588, + "eval_steps_per_second": 0.865, + "step": 2560 + }, + { + "epoch": 2.8065753424657536, + "grad_norm": 3.003795862197876, + "learning_rate": 3.9195775792038996e-07, + "loss": 1.5653, + "step": 2561 + }, + { + "epoch": 2.807671232876712, + "grad_norm": 2.8089170455932617, + "learning_rate": 3.8992688870836723e-07, + "loss": 1.5565, + "step": 2562 + }, + { + "epoch": 2.808767123287671, + "grad_norm": 2.8321988582611084, + "learning_rate": 3.8789601949634446e-07, + "loss": 1.4748, + "step": 2563 + }, + { + "epoch": 2.8098630136986302, + "grad_norm": 2.8655641078948975, + "learning_rate": 3.8586515028432173e-07, + "loss": 1.5502, + "step": 2564 + }, + { + "epoch": 2.8109589041095893, + "grad_norm": 2.7576475143432617, + "learning_rate": 3.83834281072299e-07, + "loss": 1.5138, + "step": 2565 + }, + { + "epoch": 2.812054794520548, + "grad_norm": 2.89864182472229, + "learning_rate": 3.818034118602762e-07, + "loss": 1.5553, + "step": 2566 + }, + { + "epoch": 2.813150684931507, + "grad_norm": 3.032454490661621, + "learning_rate": 3.7977254264825345e-07, + "loss": 1.5683, + "step": 2567 + }, + { + "epoch": 2.814246575342466, + "grad_norm": 2.8610739707946777, + "learning_rate": 3.7774167343623073e-07, + "loss": 1.5897, + "step": 2568 + }, + { + "epoch": 2.8153424657534245, + "grad_norm": 2.927236795425415, + "learning_rate": 3.75710804224208e-07, + "loss": 1.5342, + "step": 2569 + }, + { + "epoch": 2.8164383561643835, + "grad_norm": 2.75993013381958, + "learning_rate": 3.7367993501218523e-07, + "loss": 1.5331, + "step": 2570 + }, + { + "epoch": 2.8164383561643835, + "eval_loss": 1.8160063028335571, + "eval_runtime": 50.568, + "eval_samples_per_second": 19.775, + "eval_steps_per_second": 0.831, + "step": 2570 + }, + { + "epoch": 2.8175342465753426, + "grad_norm": 2.781172275543213, + "learning_rate": 3.716490658001625e-07, + "loss": 1.5071, + "step": 2571 + }, + { + "epoch": 2.818630136986301, + "grad_norm": 2.876645565032959, + "learning_rate": 3.696181965881398e-07, + "loss": 1.5586, + "step": 2572 + }, + { + "epoch": 2.81972602739726, + "grad_norm": 2.871903896331787, + "learning_rate": 3.67587327376117e-07, + "loss": 1.5492, + "step": 2573 + }, + { + "epoch": 2.8208219178082192, + "grad_norm": 2.888033390045166, + "learning_rate": 3.655564581640943e-07, + "loss": 1.5123, + "step": 2574 + }, + { + "epoch": 2.821917808219178, + "grad_norm": 2.7957611083984375, + "learning_rate": 3.6352558895207155e-07, + "loss": 1.5675, + "step": 2575 + }, + { + "epoch": 2.823013698630137, + "grad_norm": 2.8597512245178223, + "learning_rate": 3.614947197400488e-07, + "loss": 1.5578, + "step": 2576 + }, + { + "epoch": 2.824109589041096, + "grad_norm": 2.9219770431518555, + "learning_rate": 3.5946385052802605e-07, + "loss": 1.5407, + "step": 2577 + }, + { + "epoch": 2.825205479452055, + "grad_norm": 2.880618095397949, + "learning_rate": 3.5743298131600333e-07, + "loss": 1.6036, + "step": 2578 + }, + { + "epoch": 2.826301369863014, + "grad_norm": 2.857084274291992, + "learning_rate": 3.554021121039805e-07, + "loss": 1.5496, + "step": 2579 + }, + { + "epoch": 2.8273972602739725, + "grad_norm": 2.7948296070098877, + "learning_rate": 3.533712428919578e-07, + "loss": 1.5631, + "step": 2580 + }, + { + "epoch": 2.8273972602739725, + "eval_loss": 1.8157116174697876, + "eval_runtime": 51.0427, + "eval_samples_per_second": 19.591, + "eval_steps_per_second": 0.823, + "step": 2580 + }, + { + "epoch": 2.8284931506849316, + "grad_norm": 2.9175124168395996, + "learning_rate": 3.5134037367993505e-07, + "loss": 1.5187, + "step": 2581 + }, + { + "epoch": 2.8295890410958906, + "grad_norm": 2.923126220703125, + "learning_rate": 3.4930950446791227e-07, + "loss": 1.5466, + "step": 2582 + }, + { + "epoch": 2.830684931506849, + "grad_norm": 2.853585958480835, + "learning_rate": 3.4727863525588955e-07, + "loss": 1.5941, + "step": 2583 + }, + { + "epoch": 2.831780821917808, + "grad_norm": 2.9729127883911133, + "learning_rate": 3.452477660438668e-07, + "loss": 1.5626, + "step": 2584 + }, + { + "epoch": 2.8328767123287673, + "grad_norm": 2.900005578994751, + "learning_rate": 3.4321689683184405e-07, + "loss": 1.5117, + "step": 2585 + }, + { + "epoch": 2.833972602739726, + "grad_norm": 2.8693535327911377, + "learning_rate": 3.411860276198213e-07, + "loss": 1.5607, + "step": 2586 + }, + { + "epoch": 2.835068493150685, + "grad_norm": 2.9128522872924805, + "learning_rate": 3.391551584077986e-07, + "loss": 1.5625, + "step": 2587 + }, + { + "epoch": 2.836164383561644, + "grad_norm": 2.783365249633789, + "learning_rate": 3.3712428919577577e-07, + "loss": 1.5202, + "step": 2588 + }, + { + "epoch": 2.8372602739726025, + "grad_norm": 2.846964120864868, + "learning_rate": 3.3509341998375304e-07, + "loss": 1.5489, + "step": 2589 + }, + { + "epoch": 2.8383561643835615, + "grad_norm": 2.9657390117645264, + "learning_rate": 3.330625507717303e-07, + "loss": 1.6139, + "step": 2590 + }, + { + "epoch": 2.8383561643835615, + "eval_loss": 1.8159494400024414, + "eval_runtime": 47.7292, + "eval_samples_per_second": 20.952, + "eval_steps_per_second": 0.88, + "step": 2590 + }, + { + "epoch": 2.8394520547945206, + "grad_norm": 3.2273998260498047, + "learning_rate": 3.3103168155970754e-07, + "loss": 1.6173, + "step": 2591 + }, + { + "epoch": 2.8405479452054796, + "grad_norm": 2.946399211883545, + "learning_rate": 3.290008123476848e-07, + "loss": 1.5546, + "step": 2592 + }, + { + "epoch": 2.8416438356164386, + "grad_norm": 2.792555809020996, + "learning_rate": 3.269699431356621e-07, + "loss": 1.4948, + "step": 2593 + }, + { + "epoch": 2.842739726027397, + "grad_norm": 2.91678524017334, + "learning_rate": 3.249390739236393e-07, + "loss": 1.6064, + "step": 2594 + }, + { + "epoch": 2.8438356164383563, + "grad_norm": 2.9972245693206787, + "learning_rate": 3.229082047116166e-07, + "loss": 1.5914, + "step": 2595 + }, + { + "epoch": 2.8449315068493153, + "grad_norm": 2.853240728378296, + "learning_rate": 3.2087733549959387e-07, + "loss": 1.575, + "step": 2596 + }, + { + "epoch": 2.846027397260274, + "grad_norm": 2.9793026447296143, + "learning_rate": 3.1884646628757114e-07, + "loss": 1.5828, + "step": 2597 + }, + { + "epoch": 2.847123287671233, + "grad_norm": 2.803582191467285, + "learning_rate": 3.1681559707554837e-07, + "loss": 1.5277, + "step": 2598 + }, + { + "epoch": 2.848219178082192, + "grad_norm": 2.8925442695617676, + "learning_rate": 3.1478472786352564e-07, + "loss": 1.5616, + "step": 2599 + }, + { + "epoch": 2.8493150684931505, + "grad_norm": 2.86979603767395, + "learning_rate": 3.127538586515029e-07, + "loss": 1.5452, + "step": 2600 + }, + { + "epoch": 2.8493150684931505, + "eval_loss": 1.815985918045044, + "eval_runtime": 50.7957, + "eval_samples_per_second": 19.687, + "eval_steps_per_second": 0.827, + "step": 2600 + }, + { + "epoch": 2.8504109589041096, + "grad_norm": 4.195884704589844, + "learning_rate": 3.107229894394801e-07, + "loss": 1.4924, + "step": 2601 + }, + { + "epoch": 2.8515068493150686, + "grad_norm": 2.797089099884033, + "learning_rate": 3.0869212022745736e-07, + "loss": 1.5494, + "step": 2602 + }, + { + "epoch": 2.852602739726027, + "grad_norm": 2.924328327178955, + "learning_rate": 3.0666125101543464e-07, + "loss": 1.6201, + "step": 2603 + }, + { + "epoch": 2.853698630136986, + "grad_norm": 2.8931140899658203, + "learning_rate": 3.046303818034119e-07, + "loss": 1.5743, + "step": 2604 + }, + { + "epoch": 2.8547945205479452, + "grad_norm": 2.911238193511963, + "learning_rate": 3.0259951259138914e-07, + "loss": 1.4613, + "step": 2605 + }, + { + "epoch": 2.855890410958904, + "grad_norm": 2.8715076446533203, + "learning_rate": 3.0056864337936636e-07, + "loss": 1.54, + "step": 2606 + }, + { + "epoch": 2.856986301369863, + "grad_norm": 2.8253564834594727, + "learning_rate": 2.9853777416734364e-07, + "loss": 1.5781, + "step": 2607 + }, + { + "epoch": 2.858082191780822, + "grad_norm": 2.7086126804351807, + "learning_rate": 2.965069049553209e-07, + "loss": 1.5114, + "step": 2608 + }, + { + "epoch": 2.859178082191781, + "grad_norm": 2.8913967609405518, + "learning_rate": 2.9447603574329814e-07, + "loss": 1.5214, + "step": 2609 + }, + { + "epoch": 2.86027397260274, + "grad_norm": 2.855767250061035, + "learning_rate": 2.924451665312754e-07, + "loss": 1.4577, + "step": 2610 + }, + { + "epoch": 2.86027397260274, + "eval_loss": 1.816326379776001, + "eval_runtime": 46.3294, + "eval_samples_per_second": 21.585, + "eval_steps_per_second": 0.907, + "step": 2610 + }, + { + "epoch": 2.8613698630136986, + "grad_norm": 2.887206792831421, + "learning_rate": 2.904142973192527e-07, + "loss": 1.5266, + "step": 2611 + }, + { + "epoch": 2.8624657534246576, + "grad_norm": 2.8591439723968506, + "learning_rate": 2.883834281072299e-07, + "loss": 1.5306, + "step": 2612 + }, + { + "epoch": 2.8635616438356166, + "grad_norm": 2.872152090072632, + "learning_rate": 2.863525588952072e-07, + "loss": 1.4762, + "step": 2613 + }, + { + "epoch": 2.864657534246575, + "grad_norm": 2.8613619804382324, + "learning_rate": 2.843216896831844e-07, + "loss": 1.5567, + "step": 2614 + }, + { + "epoch": 2.8657534246575342, + "grad_norm": 2.851552963256836, + "learning_rate": 2.822908204711617e-07, + "loss": 1.5407, + "step": 2615 + }, + { + "epoch": 2.8668493150684933, + "grad_norm": 2.894739866256714, + "learning_rate": 2.8025995125913896e-07, + "loss": 1.58, + "step": 2616 + }, + { + "epoch": 2.867945205479452, + "grad_norm": 2.9518320560455322, + "learning_rate": 2.782290820471162e-07, + "loss": 1.5454, + "step": 2617 + }, + { + "epoch": 2.869041095890411, + "grad_norm": 2.912980079650879, + "learning_rate": 2.761982128350934e-07, + "loss": 1.5742, + "step": 2618 + }, + { + "epoch": 2.87013698630137, + "grad_norm": 2.855559825897217, + "learning_rate": 2.741673436230707e-07, + "loss": 1.5594, + "step": 2619 + }, + { + "epoch": 2.8712328767123285, + "grad_norm": 2.7704861164093018, + "learning_rate": 2.7213647441104796e-07, + "loss": 1.4797, + "step": 2620 + }, + { + "epoch": 2.8712328767123285, + "eval_loss": 1.8167873620986938, + "eval_runtime": 47.3047, + "eval_samples_per_second": 21.14, + "eval_steps_per_second": 0.888, + "step": 2620 + }, + { + "epoch": 2.8723287671232876, + "grad_norm": 2.8811490535736084, + "learning_rate": 2.7010560519902523e-07, + "loss": 1.5612, + "step": 2621 + }, + { + "epoch": 2.8734246575342466, + "grad_norm": 2.871553659439087, + "learning_rate": 2.6807473598700246e-07, + "loss": 1.5121, + "step": 2622 + }, + { + "epoch": 2.8745205479452056, + "grad_norm": 2.8768832683563232, + "learning_rate": 2.660438667749797e-07, + "loss": 1.5317, + "step": 2623 + }, + { + "epoch": 2.8756164383561647, + "grad_norm": 2.8615529537200928, + "learning_rate": 2.6401299756295696e-07, + "loss": 1.5397, + "step": 2624 + }, + { + "epoch": 2.8767123287671232, + "grad_norm": 2.841799020767212, + "learning_rate": 2.6198212835093423e-07, + "loss": 1.4968, + "step": 2625 + }, + { + "epoch": 2.8778082191780823, + "grad_norm": 2.8024556636810303, + "learning_rate": 2.5995125913891145e-07, + "loss": 1.5148, + "step": 2626 + }, + { + "epoch": 2.8789041095890413, + "grad_norm": 2.8324098587036133, + "learning_rate": 2.5792038992688873e-07, + "loss": 1.5472, + "step": 2627 + }, + { + "epoch": 2.88, + "grad_norm": 2.966639757156372, + "learning_rate": 2.55889520714866e-07, + "loss": 1.5724, + "step": 2628 + }, + { + "epoch": 2.881095890410959, + "grad_norm": 2.9619524478912354, + "learning_rate": 2.5385865150284323e-07, + "loss": 1.5554, + "step": 2629 + }, + { + "epoch": 2.882191780821918, + "grad_norm": 2.8983805179595947, + "learning_rate": 2.518277822908205e-07, + "loss": 1.5893, + "step": 2630 + }, + { + "epoch": 2.882191780821918, + "eval_loss": 1.8153728246688843, + "eval_runtime": 48.9891, + "eval_samples_per_second": 20.413, + "eval_steps_per_second": 0.857, + "step": 2630 + }, + { + "epoch": 2.8832876712328765, + "grad_norm": 2.887132167816162, + "learning_rate": 2.4979691307879773e-07, + "loss": 1.5001, + "step": 2631 + }, + { + "epoch": 2.8843835616438356, + "grad_norm": 2.8406999111175537, + "learning_rate": 2.47766043866775e-07, + "loss": 1.4938, + "step": 2632 + }, + { + "epoch": 2.8854794520547946, + "grad_norm": 2.84665846824646, + "learning_rate": 2.457351746547523e-07, + "loss": 1.5681, + "step": 2633 + }, + { + "epoch": 2.886575342465753, + "grad_norm": 2.8989057540893555, + "learning_rate": 2.437043054427295e-07, + "loss": 1.5454, + "step": 2634 + }, + { + "epoch": 2.8876712328767122, + "grad_norm": 2.863280773162842, + "learning_rate": 2.416734362307068e-07, + "loss": 1.4964, + "step": 2635 + }, + { + "epoch": 2.8887671232876713, + "grad_norm": 2.8970446586608887, + "learning_rate": 2.39642567018684e-07, + "loss": 1.5607, + "step": 2636 + }, + { + "epoch": 2.8898630136986303, + "grad_norm": 2.820486307144165, + "learning_rate": 2.3761169780666125e-07, + "loss": 1.5308, + "step": 2637 + }, + { + "epoch": 2.890958904109589, + "grad_norm": 2.8893818855285645, + "learning_rate": 2.3558082859463852e-07, + "loss": 1.4902, + "step": 2638 + }, + { + "epoch": 2.892054794520548, + "grad_norm": 2.8433971405029297, + "learning_rate": 2.3354995938261577e-07, + "loss": 1.585, + "step": 2639 + }, + { + "epoch": 2.893150684931507, + "grad_norm": 2.9042558670043945, + "learning_rate": 2.3151909017059302e-07, + "loss": 1.5639, + "step": 2640 + }, + { + "epoch": 2.893150684931507, + "eval_loss": 1.8146179914474487, + "eval_runtime": 46.34, + "eval_samples_per_second": 21.58, + "eval_steps_per_second": 0.906, + "step": 2640 + }, + { + "epoch": 2.894246575342466, + "grad_norm": 2.787599802017212, + "learning_rate": 2.294882209585703e-07, + "loss": 1.5085, + "step": 2641 + }, + { + "epoch": 2.8953424657534246, + "grad_norm": 2.8571665287017822, + "learning_rate": 2.2745735174654755e-07, + "loss": 1.5543, + "step": 2642 + }, + { + "epoch": 2.8964383561643836, + "grad_norm": 3.043341875076294, + "learning_rate": 2.2542648253452477e-07, + "loss": 1.5927, + "step": 2643 + }, + { + "epoch": 2.8975342465753426, + "grad_norm": 3.0167174339294434, + "learning_rate": 2.2339561332250205e-07, + "loss": 1.5806, + "step": 2644 + }, + { + "epoch": 2.8986301369863012, + "grad_norm": 2.8215010166168213, + "learning_rate": 2.213647441104793e-07, + "loss": 1.5387, + "step": 2645 + }, + { + "epoch": 2.8997260273972603, + "grad_norm": 2.868828058242798, + "learning_rate": 2.1933387489845655e-07, + "loss": 1.5346, + "step": 2646 + }, + { + "epoch": 2.9008219178082193, + "grad_norm": 2.8423750400543213, + "learning_rate": 2.1730300568643382e-07, + "loss": 1.5666, + "step": 2647 + }, + { + "epoch": 2.901917808219178, + "grad_norm": 2.9261488914489746, + "learning_rate": 2.1527213647441107e-07, + "loss": 1.5504, + "step": 2648 + }, + { + "epoch": 2.903013698630137, + "grad_norm": 2.849473237991333, + "learning_rate": 2.1324126726238835e-07, + "loss": 1.5618, + "step": 2649 + }, + { + "epoch": 2.904109589041096, + "grad_norm": 3.0464186668395996, + "learning_rate": 2.1121039805036557e-07, + "loss": 1.5992, + "step": 2650 + }, + { + "epoch": 2.904109589041096, + "eval_loss": 1.8141461610794067, + "eval_runtime": 44.6933, + "eval_samples_per_second": 22.375, + "eval_steps_per_second": 0.94, + "step": 2650 + }, + { + "epoch": 2.9052054794520545, + "grad_norm": 2.865623950958252, + "learning_rate": 2.0917952883834282e-07, + "loss": 1.5146, + "step": 2651 + }, + { + "epoch": 2.9063013698630136, + "grad_norm": 2.8237428665161133, + "learning_rate": 2.071486596263201e-07, + "loss": 1.5105, + "step": 2652 + }, + { + "epoch": 2.9073972602739726, + "grad_norm": 2.9813053607940674, + "learning_rate": 2.0511779041429734e-07, + "loss": 1.5767, + "step": 2653 + }, + { + "epoch": 2.9084931506849316, + "grad_norm": 2.9538450241088867, + "learning_rate": 2.0308692120227457e-07, + "loss": 1.5854, + "step": 2654 + }, + { + "epoch": 2.9095890410958907, + "grad_norm": 3.3947484493255615, + "learning_rate": 2.0105605199025184e-07, + "loss": 1.6149, + "step": 2655 + }, + { + "epoch": 2.9106849315068493, + "grad_norm": 2.8438758850097656, + "learning_rate": 1.990251827782291e-07, + "loss": 1.4919, + "step": 2656 + }, + { + "epoch": 2.9117808219178083, + "grad_norm": 2.9257078170776367, + "learning_rate": 1.9699431356620634e-07, + "loss": 1.5588, + "step": 2657 + }, + { + "epoch": 2.9128767123287673, + "grad_norm": 2.8315038681030273, + "learning_rate": 1.9496344435418362e-07, + "loss": 1.5438, + "step": 2658 + }, + { + "epoch": 2.913972602739726, + "grad_norm": 2.7881860733032227, + "learning_rate": 1.9293257514216087e-07, + "loss": 1.5323, + "step": 2659 + }, + { + "epoch": 2.915068493150685, + "grad_norm": 2.8819425106048584, + "learning_rate": 1.909017059301381e-07, + "loss": 1.5303, + "step": 2660 + }, + { + "epoch": 2.915068493150685, + "eval_loss": 1.8143864870071411, + "eval_runtime": 45.7876, + "eval_samples_per_second": 21.84, + "eval_steps_per_second": 0.917, + "step": 2660 + }, + { + "epoch": 2.916164383561644, + "grad_norm": 3.0435237884521484, + "learning_rate": 1.8887083671811536e-07, + "loss": 1.5721, + "step": 2661 + }, + { + "epoch": 2.9172602739726026, + "grad_norm": 2.91766619682312, + "learning_rate": 1.8683996750609261e-07, + "loss": 1.6246, + "step": 2662 + }, + { + "epoch": 2.9183561643835616, + "grad_norm": 2.804697036743164, + "learning_rate": 1.848090982940699e-07, + "loss": 1.5332, + "step": 2663 + }, + { + "epoch": 2.9194520547945206, + "grad_norm": 2.883101224899292, + "learning_rate": 1.8277822908204714e-07, + "loss": 1.5536, + "step": 2664 + }, + { + "epoch": 2.9205479452054792, + "grad_norm": 3.073230028152466, + "learning_rate": 1.807473598700244e-07, + "loss": 1.6428, + "step": 2665 + }, + { + "epoch": 2.9216438356164383, + "grad_norm": 2.8450233936309814, + "learning_rate": 1.7871649065800166e-07, + "loss": 1.5529, + "step": 2666 + }, + { + "epoch": 2.9227397260273973, + "grad_norm": 2.784237861633301, + "learning_rate": 1.766856214459789e-07, + "loss": 1.5477, + "step": 2667 + }, + { + "epoch": 2.9238356164383563, + "grad_norm": 2.7653286457061768, + "learning_rate": 1.7465475223395614e-07, + "loss": 1.5405, + "step": 2668 + }, + { + "epoch": 2.924931506849315, + "grad_norm": 2.8873486518859863, + "learning_rate": 1.726238830219334e-07, + "loss": 1.526, + "step": 2669 + }, + { + "epoch": 2.926027397260274, + "grad_norm": 2.749988555908203, + "learning_rate": 1.7059301380991066e-07, + "loss": 1.5021, + "step": 2670 + }, + { + "epoch": 2.926027397260274, + "eval_loss": 1.8143601417541504, + "eval_runtime": 48.6564, + "eval_samples_per_second": 20.552, + "eval_steps_per_second": 0.863, + "step": 2670 + }, + { + "epoch": 2.927123287671233, + "grad_norm": 2.9263057708740234, + "learning_rate": 1.6856214459788788e-07, + "loss": 1.5786, + "step": 2671 + }, + { + "epoch": 2.928219178082192, + "grad_norm": 2.774091958999634, + "learning_rate": 1.6653127538586516e-07, + "loss": 1.5095, + "step": 2672 + }, + { + "epoch": 2.9293150684931506, + "grad_norm": 2.8333022594451904, + "learning_rate": 1.645004061738424e-07, + "loss": 1.5497, + "step": 2673 + }, + { + "epoch": 2.9304109589041096, + "grad_norm": 2.833460807800293, + "learning_rate": 1.6246953696181966e-07, + "loss": 1.5869, + "step": 2674 + }, + { + "epoch": 2.9315068493150687, + "grad_norm": 2.8933019638061523, + "learning_rate": 1.6043866774979693e-07, + "loss": 1.5535, + "step": 2675 + }, + { + "epoch": 2.9326027397260273, + "grad_norm": 2.913149356842041, + "learning_rate": 1.5840779853777418e-07, + "loss": 1.526, + "step": 2676 + }, + { + "epoch": 2.9336986301369863, + "grad_norm": 2.863145351409912, + "learning_rate": 1.5637692932575146e-07, + "loss": 1.5894, + "step": 2677 + }, + { + "epoch": 2.9347945205479453, + "grad_norm": 2.8492307662963867, + "learning_rate": 1.5434606011372868e-07, + "loss": 1.581, + "step": 2678 + }, + { + "epoch": 2.935890410958904, + "grad_norm": 2.8164520263671875, + "learning_rate": 1.5231519090170596e-07, + "loss": 1.5682, + "step": 2679 + }, + { + "epoch": 2.936986301369863, + "grad_norm": 2.89471435546875, + "learning_rate": 1.5028432168968318e-07, + "loss": 1.5903, + "step": 2680 + }, + { + "epoch": 2.936986301369863, + "eval_loss": 1.814650535583496, + "eval_runtime": 45.9006, + "eval_samples_per_second": 21.786, + "eval_steps_per_second": 0.915, + "step": 2680 + }, + { + "epoch": 2.938082191780822, + "grad_norm": 2.811854839324951, + "learning_rate": 1.4825345247766046e-07, + "loss": 1.538, + "step": 2681 + }, + { + "epoch": 2.9391780821917806, + "grad_norm": 2.9468424320220947, + "learning_rate": 1.462225832656377e-07, + "loss": 1.5188, + "step": 2682 + }, + { + "epoch": 2.9402739726027396, + "grad_norm": 2.85314679145813, + "learning_rate": 1.4419171405361496e-07, + "loss": 1.5736, + "step": 2683 + }, + { + "epoch": 2.9413698630136986, + "grad_norm": 2.9153804779052734, + "learning_rate": 1.421608448415922e-07, + "loss": 1.5865, + "step": 2684 + }, + { + "epoch": 2.9424657534246577, + "grad_norm": 2.951738119125366, + "learning_rate": 1.4012997562956948e-07, + "loss": 1.5958, + "step": 2685 + }, + { + "epoch": 2.9435616438356167, + "grad_norm": 2.9108781814575195, + "learning_rate": 1.380991064175467e-07, + "loss": 1.5919, + "step": 2686 + }, + { + "epoch": 2.9446575342465753, + "grad_norm": 2.932921886444092, + "learning_rate": 1.3606823720552398e-07, + "loss": 1.5669, + "step": 2687 + }, + { + "epoch": 2.9457534246575343, + "grad_norm": 2.86723256111145, + "learning_rate": 1.3403736799350123e-07, + "loss": 1.5703, + "step": 2688 + }, + { + "epoch": 2.9468493150684933, + "grad_norm": 2.89878511428833, + "learning_rate": 1.3200649878147848e-07, + "loss": 1.5979, + "step": 2689 + }, + { + "epoch": 2.947945205479452, + "grad_norm": 2.8970377445220947, + "learning_rate": 1.2997562956945573e-07, + "loss": 1.5701, + "step": 2690 + }, + { + "epoch": 2.947945205479452, + "eval_loss": 1.8143481016159058, + "eval_runtime": 46.4632, + "eval_samples_per_second": 21.522, + "eval_steps_per_second": 0.904, + "step": 2690 + }, + { + "epoch": 2.949041095890411, + "grad_norm": 2.899817705154419, + "learning_rate": 1.27944760357433e-07, + "loss": 1.5342, + "step": 2691 + }, + { + "epoch": 2.95013698630137, + "grad_norm": 2.82865834236145, + "learning_rate": 1.2591389114541025e-07, + "loss": 1.5004, + "step": 2692 + }, + { + "epoch": 2.9512328767123286, + "grad_norm": 2.7688701152801514, + "learning_rate": 1.238830219333875e-07, + "loss": 1.5139, + "step": 2693 + }, + { + "epoch": 2.9523287671232876, + "grad_norm": 2.8859894275665283, + "learning_rate": 1.2185215272136475e-07, + "loss": 1.573, + "step": 2694 + }, + { + "epoch": 2.9534246575342467, + "grad_norm": 2.8690452575683594, + "learning_rate": 1.19821283509342e-07, + "loss": 1.5875, + "step": 2695 + }, + { + "epoch": 2.9545205479452052, + "grad_norm": 2.8701164722442627, + "learning_rate": 1.1779041429731926e-07, + "loss": 1.5144, + "step": 2696 + }, + { + "epoch": 2.9556164383561643, + "grad_norm": 2.8610260486602783, + "learning_rate": 1.1575954508529651e-07, + "loss": 1.5462, + "step": 2697 + }, + { + "epoch": 2.9567123287671233, + "grad_norm": 2.9622511863708496, + "learning_rate": 1.1372867587327377e-07, + "loss": 1.5883, + "step": 2698 + }, + { + "epoch": 2.9578082191780823, + "grad_norm": 2.773635149002075, + "learning_rate": 1.1169780666125102e-07, + "loss": 1.5379, + "step": 2699 + }, + { + "epoch": 2.958904109589041, + "grad_norm": 2.8588006496429443, + "learning_rate": 1.0966693744922827e-07, + "loss": 1.5279, + "step": 2700 + }, + { + "epoch": 2.958904109589041, + "eval_loss": 1.814355731010437, + "eval_runtime": 47.1757, + "eval_samples_per_second": 21.197, + "eval_steps_per_second": 0.89, + "step": 2700 + }, + { + "epoch": 2.96, + "grad_norm": 2.9160079956054688, + "learning_rate": 1.0763606823720554e-07, + "loss": 1.5674, + "step": 2701 + }, + { + "epoch": 2.961095890410959, + "grad_norm": 2.803605318069458, + "learning_rate": 1.0560519902518278e-07, + "loss": 1.5683, + "step": 2702 + }, + { + "epoch": 2.962191780821918, + "grad_norm": 2.8092551231384277, + "learning_rate": 1.0357432981316005e-07, + "loss": 1.6065, + "step": 2703 + }, + { + "epoch": 2.9632876712328766, + "grad_norm": 2.84529185295105, + "learning_rate": 1.0154346060113728e-07, + "loss": 1.541, + "step": 2704 + }, + { + "epoch": 2.9643835616438357, + "grad_norm": 2.8634560108184814, + "learning_rate": 9.951259138911455e-08, + "loss": 1.5629, + "step": 2705 + }, + { + "epoch": 2.9654794520547947, + "grad_norm": 2.8337323665618896, + "learning_rate": 9.748172217709181e-08, + "loss": 1.5569, + "step": 2706 + }, + { + "epoch": 2.9665753424657533, + "grad_norm": 2.856780529022217, + "learning_rate": 9.545085296506904e-08, + "loss": 1.5373, + "step": 2707 + }, + { + "epoch": 2.9676712328767123, + "grad_norm": 2.886672019958496, + "learning_rate": 9.341998375304631e-08, + "loss": 1.5336, + "step": 2708 + }, + { + "epoch": 2.9687671232876713, + "grad_norm": 2.8349647521972656, + "learning_rate": 9.138911454102357e-08, + "loss": 1.5891, + "step": 2709 + }, + { + "epoch": 2.96986301369863, + "grad_norm": 2.816354274749756, + "learning_rate": 8.935824532900083e-08, + "loss": 1.5382, + "step": 2710 + }, + { + "epoch": 2.96986301369863, + "eval_loss": 1.8138083219528198, + "eval_runtime": 44.7215, + "eval_samples_per_second": 22.361, + "eval_steps_per_second": 0.939, + "step": 2710 + }, + { + "epoch": 2.970958904109589, + "grad_norm": 2.7551629543304443, + "learning_rate": 8.732737611697807e-08, + "loss": 1.4825, + "step": 2711 + }, + { + "epoch": 2.972054794520548, + "grad_norm": 2.8292622566223145, + "learning_rate": 8.529650690495533e-08, + "loss": 1.5658, + "step": 2712 + }, + { + "epoch": 2.9731506849315066, + "grad_norm": 2.9308197498321533, + "learning_rate": 8.326563769293258e-08, + "loss": 1.5939, + "step": 2713 + }, + { + "epoch": 2.9742465753424656, + "grad_norm": 2.8841938972473145, + "learning_rate": 8.123476848090983e-08, + "loss": 1.5677, + "step": 2714 + }, + { + "epoch": 2.9753424657534246, + "grad_norm": 2.9214682579040527, + "learning_rate": 7.920389926888709e-08, + "loss": 1.5646, + "step": 2715 + }, + { + "epoch": 2.9764383561643837, + "grad_norm": 2.8754327297210693, + "learning_rate": 7.717303005686434e-08, + "loss": 1.6127, + "step": 2716 + }, + { + "epoch": 2.9775342465753427, + "grad_norm": 2.7910163402557373, + "learning_rate": 7.514216084484159e-08, + "loss": 1.55, + "step": 2717 + }, + { + "epoch": 2.9786301369863013, + "grad_norm": 2.8687946796417236, + "learning_rate": 7.311129163281885e-08, + "loss": 1.5808, + "step": 2718 + }, + { + "epoch": 2.9797260273972603, + "grad_norm": 2.8112573623657227, + "learning_rate": 7.10804224207961e-08, + "loss": 1.5494, + "step": 2719 + }, + { + "epoch": 2.9808219178082194, + "grad_norm": 2.967491626739502, + "learning_rate": 6.904955320877335e-08, + "loss": 1.5683, + "step": 2720 + }, + { + "epoch": 2.9808219178082194, + "eval_loss": 1.814038872718811, + "eval_runtime": 48.8728, + "eval_samples_per_second": 20.461, + "eval_steps_per_second": 0.859, + "step": 2720 + }, + { + "epoch": 2.981917808219178, + "grad_norm": 2.8173718452453613, + "learning_rate": 6.701868399675061e-08, + "loss": 1.5512, + "step": 2721 + }, + { + "epoch": 2.983013698630137, + "grad_norm": 2.8573005199432373, + "learning_rate": 6.498781478472786e-08, + "loss": 1.6114, + "step": 2722 + }, + { + "epoch": 2.984109589041096, + "grad_norm": 2.8853330612182617, + "learning_rate": 6.295694557270513e-08, + "loss": 1.5679, + "step": 2723 + }, + { + "epoch": 2.9852054794520546, + "grad_norm": 2.9632973670959473, + "learning_rate": 6.092607636068238e-08, + "loss": 1.5592, + "step": 2724 + }, + { + "epoch": 2.9863013698630136, + "grad_norm": 2.835488796234131, + "learning_rate": 5.889520714865963e-08, + "loss": 1.5529, + "step": 2725 + }, + { + "epoch": 2.9873972602739727, + "grad_norm": 2.8867247104644775, + "learning_rate": 5.686433793663689e-08, + "loss": 1.6375, + "step": 2726 + }, + { + "epoch": 2.9884931506849313, + "grad_norm": 2.8852715492248535, + "learning_rate": 5.4833468724614136e-08, + "loss": 1.566, + "step": 2727 + }, + { + "epoch": 2.9895890410958903, + "grad_norm": 2.841155767440796, + "learning_rate": 5.280259951259139e-08, + "loss": 1.4918, + "step": 2728 + }, + { + "epoch": 2.9906849315068493, + "grad_norm": 2.9603850841522217, + "learning_rate": 5.077173030056864e-08, + "loss": 1.5116, + "step": 2729 + }, + { + "epoch": 2.9917808219178084, + "grad_norm": 2.9059109687805176, + "learning_rate": 4.8740861088545904e-08, + "loss": 1.6116, + "step": 2730 + }, + { + "epoch": 2.9917808219178084, + "eval_loss": 1.8143436908721924, + "eval_runtime": 52.1426, + "eval_samples_per_second": 19.178, + "eval_steps_per_second": 0.805, + "step": 2730 + }, + { + "epoch": 2.9928767123287674, + "grad_norm": 2.8827717304229736, + "learning_rate": 4.6709991876523154e-08, + "loss": 1.6186, + "step": 2731 + }, + { + "epoch": 2.993972602739726, + "grad_norm": 2.85530948638916, + "learning_rate": 4.4679122664500416e-08, + "loss": 1.5458, + "step": 2732 + }, + { + "epoch": 2.995068493150685, + "grad_norm": 3.1319854259490967, + "learning_rate": 4.2648253452477665e-08, + "loss": 1.605, + "step": 2733 + }, + { + "epoch": 2.996164383561644, + "grad_norm": 2.914677143096924, + "learning_rate": 4.0617384240454915e-08, + "loss": 1.5743, + "step": 2734 + }, + { + "epoch": 2.9972602739726026, + "grad_norm": 2.8031442165374756, + "learning_rate": 3.858651502843217e-08, + "loss": 1.556, + "step": 2735 + }, + { + "epoch": 2.9983561643835617, + "grad_norm": 2.9505374431610107, + "learning_rate": 3.6555645816409427e-08, + "loss": 1.602, + "step": 2736 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 8.528958544014213e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-2736/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/generation_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/latest b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/latest new file mode 100644 index 0000000000000000000000000000000000000000..f0b47ce15fff9a01b2a416a473b2148085048a50 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/latest @@ -0,0 +1 @@ +global_step500 \ No newline at end of file diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/scheduler.pt b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..05bb43cf1ef0500c9052c921cddc3759722936cc --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6c29c3e135ab80dd07845a2238d1e0d3fa5d1e843e3b253e53f208d3a15409e2 +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/special_tokens_map.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/tokenizer.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..9c92c0270bdc0f9469f111f0085c1673351a3860 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/trainer_state.json @@ -0,0 +1,3933 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.547945205479452, + "eval_steps": 10, + "global_step": 500, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.55865470468096e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-500/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/generation_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/generation_config.json new file mode 100644 index 0000000000000000000000000000000000000000..9c389b841a9a29ddd904e02d1eb0e08dcce82ad9 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/generation_config.json @@ -0,0 +1,9 @@ +{ + "_from_model_config": true, + "bos_token_id": 128000, + "do_sample": true, + "eos_token_id": 128001, + "temperature": 0.6, + "top_p": 0.9, + "transformers_version": "4.45.1" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/latest b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/latest new file mode 100644 index 0000000000000000000000000000000000000000..12cae1adf3af8546b4141c6f62261c8e99839a54 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/latest @@ -0,0 +1 @@ +global_step600 \ No newline at end of file diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/rng_state_0.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..d5f37008b65b2b6b774ba8ca3e08ea7f2886beeb --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:898e9b2b9cf458e184c0b3debd303a4fd7c10dad3b82242d58682e89cace8b1e +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/rng_state_1.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..43698f33ac8bdb91266f12645eda8b36787da828 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:050132a910188198ac7405dcc27a0ec731bdadfd7fd9c07e7468ac838c90fae0 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..90d0b3f139f4cb1368a680fad6f12df4c66f078f --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:21ab24991874894fe74f6d6882102cbe2b931a423ab076862a8f74e33fa92f52 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/scheduler.pt b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..5dd84a5da536d2e95eb95bb80697a368ccec01b4 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fbf17210f2869589626b5383f7a713ce38ea6571d2f633aa0d0ba53e1aca6163 +size 627 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/special_tokens_map.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/tokenizer_config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..5684b14e2756b56aa397c1324e239b442162b34a --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/tokenizer_config.json @@ -0,0 +1,2070 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128256": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "[PAD]", + "tokenizer_class": "PreTrainedTokenizerFast" +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/trainer_state.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..7ae1da883921438a625e30c353f038b31d85a058 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/trainer_state.json @@ -0,0 +1,4713 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.6575342465753424, + "eval_steps": 10, + "global_step": 600, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.001095890410958904, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0382, + "step": 1 + }, + { + "epoch": 0.002191780821917808, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0278, + "step": 2 + }, + { + "epoch": 0.0032876712328767125, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9992, + "step": 3 + }, + { + "epoch": 0.004383561643835616, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0187, + "step": 4 + }, + { + "epoch": 0.005479452054794521, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0015, + "step": 5 + }, + { + "epoch": 0.006575342465753425, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.1223, + "step": 6 + }, + { + "epoch": 0.007671232876712329, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0383, + "step": 7 + }, + { + "epoch": 0.008767123287671232, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.058, + "step": 8 + }, + { + "epoch": 0.009863013698630137, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0337, + "step": 9 + }, + { + "epoch": 0.010958904109589041, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0863, + "step": 10 + }, + { + "epoch": 0.010958904109589041, + "eval_loss": 3.080348253250122, + "eval_runtime": 43.9716, + "eval_samples_per_second": 22.742, + "eval_steps_per_second": 0.955, + "step": 10 + }, + { + "epoch": 0.012054794520547946, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0636, + "step": 11 + }, + { + "epoch": 0.01315068493150685, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 2.9879, + "step": 12 + }, + { + "epoch": 0.014246575342465753, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0493, + "step": 13 + }, + { + "epoch": 0.015342465753424657, + "grad_norm": 0.0, + "learning_rate": 0.0, + "loss": 3.0416, + "step": 14 + }, + { + "epoch": 0.01643835616438356, + "grad_norm": 5.838740825653076, + "learning_rate": 1.8248175182481753e-08, + "loss": 3.0777, + "step": 15 + }, + { + "epoch": 0.017534246575342465, + "grad_norm": 6.277613162994385, + "learning_rate": 3.6496350364963505e-08, + "loss": 3.0301, + "step": 16 + }, + { + "epoch": 0.01863013698630137, + "grad_norm": 6.362560272216797, + "learning_rate": 5.4744525547445264e-08, + "loss": 3.0286, + "step": 17 + }, + { + "epoch": 0.019726027397260273, + "grad_norm": 6.523171424865723, + "learning_rate": 7.299270072992701e-08, + "loss": 3.0652, + "step": 18 + }, + { + "epoch": 0.020821917808219178, + "grad_norm": 6.391225337982178, + "learning_rate": 9.124087591240876e-08, + "loss": 3.0152, + "step": 19 + }, + { + "epoch": 0.021917808219178082, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0877, + "step": 20 + }, + { + "epoch": 0.021917808219178082, + "eval_loss": 3.0788941383361816, + "eval_runtime": 43.1314, + "eval_samples_per_second": 23.185, + "eval_steps_per_second": 0.974, + "step": 20 + }, + { + "epoch": 0.023013698630136987, + "grad_norm": 5.987354278564453, + "learning_rate": 1.0948905109489053e-07, + "loss": 3.0674, + "step": 21 + }, + { + "epoch": 0.02410958904109589, + "grad_norm": 6.226771354675293, + "learning_rate": 1.2773722627737227e-07, + "loss": 2.9923, + "step": 22 + }, + { + "epoch": 0.025205479452054796, + "grad_norm": 6.025713920593262, + "learning_rate": 1.4598540145985402e-07, + "loss": 3.0266, + "step": 23 + }, + { + "epoch": 0.0263013698630137, + "grad_norm": 6.291508197784424, + "learning_rate": 1.642335766423358e-07, + "loss": 3.0645, + "step": 24 + }, + { + "epoch": 0.0273972602739726, + "grad_norm": 5.723764896392822, + "learning_rate": 1.8248175182481753e-07, + "loss": 3.0645, + "step": 25 + }, + { + "epoch": 0.028493150684931506, + "grad_norm": 5.392496585845947, + "learning_rate": 2.0072992700729928e-07, + "loss": 3.0524, + "step": 26 + }, + { + "epoch": 0.02958904109589041, + "grad_norm": 5.1719255447387695, + "learning_rate": 2.1897810218978106e-07, + "loss": 3.0584, + "step": 27 + }, + { + "epoch": 0.030684931506849315, + "grad_norm": 4.762581825256348, + "learning_rate": 2.3722627737226278e-07, + "loss": 3.0029, + "step": 28 + }, + { + "epoch": 0.03178082191780822, + "grad_norm": 4.742554187774658, + "learning_rate": 2.5547445255474454e-07, + "loss": 3.0053, + "step": 29 + }, + { + "epoch": 0.03287671232876712, + "grad_norm": 4.693714141845703, + "learning_rate": 2.737226277372263e-07, + "loss": 2.9717, + "step": 30 + }, + { + "epoch": 0.03287671232876712, + "eval_loss": 3.0545272827148438, + "eval_runtime": 42.3621, + "eval_samples_per_second": 23.606, + "eval_steps_per_second": 0.991, + "step": 30 + }, + { + "epoch": 0.03397260273972603, + "grad_norm": 4.737666606903076, + "learning_rate": 2.9197080291970804e-07, + "loss": 3.0199, + "step": 31 + }, + { + "epoch": 0.03506849315068493, + "grad_norm": 4.3344011306762695, + "learning_rate": 3.102189781021898e-07, + "loss": 2.9753, + "step": 32 + }, + { + "epoch": 0.03616438356164384, + "grad_norm": 4.522485256195068, + "learning_rate": 3.284671532846716e-07, + "loss": 2.9572, + "step": 33 + }, + { + "epoch": 0.03726027397260274, + "grad_norm": 4.1835713386535645, + "learning_rate": 3.467153284671533e-07, + "loss": 2.9538, + "step": 34 + }, + { + "epoch": 0.038356164383561646, + "grad_norm": 4.331220626831055, + "learning_rate": 3.6496350364963505e-07, + "loss": 2.9624, + "step": 35 + }, + { + "epoch": 0.03945205479452055, + "grad_norm": 4.14863920211792, + "learning_rate": 3.8321167883211686e-07, + "loss": 2.9951, + "step": 36 + }, + { + "epoch": 0.040547945205479455, + "grad_norm": 3.8865740299224854, + "learning_rate": 4.0145985401459856e-07, + "loss": 2.9548, + "step": 37 + }, + { + "epoch": 0.041643835616438356, + "grad_norm": 3.7169909477233887, + "learning_rate": 4.197080291970803e-07, + "loss": 2.8997, + "step": 38 + }, + { + "epoch": 0.042739726027397264, + "grad_norm": 3.520193576812744, + "learning_rate": 4.379562043795621e-07, + "loss": 2.9893, + "step": 39 + }, + { + "epoch": 0.043835616438356165, + "grad_norm": 3.556771993637085, + "learning_rate": 4.562043795620438e-07, + "loss": 2.8876, + "step": 40 + }, + { + "epoch": 0.043835616438356165, + "eval_loss": 2.957977294921875, + "eval_runtime": 42.8306, + "eval_samples_per_second": 23.348, + "eval_steps_per_second": 0.981, + "step": 40 + }, + { + "epoch": 0.044931506849315066, + "grad_norm": 3.554539918899536, + "learning_rate": 4.7445255474452557e-07, + "loss": 2.9108, + "step": 41 + }, + { + "epoch": 0.046027397260273974, + "grad_norm": 3.3728439807891846, + "learning_rate": 4.927007299270073e-07, + "loss": 2.9426, + "step": 42 + }, + { + "epoch": 0.047123287671232875, + "grad_norm": 3.247314453125, + "learning_rate": 5.109489051094891e-07, + "loss": 2.9104, + "step": 43 + }, + { + "epoch": 0.04821917808219178, + "grad_norm": 3.1744000911712646, + "learning_rate": 5.291970802919708e-07, + "loss": 2.9222, + "step": 44 + }, + { + "epoch": 0.049315068493150684, + "grad_norm": 2.894174098968506, + "learning_rate": 5.474452554744526e-07, + "loss": 2.9061, + "step": 45 + }, + { + "epoch": 0.05041095890410959, + "grad_norm": 2.8599677085876465, + "learning_rate": 5.656934306569343e-07, + "loss": 2.8591, + "step": 46 + }, + { + "epoch": 0.05150684931506849, + "grad_norm": 2.9902148246765137, + "learning_rate": 5.839416058394161e-07, + "loss": 2.8385, + "step": 47 + }, + { + "epoch": 0.0526027397260274, + "grad_norm": 2.9698832035064697, + "learning_rate": 6.021897810218979e-07, + "loss": 2.865, + "step": 48 + }, + { + "epoch": 0.0536986301369863, + "grad_norm": 3.2346339225769043, + "learning_rate": 6.204379562043796e-07, + "loss": 2.7998, + "step": 49 + }, + { + "epoch": 0.0547945205479452, + "grad_norm": 2.5236120223999023, + "learning_rate": 6.386861313868614e-07, + "loss": 2.906, + "step": 50 + }, + { + "epoch": 0.0547945205479452, + "eval_loss": 2.862985849380493, + "eval_runtime": 41.7778, + "eval_samples_per_second": 23.936, + "eval_steps_per_second": 1.005, + "step": 50 + }, + { + "epoch": 0.05589041095890411, + "grad_norm": 2.6808536052703857, + "learning_rate": 6.569343065693432e-07, + "loss": 2.7992, + "step": 51 + }, + { + "epoch": 0.05698630136986301, + "grad_norm": 2.5073208808898926, + "learning_rate": 6.751824817518248e-07, + "loss": 2.8256, + "step": 52 + }, + { + "epoch": 0.05808219178082192, + "grad_norm": 2.3364014625549316, + "learning_rate": 6.934306569343066e-07, + "loss": 2.7532, + "step": 53 + }, + { + "epoch": 0.05917808219178082, + "grad_norm": 2.2930450439453125, + "learning_rate": 7.116788321167883e-07, + "loss": 2.7555, + "step": 54 + }, + { + "epoch": 0.06027397260273973, + "grad_norm": 2.165067672729492, + "learning_rate": 7.299270072992701e-07, + "loss": 2.7711, + "step": 55 + }, + { + "epoch": 0.06136986301369863, + "grad_norm": 2.165930986404419, + "learning_rate": 7.48175182481752e-07, + "loss": 2.7825, + "step": 56 + }, + { + "epoch": 0.06246575342465754, + "grad_norm": 2.3751773834228516, + "learning_rate": 7.664233576642337e-07, + "loss": 2.7904, + "step": 57 + }, + { + "epoch": 0.06356164383561644, + "grad_norm": 2.2055346965789795, + "learning_rate": 7.846715328467154e-07, + "loss": 2.769, + "step": 58 + }, + { + "epoch": 0.06465753424657535, + "grad_norm": 2.1593515872955322, + "learning_rate": 8.029197080291971e-07, + "loss": 2.7125, + "step": 59 + }, + { + "epoch": 0.06575342465753424, + "grad_norm": 2.033651828765869, + "learning_rate": 8.211678832116789e-07, + "loss": 2.6806, + "step": 60 + }, + { + "epoch": 0.06575342465753424, + "eval_loss": 2.783937692642212, + "eval_runtime": 42.4961, + "eval_samples_per_second": 23.532, + "eval_steps_per_second": 0.988, + "step": 60 + }, + { + "epoch": 0.06684931506849315, + "grad_norm": 2.1687231063842773, + "learning_rate": 8.394160583941606e-07, + "loss": 2.7175, + "step": 61 + }, + { + "epoch": 0.06794520547945206, + "grad_norm": 2.0312962532043457, + "learning_rate": 8.576642335766425e-07, + "loss": 2.6937, + "step": 62 + }, + { + "epoch": 0.06904109589041096, + "grad_norm": 1.968230128288269, + "learning_rate": 8.759124087591242e-07, + "loss": 2.7144, + "step": 63 + }, + { + "epoch": 0.07013698630136986, + "grad_norm": 1.9065202474594116, + "learning_rate": 8.94160583941606e-07, + "loss": 2.7199, + "step": 64 + }, + { + "epoch": 0.07123287671232877, + "grad_norm": 1.9251646995544434, + "learning_rate": 9.124087591240876e-07, + "loss": 2.6797, + "step": 65 + }, + { + "epoch": 0.07232876712328767, + "grad_norm": 1.9477510452270508, + "learning_rate": 9.306569343065694e-07, + "loss": 2.7, + "step": 66 + }, + { + "epoch": 0.07342465753424658, + "grad_norm": 2.09024977684021, + "learning_rate": 9.489051094890511e-07, + "loss": 2.6857, + "step": 67 + }, + { + "epoch": 0.07452054794520548, + "grad_norm": 2.124732732772827, + "learning_rate": 9.67153284671533e-07, + "loss": 2.7261, + "step": 68 + }, + { + "epoch": 0.07561643835616438, + "grad_norm": 2.019111394882202, + "learning_rate": 9.854014598540146e-07, + "loss": 2.6651, + "step": 69 + }, + { + "epoch": 0.07671232876712329, + "grad_norm": 2.086594820022583, + "learning_rate": 1.0036496350364965e-06, + "loss": 2.6227, + "step": 70 + }, + { + "epoch": 0.07671232876712329, + "eval_loss": 2.710885763168335, + "eval_runtime": 45.9727, + "eval_samples_per_second": 21.752, + "eval_steps_per_second": 0.914, + "step": 70 + }, + { + "epoch": 0.07780821917808219, + "grad_norm": 1.9335044622421265, + "learning_rate": 1.0218978102189781e-06, + "loss": 2.6935, + "step": 71 + }, + { + "epoch": 0.0789041095890411, + "grad_norm": 1.9765923023223877, + "learning_rate": 1.04014598540146e-06, + "loss": 2.5846, + "step": 72 + }, + { + "epoch": 0.08, + "grad_norm": 2.1063954830169678, + "learning_rate": 1.0583941605839416e-06, + "loss": 2.5799, + "step": 73 + }, + { + "epoch": 0.08109589041095891, + "grad_norm": 2.2129030227661133, + "learning_rate": 1.0766423357664233e-06, + "loss": 2.5978, + "step": 74 + }, + { + "epoch": 0.0821917808219178, + "grad_norm": 1.9766309261322021, + "learning_rate": 1.0948905109489052e-06, + "loss": 2.6169, + "step": 75 + }, + { + "epoch": 0.08328767123287671, + "grad_norm": 2.087632179260254, + "learning_rate": 1.113138686131387e-06, + "loss": 2.6272, + "step": 76 + }, + { + "epoch": 0.08438356164383562, + "grad_norm": 2.062459707260132, + "learning_rate": 1.1313868613138687e-06, + "loss": 2.5499, + "step": 77 + }, + { + "epoch": 0.08547945205479453, + "grad_norm": 1.9880034923553467, + "learning_rate": 1.1496350364963505e-06, + "loss": 2.5344, + "step": 78 + }, + { + "epoch": 0.08657534246575342, + "grad_norm": 2.105504274368286, + "learning_rate": 1.1678832116788322e-06, + "loss": 2.5404, + "step": 79 + }, + { + "epoch": 0.08767123287671233, + "grad_norm": 2.148643970489502, + "learning_rate": 1.1861313868613138e-06, + "loss": 2.5325, + "step": 80 + }, + { + "epoch": 0.08767123287671233, + "eval_loss": 2.624393939971924, + "eval_runtime": 44.3157, + "eval_samples_per_second": 22.565, + "eval_steps_per_second": 0.948, + "step": 80 + }, + { + "epoch": 0.08876712328767124, + "grad_norm": 2.437154531478882, + "learning_rate": 1.2043795620437959e-06, + "loss": 2.5371, + "step": 81 + }, + { + "epoch": 0.08986301369863013, + "grad_norm": 2.3247954845428467, + "learning_rate": 1.2226277372262775e-06, + "loss": 2.5658, + "step": 82 + }, + { + "epoch": 0.09095890410958904, + "grad_norm": 2.3776540756225586, + "learning_rate": 1.2408759124087592e-06, + "loss": 2.5187, + "step": 83 + }, + { + "epoch": 0.09205479452054795, + "grad_norm": 2.66748046875, + "learning_rate": 1.259124087591241e-06, + "loss": 2.5386, + "step": 84 + }, + { + "epoch": 0.09315068493150686, + "grad_norm": 2.408660888671875, + "learning_rate": 1.2773722627737229e-06, + "loss": 2.5182, + "step": 85 + }, + { + "epoch": 0.09424657534246575, + "grad_norm": 2.84110689163208, + "learning_rate": 1.2956204379562045e-06, + "loss": 2.465, + "step": 86 + }, + { + "epoch": 0.09534246575342466, + "grad_norm": 2.413292407989502, + "learning_rate": 1.3138686131386864e-06, + "loss": 2.4806, + "step": 87 + }, + { + "epoch": 0.09643835616438357, + "grad_norm": 2.380706310272217, + "learning_rate": 1.332116788321168e-06, + "loss": 2.4039, + "step": 88 + }, + { + "epoch": 0.09753424657534246, + "grad_norm": 2.974210262298584, + "learning_rate": 1.3503649635036497e-06, + "loss": 2.4513, + "step": 89 + }, + { + "epoch": 0.09863013698630137, + "grad_norm": 2.645435333251953, + "learning_rate": 1.3686131386861315e-06, + "loss": 2.3779, + "step": 90 + }, + { + "epoch": 0.09863013698630137, + "eval_loss": 2.53607439994812, + "eval_runtime": 45.3275, + "eval_samples_per_second": 22.062, + "eval_steps_per_second": 0.927, + "step": 90 + }, + { + "epoch": 0.09972602739726028, + "grad_norm": 2.8195626735687256, + "learning_rate": 1.3868613138686132e-06, + "loss": 2.4685, + "step": 91 + }, + { + "epoch": 0.10082191780821918, + "grad_norm": 2.497040033340454, + "learning_rate": 1.405109489051095e-06, + "loss": 2.4279, + "step": 92 + }, + { + "epoch": 0.10191780821917808, + "grad_norm": 2.9897544384002686, + "learning_rate": 1.4233576642335767e-06, + "loss": 2.4563, + "step": 93 + }, + { + "epoch": 0.10301369863013699, + "grad_norm": 2.9979658126831055, + "learning_rate": 1.4416058394160583e-06, + "loss": 2.4162, + "step": 94 + }, + { + "epoch": 0.10410958904109589, + "grad_norm": 3.053898572921753, + "learning_rate": 1.4598540145985402e-06, + "loss": 2.381, + "step": 95 + }, + { + "epoch": 0.1052054794520548, + "grad_norm": 2.5075783729553223, + "learning_rate": 1.478102189781022e-06, + "loss": 2.4455, + "step": 96 + }, + { + "epoch": 0.1063013698630137, + "grad_norm": 2.3086116313934326, + "learning_rate": 1.496350364963504e-06, + "loss": 2.4191, + "step": 97 + }, + { + "epoch": 0.1073972602739726, + "grad_norm": 2.3600242137908936, + "learning_rate": 1.5145985401459856e-06, + "loss": 2.4246, + "step": 98 + }, + { + "epoch": 0.10849315068493151, + "grad_norm": 2.6032862663269043, + "learning_rate": 1.5328467153284674e-06, + "loss": 2.383, + "step": 99 + }, + { + "epoch": 0.1095890410958904, + "grad_norm": 2.326075315475464, + "learning_rate": 1.551094890510949e-06, + "loss": 2.326, + "step": 100 + }, + { + "epoch": 0.1095890410958904, + "eval_loss": 2.4669740200042725, + "eval_runtime": 44.8776, + "eval_samples_per_second": 22.283, + "eval_steps_per_second": 0.936, + "step": 100 + }, + { + "epoch": 0.11068493150684931, + "grad_norm": 2.90632963180542, + "learning_rate": 1.5693430656934307e-06, + "loss": 2.3305, + "step": 101 + }, + { + "epoch": 0.11178082191780822, + "grad_norm": 2.6117913722991943, + "learning_rate": 1.5875912408759126e-06, + "loss": 2.3863, + "step": 102 + }, + { + "epoch": 0.11287671232876713, + "grad_norm": 2.6096816062927246, + "learning_rate": 1.6058394160583942e-06, + "loss": 2.435, + "step": 103 + }, + { + "epoch": 0.11397260273972602, + "grad_norm": 2.7738752365112305, + "learning_rate": 1.624087591240876e-06, + "loss": 2.3441, + "step": 104 + }, + { + "epoch": 0.11506849315068493, + "grad_norm": 2.6298956871032715, + "learning_rate": 1.6423357664233577e-06, + "loss": 2.2895, + "step": 105 + }, + { + "epoch": 0.11616438356164384, + "grad_norm": 2.4780426025390625, + "learning_rate": 1.6605839416058394e-06, + "loss": 2.3218, + "step": 106 + }, + { + "epoch": 0.11726027397260275, + "grad_norm": 2.4928417205810547, + "learning_rate": 1.6788321167883212e-06, + "loss": 2.3345, + "step": 107 + }, + { + "epoch": 0.11835616438356164, + "grad_norm": 2.952632427215576, + "learning_rate": 1.6970802919708029e-06, + "loss": 2.2623, + "step": 108 + }, + { + "epoch": 0.11945205479452055, + "grad_norm": 2.6794142723083496, + "learning_rate": 1.715328467153285e-06, + "loss": 2.317, + "step": 109 + }, + { + "epoch": 0.12054794520547946, + "grad_norm": 2.6773149967193604, + "learning_rate": 1.7335766423357666e-06, + "loss": 2.3162, + "step": 110 + }, + { + "epoch": 0.12054794520547946, + "eval_loss": 2.4090704917907715, + "eval_runtime": 42.9751, + "eval_samples_per_second": 23.269, + "eval_steps_per_second": 0.977, + "step": 110 + }, + { + "epoch": 0.12164383561643835, + "grad_norm": 2.8382351398468018, + "learning_rate": 1.7518248175182485e-06, + "loss": 2.3507, + "step": 111 + }, + { + "epoch": 0.12273972602739726, + "grad_norm": 2.862053871154785, + "learning_rate": 1.77007299270073e-06, + "loss": 2.3239, + "step": 112 + }, + { + "epoch": 0.12383561643835617, + "grad_norm": 2.6093366146087646, + "learning_rate": 1.788321167883212e-06, + "loss": 2.3677, + "step": 113 + }, + { + "epoch": 0.12493150684931507, + "grad_norm": 2.586615562438965, + "learning_rate": 1.8065693430656936e-06, + "loss": 2.3141, + "step": 114 + }, + { + "epoch": 0.12602739726027398, + "grad_norm": 2.8091650009155273, + "learning_rate": 1.8248175182481753e-06, + "loss": 2.2529, + "step": 115 + }, + { + "epoch": 0.12712328767123288, + "grad_norm": 2.581000566482544, + "learning_rate": 1.8430656934306571e-06, + "loss": 2.2732, + "step": 116 + }, + { + "epoch": 0.12821917808219177, + "grad_norm": 2.636491298675537, + "learning_rate": 1.8613138686131388e-06, + "loss": 2.267, + "step": 117 + }, + { + "epoch": 0.1293150684931507, + "grad_norm": 2.5760490894317627, + "learning_rate": 1.8795620437956206e-06, + "loss": 2.2642, + "step": 118 + }, + { + "epoch": 0.1304109589041096, + "grad_norm": 2.506948709487915, + "learning_rate": 1.8978102189781023e-06, + "loss": 2.2695, + "step": 119 + }, + { + "epoch": 0.13150684931506848, + "grad_norm": 2.3089029788970947, + "learning_rate": 1.916058394160584e-06, + "loss": 2.289, + "step": 120 + }, + { + "epoch": 0.13150684931506848, + "eval_loss": 2.365485429763794, + "eval_runtime": 44.495, + "eval_samples_per_second": 22.474, + "eval_steps_per_second": 0.944, + "step": 120 + }, + { + "epoch": 0.1326027397260274, + "grad_norm": 3.5055298805236816, + "learning_rate": 1.934306569343066e-06, + "loss": 2.184, + "step": 121 + }, + { + "epoch": 0.1336986301369863, + "grad_norm": 2.3220107555389404, + "learning_rate": 1.952554744525548e-06, + "loss": 2.2337, + "step": 122 + }, + { + "epoch": 0.13479452054794522, + "grad_norm": 3.1137821674346924, + "learning_rate": 1.9708029197080293e-06, + "loss": 2.2397, + "step": 123 + }, + { + "epoch": 0.1358904109589041, + "grad_norm": 2.8009350299835205, + "learning_rate": 1.989051094890511e-06, + "loss": 2.2347, + "step": 124 + }, + { + "epoch": 0.136986301369863, + "grad_norm": 2.4079606533050537, + "learning_rate": 2.007299270072993e-06, + "loss": 2.2879, + "step": 125 + }, + { + "epoch": 0.13808219178082193, + "grad_norm": 3.0863444805145264, + "learning_rate": 2.025547445255475e-06, + "loss": 2.2222, + "step": 126 + }, + { + "epoch": 0.13917808219178082, + "grad_norm": 2.796104669570923, + "learning_rate": 2.0437956204379563e-06, + "loss": 2.2373, + "step": 127 + }, + { + "epoch": 0.14027397260273972, + "grad_norm": 2.8173458576202393, + "learning_rate": 2.062043795620438e-06, + "loss": 2.2273, + "step": 128 + }, + { + "epoch": 0.14136986301369864, + "grad_norm": 3.3563573360443115, + "learning_rate": 2.08029197080292e-06, + "loss": 2.2864, + "step": 129 + }, + { + "epoch": 0.14246575342465753, + "grad_norm": 2.7632670402526855, + "learning_rate": 2.0985401459854014e-06, + "loss": 2.1808, + "step": 130 + }, + { + "epoch": 0.14246575342465753, + "eval_loss": 2.327371597290039, + "eval_runtime": 44.9288, + "eval_samples_per_second": 22.257, + "eval_steps_per_second": 0.935, + "step": 130 + }, + { + "epoch": 0.14356164383561643, + "grad_norm": 2.587303876876831, + "learning_rate": 2.1167883211678833e-06, + "loss": 2.2439, + "step": 131 + }, + { + "epoch": 0.14465753424657535, + "grad_norm": 2.5574934482574463, + "learning_rate": 2.135036496350365e-06, + "loss": 2.2645, + "step": 132 + }, + { + "epoch": 0.14575342465753424, + "grad_norm": 2.4953293800354004, + "learning_rate": 2.1532846715328466e-06, + "loss": 2.1784, + "step": 133 + }, + { + "epoch": 0.14684931506849316, + "grad_norm": 2.3953003883361816, + "learning_rate": 2.171532846715329e-06, + "loss": 2.1843, + "step": 134 + }, + { + "epoch": 0.14794520547945206, + "grad_norm": 2.4708609580993652, + "learning_rate": 2.1897810218978103e-06, + "loss": 2.2167, + "step": 135 + }, + { + "epoch": 0.14904109589041095, + "grad_norm": 2.6704142093658447, + "learning_rate": 2.208029197080292e-06, + "loss": 2.2024, + "step": 136 + }, + { + "epoch": 0.15013698630136987, + "grad_norm": 2.6712591648101807, + "learning_rate": 2.226277372262774e-06, + "loss": 2.2072, + "step": 137 + }, + { + "epoch": 0.15123287671232877, + "grad_norm": 2.592298984527588, + "learning_rate": 2.244525547445256e-06, + "loss": 2.231, + "step": 138 + }, + { + "epoch": 0.15232876712328766, + "grad_norm": 2.6005120277404785, + "learning_rate": 2.2627737226277373e-06, + "loss": 2.2285, + "step": 139 + }, + { + "epoch": 0.15342465753424658, + "grad_norm": 2.3024377822875977, + "learning_rate": 2.281021897810219e-06, + "loss": 2.1843, + "step": 140 + }, + { + "epoch": 0.15342465753424658, + "eval_loss": 2.3062543869018555, + "eval_runtime": 42.7457, + "eval_samples_per_second": 23.394, + "eval_steps_per_second": 0.983, + "step": 140 + }, + { + "epoch": 0.15452054794520548, + "grad_norm": 2.753288984298706, + "learning_rate": 2.299270072992701e-06, + "loss": 2.1664, + "step": 141 + }, + { + "epoch": 0.15561643835616437, + "grad_norm": 2.9256935119628906, + "learning_rate": 2.3175182481751825e-06, + "loss": 2.1899, + "step": 142 + }, + { + "epoch": 0.1567123287671233, + "grad_norm": 2.7583601474761963, + "learning_rate": 2.3357664233576643e-06, + "loss": 2.154, + "step": 143 + }, + { + "epoch": 0.1578082191780822, + "grad_norm": 2.6788907051086426, + "learning_rate": 2.354014598540146e-06, + "loss": 2.1832, + "step": 144 + }, + { + "epoch": 0.1589041095890411, + "grad_norm": 2.4809155464172363, + "learning_rate": 2.3722627737226276e-06, + "loss": 2.1859, + "step": 145 + }, + { + "epoch": 0.16, + "grad_norm": 2.4358725547790527, + "learning_rate": 2.39051094890511e-06, + "loss": 2.1244, + "step": 146 + }, + { + "epoch": 0.1610958904109589, + "grad_norm": 2.6939032077789307, + "learning_rate": 2.4087591240875918e-06, + "loss": 2.1459, + "step": 147 + }, + { + "epoch": 0.16219178082191782, + "grad_norm": 2.461724281311035, + "learning_rate": 2.427007299270073e-06, + "loss": 2.1592, + "step": 148 + }, + { + "epoch": 0.1632876712328767, + "grad_norm": 2.302741050720215, + "learning_rate": 2.445255474452555e-06, + "loss": 2.2066, + "step": 149 + }, + { + "epoch": 0.1643835616438356, + "grad_norm": 3.0586416721343994, + "learning_rate": 2.463503649635037e-06, + "loss": 2.1104, + "step": 150 + }, + { + "epoch": 0.1643835616438356, + "eval_loss": 2.2823894023895264, + "eval_runtime": 42.6172, + "eval_samples_per_second": 23.465, + "eval_steps_per_second": 0.986, + "step": 150 + }, + { + "epoch": 0.16547945205479453, + "grad_norm": 2.6842763423919678, + "learning_rate": 2.4817518248175183e-06, + "loss": 2.0873, + "step": 151 + }, + { + "epoch": 0.16657534246575342, + "grad_norm": 3.1450300216674805, + "learning_rate": 2.5e-06, + "loss": 2.2092, + "step": 152 + }, + { + "epoch": 0.16767123287671232, + "grad_norm": 2.8436853885650635, + "learning_rate": 2.518248175182482e-06, + "loss": 2.1182, + "step": 153 + }, + { + "epoch": 0.16876712328767124, + "grad_norm": 2.50284481048584, + "learning_rate": 2.5364963503649635e-06, + "loss": 2.1134, + "step": 154 + }, + { + "epoch": 0.16986301369863013, + "grad_norm": 2.6976687908172607, + "learning_rate": 2.5547445255474458e-06, + "loss": 2.169, + "step": 155 + }, + { + "epoch": 0.17095890410958905, + "grad_norm": 2.9173824787139893, + "learning_rate": 2.5729927007299272e-06, + "loss": 2.1423, + "step": 156 + }, + { + "epoch": 0.17205479452054795, + "grad_norm": 2.4598944187164307, + "learning_rate": 2.591240875912409e-06, + "loss": 2.1415, + "step": 157 + }, + { + "epoch": 0.17315068493150684, + "grad_norm": 2.7689433097839355, + "learning_rate": 2.6094890510948905e-06, + "loss": 2.1634, + "step": 158 + }, + { + "epoch": 0.17424657534246576, + "grad_norm": 2.6046409606933594, + "learning_rate": 2.627737226277373e-06, + "loss": 2.1833, + "step": 159 + }, + { + "epoch": 0.17534246575342466, + "grad_norm": 2.8656468391418457, + "learning_rate": 2.6459854014598542e-06, + "loss": 2.141, + "step": 160 + }, + { + "epoch": 0.17534246575342466, + "eval_loss": 2.2605931758880615, + "eval_runtime": 42.862, + "eval_samples_per_second": 23.331, + "eval_steps_per_second": 0.98, + "step": 160 + }, + { + "epoch": 0.17643835616438355, + "grad_norm": 2.7016804218292236, + "learning_rate": 2.664233576642336e-06, + "loss": 2.1138, + "step": 161 + }, + { + "epoch": 0.17753424657534247, + "grad_norm": 2.4924707412719727, + "learning_rate": 2.6824817518248175e-06, + "loss": 2.1417, + "step": 162 + }, + { + "epoch": 0.17863013698630137, + "grad_norm": 2.4874186515808105, + "learning_rate": 2.7007299270072994e-06, + "loss": 2.1156, + "step": 163 + }, + { + "epoch": 0.17972602739726026, + "grad_norm": 2.3887176513671875, + "learning_rate": 2.718978102189781e-06, + "loss": 2.182, + "step": 164 + }, + { + "epoch": 0.18082191780821918, + "grad_norm": 2.5159378051757812, + "learning_rate": 2.737226277372263e-06, + "loss": 2.1501, + "step": 165 + }, + { + "epoch": 0.18191780821917808, + "grad_norm": 2.817784309387207, + "learning_rate": 2.755474452554745e-06, + "loss": 2.1226, + "step": 166 + }, + { + "epoch": 0.18301369863013697, + "grad_norm": 2.3638343811035156, + "learning_rate": 2.7737226277372264e-06, + "loss": 2.2195, + "step": 167 + }, + { + "epoch": 0.1841095890410959, + "grad_norm": 2.598869800567627, + "learning_rate": 2.7919708029197082e-06, + "loss": 2.1645, + "step": 168 + }, + { + "epoch": 0.1852054794520548, + "grad_norm": 2.4598543643951416, + "learning_rate": 2.81021897810219e-06, + "loss": 2.1029, + "step": 169 + }, + { + "epoch": 0.1863013698630137, + "grad_norm": 2.395592212677002, + "learning_rate": 2.828467153284672e-06, + "loss": 2.1369, + "step": 170 + }, + { + "epoch": 0.1863013698630137, + "eval_loss": 2.248337507247925, + "eval_runtime": 42.3651, + "eval_samples_per_second": 23.604, + "eval_steps_per_second": 0.991, + "step": 170 + }, + { + "epoch": 0.1873972602739726, + "grad_norm": 2.527686357498169, + "learning_rate": 2.8467153284671534e-06, + "loss": 2.0586, + "step": 171 + }, + { + "epoch": 0.1884931506849315, + "grad_norm": 2.7667572498321533, + "learning_rate": 2.8649635036496353e-06, + "loss": 2.1043, + "step": 172 + }, + { + "epoch": 0.18958904109589042, + "grad_norm": 2.4474146366119385, + "learning_rate": 2.8832116788321167e-06, + "loss": 2.0626, + "step": 173 + }, + { + "epoch": 0.19068493150684931, + "grad_norm": 2.4016106128692627, + "learning_rate": 2.901459854014599e-06, + "loss": 2.1567, + "step": 174 + }, + { + "epoch": 0.1917808219178082, + "grad_norm": 2.4824960231781006, + "learning_rate": 2.9197080291970804e-06, + "loss": 2.1333, + "step": 175 + }, + { + "epoch": 0.19287671232876713, + "grad_norm": 2.5006752014160156, + "learning_rate": 2.9379562043795623e-06, + "loss": 2.122, + "step": 176 + }, + { + "epoch": 0.19397260273972602, + "grad_norm": 2.4381911754608154, + "learning_rate": 2.956204379562044e-06, + "loss": 2.1613, + "step": 177 + }, + { + "epoch": 0.19506849315068492, + "grad_norm": 2.6550545692443848, + "learning_rate": 2.9744525547445256e-06, + "loss": 2.1248, + "step": 178 + }, + { + "epoch": 0.19616438356164384, + "grad_norm": 2.9135735034942627, + "learning_rate": 2.992700729927008e-06, + "loss": 2.0758, + "step": 179 + }, + { + "epoch": 0.19726027397260273, + "grad_norm": 2.97813081741333, + "learning_rate": 3.0109489051094893e-06, + "loss": 2.065, + "step": 180 + }, + { + "epoch": 0.19726027397260273, + "eval_loss": 2.2347888946533203, + "eval_runtime": 43.4361, + "eval_samples_per_second": 23.022, + "eval_steps_per_second": 0.967, + "step": 180 + }, + { + "epoch": 0.19835616438356166, + "grad_norm": 2.600041627883911, + "learning_rate": 3.029197080291971e-06, + "loss": 2.1524, + "step": 181 + }, + { + "epoch": 0.19945205479452055, + "grad_norm": 2.9430601596832275, + "learning_rate": 3.0474452554744526e-06, + "loss": 2.1449, + "step": 182 + }, + { + "epoch": 0.20054794520547944, + "grad_norm": 2.6751530170440674, + "learning_rate": 3.065693430656935e-06, + "loss": 2.1188, + "step": 183 + }, + { + "epoch": 0.20164383561643837, + "grad_norm": 2.9019253253936768, + "learning_rate": 3.0839416058394163e-06, + "loss": 2.1442, + "step": 184 + }, + { + "epoch": 0.20273972602739726, + "grad_norm": 2.6712942123413086, + "learning_rate": 3.102189781021898e-06, + "loss": 2.1288, + "step": 185 + }, + { + "epoch": 0.20383561643835615, + "grad_norm": 2.909146308898926, + "learning_rate": 3.1204379562043796e-06, + "loss": 2.0614, + "step": 186 + }, + { + "epoch": 0.20493150684931508, + "grad_norm": 3.3119146823883057, + "learning_rate": 3.1386861313868614e-06, + "loss": 2.0403, + "step": 187 + }, + { + "epoch": 0.20602739726027397, + "grad_norm": 3.32538104057312, + "learning_rate": 3.156934306569343e-06, + "loss": 2.1503, + "step": 188 + }, + { + "epoch": 0.20712328767123286, + "grad_norm": 2.8644299507141113, + "learning_rate": 3.175182481751825e-06, + "loss": 2.11, + "step": 189 + }, + { + "epoch": 0.20821917808219179, + "grad_norm": 3.1166255474090576, + "learning_rate": 3.193430656934307e-06, + "loss": 2.0754, + "step": 190 + }, + { + "epoch": 0.20821917808219179, + "eval_loss": 2.2198550701141357, + "eval_runtime": 43.9737, + "eval_samples_per_second": 22.741, + "eval_steps_per_second": 0.955, + "step": 190 + }, + { + "epoch": 0.20931506849315068, + "grad_norm": 2.5856869220733643, + "learning_rate": 3.2116788321167884e-06, + "loss": 2.0987, + "step": 191 + }, + { + "epoch": 0.2104109589041096, + "grad_norm": 3.138582468032837, + "learning_rate": 3.2299270072992707e-06, + "loss": 2.0791, + "step": 192 + }, + { + "epoch": 0.2115068493150685, + "grad_norm": 2.87868332862854, + "learning_rate": 3.248175182481752e-06, + "loss": 2.0104, + "step": 193 + }, + { + "epoch": 0.2126027397260274, + "grad_norm": 2.6997270584106445, + "learning_rate": 3.266423357664234e-06, + "loss": 2.1026, + "step": 194 + }, + { + "epoch": 0.2136986301369863, + "grad_norm": 3.0818371772766113, + "learning_rate": 3.2846715328467155e-06, + "loss": 2.0814, + "step": 195 + }, + { + "epoch": 0.2147945205479452, + "grad_norm": 2.8013179302215576, + "learning_rate": 3.3029197080291973e-06, + "loss": 2.1102, + "step": 196 + }, + { + "epoch": 0.2158904109589041, + "grad_norm": 2.7401671409606934, + "learning_rate": 3.3211678832116788e-06, + "loss": 2.0854, + "step": 197 + }, + { + "epoch": 0.21698630136986302, + "grad_norm": 2.8041036128997803, + "learning_rate": 3.339416058394161e-06, + "loss": 2.0851, + "step": 198 + }, + { + "epoch": 0.21808219178082192, + "grad_norm": 2.416430711746216, + "learning_rate": 3.3576642335766425e-06, + "loss": 2.0862, + "step": 199 + }, + { + "epoch": 0.2191780821917808, + "grad_norm": 2.795642852783203, + "learning_rate": 3.3759124087591243e-06, + "loss": 2.0819, + "step": 200 + }, + { + "epoch": 0.2191780821917808, + "eval_loss": 2.2115843296051025, + "eval_runtime": 42.9778, + "eval_samples_per_second": 23.268, + "eval_steps_per_second": 0.977, + "step": 200 + }, + { + "epoch": 0.22027397260273973, + "grad_norm": 2.9077513217926025, + "learning_rate": 3.3941605839416058e-06, + "loss": 2.0813, + "step": 201 + }, + { + "epoch": 0.22136986301369863, + "grad_norm": 2.910753011703491, + "learning_rate": 3.412408759124088e-06, + "loss": 2.0989, + "step": 202 + }, + { + "epoch": 0.22246575342465755, + "grad_norm": 2.6539418697357178, + "learning_rate": 3.43065693430657e-06, + "loss": 2.0247, + "step": 203 + }, + { + "epoch": 0.22356164383561644, + "grad_norm": 2.9435222148895264, + "learning_rate": 3.4489051094890513e-06, + "loss": 2.0811, + "step": 204 + }, + { + "epoch": 0.22465753424657534, + "grad_norm": 2.725790500640869, + "learning_rate": 3.467153284671533e-06, + "loss": 2.0277, + "step": 205 + }, + { + "epoch": 0.22575342465753426, + "grad_norm": 2.766728639602661, + "learning_rate": 3.4854014598540146e-06, + "loss": 2.0671, + "step": 206 + }, + { + "epoch": 0.22684931506849315, + "grad_norm": 2.546391010284424, + "learning_rate": 3.503649635036497e-06, + "loss": 2.0701, + "step": 207 + }, + { + "epoch": 0.22794520547945205, + "grad_norm": 2.5030412673950195, + "learning_rate": 3.5218978102189783e-06, + "loss": 2.0855, + "step": 208 + }, + { + "epoch": 0.22904109589041097, + "grad_norm": 2.6629745960235596, + "learning_rate": 3.54014598540146e-06, + "loss": 2.1114, + "step": 209 + }, + { + "epoch": 0.23013698630136986, + "grad_norm": 2.56917142868042, + "learning_rate": 3.5583941605839416e-06, + "loss": 2.1597, + "step": 210 + }, + { + "epoch": 0.23013698630136986, + "eval_loss": 2.199700355529785, + "eval_runtime": 44.0819, + "eval_samples_per_second": 22.685, + "eval_steps_per_second": 0.953, + "step": 210 + }, + { + "epoch": 0.23123287671232876, + "grad_norm": 2.4856061935424805, + "learning_rate": 3.576642335766424e-06, + "loss": 2.1343, + "step": 211 + }, + { + "epoch": 0.23232876712328768, + "grad_norm": 2.420940399169922, + "learning_rate": 3.5948905109489054e-06, + "loss": 1.9787, + "step": 212 + }, + { + "epoch": 0.23342465753424657, + "grad_norm": 2.573434591293335, + "learning_rate": 3.6131386861313872e-06, + "loss": 1.9825, + "step": 213 + }, + { + "epoch": 0.2345205479452055, + "grad_norm": 2.6574819087982178, + "learning_rate": 3.6313868613138687e-06, + "loss": 2.0239, + "step": 214 + }, + { + "epoch": 0.2356164383561644, + "grad_norm": 2.426731586456299, + "learning_rate": 3.6496350364963505e-06, + "loss": 2.039, + "step": 215 + }, + { + "epoch": 0.23671232876712328, + "grad_norm": 2.8172669410705566, + "learning_rate": 3.667883211678833e-06, + "loss": 2.0111, + "step": 216 + }, + { + "epoch": 0.2378082191780822, + "grad_norm": 2.6581475734710693, + "learning_rate": 3.6861313868613142e-06, + "loss": 2.0125, + "step": 217 + }, + { + "epoch": 0.2389041095890411, + "grad_norm": 2.832300901412964, + "learning_rate": 3.704379562043796e-06, + "loss": 2.1067, + "step": 218 + }, + { + "epoch": 0.24, + "grad_norm": 2.9996016025543213, + "learning_rate": 3.7226277372262775e-06, + "loss": 2.0717, + "step": 219 + }, + { + "epoch": 0.2410958904109589, + "grad_norm": 2.8467094898223877, + "learning_rate": 3.7408759124087594e-06, + "loss": 2.0117, + "step": 220 + }, + { + "epoch": 0.2410958904109589, + "eval_loss": 2.188706159591675, + "eval_runtime": 44.3101, + "eval_samples_per_second": 22.568, + "eval_steps_per_second": 0.948, + "step": 220 + }, + { + "epoch": 0.2421917808219178, + "grad_norm": 2.6468069553375244, + "learning_rate": 3.7591240875912412e-06, + "loss": 2.048, + "step": 221 + }, + { + "epoch": 0.2432876712328767, + "grad_norm": 2.697842836380005, + "learning_rate": 3.777372262773723e-06, + "loss": 2.0408, + "step": 222 + }, + { + "epoch": 0.24438356164383562, + "grad_norm": 2.5084266662597656, + "learning_rate": 3.7956204379562045e-06, + "loss": 2.0162, + "step": 223 + }, + { + "epoch": 0.24547945205479452, + "grad_norm": 2.4894039630889893, + "learning_rate": 3.8138686131386864e-06, + "loss": 2.0722, + "step": 224 + }, + { + "epoch": 0.2465753424657534, + "grad_norm": 2.6232657432556152, + "learning_rate": 3.832116788321168e-06, + "loss": 2.0214, + "step": 225 + }, + { + "epoch": 0.24767123287671233, + "grad_norm": 2.4213356971740723, + "learning_rate": 3.85036496350365e-06, + "loss": 2.0006, + "step": 226 + }, + { + "epoch": 0.24876712328767123, + "grad_norm": 2.605342388153076, + "learning_rate": 3.868613138686132e-06, + "loss": 2.0336, + "step": 227 + }, + { + "epoch": 0.24986301369863015, + "grad_norm": 2.6153266429901123, + "learning_rate": 3.886861313868613e-06, + "loss": 2.0501, + "step": 228 + }, + { + "epoch": 0.25095890410958904, + "grad_norm": 2.4511303901672363, + "learning_rate": 3.905109489051096e-06, + "loss": 2.0763, + "step": 229 + }, + { + "epoch": 0.25205479452054796, + "grad_norm": 2.833284854888916, + "learning_rate": 3.923357664233577e-06, + "loss": 2.0194, + "step": 230 + }, + { + "epoch": 0.25205479452054796, + "eval_loss": 2.1737821102142334, + "eval_runtime": 43.7157, + "eval_samples_per_second": 22.875, + "eval_steps_per_second": 0.961, + "step": 230 + }, + { + "epoch": 0.25315068493150683, + "grad_norm": 2.7150166034698486, + "learning_rate": 3.9416058394160585e-06, + "loss": 2.0892, + "step": 231 + }, + { + "epoch": 0.25424657534246575, + "grad_norm": 2.6422505378723145, + "learning_rate": 3.95985401459854e-06, + "loss": 2.1025, + "step": 232 + }, + { + "epoch": 0.2553424657534247, + "grad_norm": 2.8351335525512695, + "learning_rate": 3.978102189781022e-06, + "loss": 1.967, + "step": 233 + }, + { + "epoch": 0.25643835616438354, + "grad_norm": 2.4947831630706787, + "learning_rate": 3.996350364963504e-06, + "loss": 2.0541, + "step": 234 + }, + { + "epoch": 0.25753424657534246, + "grad_norm": 2.6740469932556152, + "learning_rate": 4.014598540145986e-06, + "loss": 1.958, + "step": 235 + }, + { + "epoch": 0.2586301369863014, + "grad_norm": 3.113481044769287, + "learning_rate": 4.032846715328467e-06, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.25972602739726025, + "grad_norm": 2.5094399452209473, + "learning_rate": 4.05109489051095e-06, + "loss": 2.0304, + "step": 237 + }, + { + "epoch": 0.2608219178082192, + "grad_norm": 2.5995097160339355, + "learning_rate": 4.069343065693431e-06, + "loss": 1.9859, + "step": 238 + }, + { + "epoch": 0.2619178082191781, + "grad_norm": 2.66732120513916, + "learning_rate": 4.0875912408759126e-06, + "loss": 2.0221, + "step": 239 + }, + { + "epoch": 0.26301369863013696, + "grad_norm": 2.6683313846588135, + "learning_rate": 4.105839416058395e-06, + "loss": 2.0873, + "step": 240 + }, + { + "epoch": 0.26301369863013696, + "eval_loss": 2.1653976440429688, + "eval_runtime": 44.8246, + "eval_samples_per_second": 22.309, + "eval_steps_per_second": 0.937, + "step": 240 + }, + { + "epoch": 0.2641095890410959, + "grad_norm": 2.83756947517395, + "learning_rate": 4.124087591240876e-06, + "loss": 2.0071, + "step": 241 + }, + { + "epoch": 0.2652054794520548, + "grad_norm": 2.5801031589508057, + "learning_rate": 4.1423357664233586e-06, + "loss": 2.003, + "step": 242 + }, + { + "epoch": 0.2663013698630137, + "grad_norm": 2.5858423709869385, + "learning_rate": 4.16058394160584e-06, + "loss": 2.0728, + "step": 243 + }, + { + "epoch": 0.2673972602739726, + "grad_norm": 3.0055713653564453, + "learning_rate": 4.1788321167883214e-06, + "loss": 1.9568, + "step": 244 + }, + { + "epoch": 0.2684931506849315, + "grad_norm": 2.814678430557251, + "learning_rate": 4.197080291970803e-06, + "loss": 2.0324, + "step": 245 + }, + { + "epoch": 0.26958904109589044, + "grad_norm": 2.947052001953125, + "learning_rate": 4.215328467153285e-06, + "loss": 1.9935, + "step": 246 + }, + { + "epoch": 0.2706849315068493, + "grad_norm": 2.638324022293091, + "learning_rate": 4.233576642335767e-06, + "loss": 1.9916, + "step": 247 + }, + { + "epoch": 0.2717808219178082, + "grad_norm": 2.8386952877044678, + "learning_rate": 4.251824817518249e-06, + "loss": 2.0158, + "step": 248 + }, + { + "epoch": 0.27287671232876715, + "grad_norm": 2.7214903831481934, + "learning_rate": 4.27007299270073e-06, + "loss": 1.9832, + "step": 249 + }, + { + "epoch": 0.273972602739726, + "grad_norm": 2.873758316040039, + "learning_rate": 4.288321167883212e-06, + "loss": 1.991, + "step": 250 + }, + { + "epoch": 0.273972602739726, + "eval_loss": 2.1588690280914307, + "eval_runtime": 46.0357, + "eval_samples_per_second": 21.722, + "eval_steps_per_second": 0.912, + "step": 250 + }, + { + "epoch": 0.27506849315068493, + "grad_norm": 2.647588014602661, + "learning_rate": 4.306569343065693e-06, + "loss": 1.9542, + "step": 251 + }, + { + "epoch": 0.27616438356164386, + "grad_norm": 2.7150375843048096, + "learning_rate": 4.3248175182481755e-06, + "loss": 2.017, + "step": 252 + }, + { + "epoch": 0.2772602739726027, + "grad_norm": 2.4526689052581787, + "learning_rate": 4.343065693430658e-06, + "loss": 2.0051, + "step": 253 + }, + { + "epoch": 0.27835616438356164, + "grad_norm": 2.770855665206909, + "learning_rate": 4.361313868613139e-06, + "loss": 1.9931, + "step": 254 + }, + { + "epoch": 0.27945205479452057, + "grad_norm": 2.618800163269043, + "learning_rate": 4.379562043795621e-06, + "loss": 2.0347, + "step": 255 + }, + { + "epoch": 0.28054794520547943, + "grad_norm": 3.069519281387329, + "learning_rate": 4.397810218978102e-06, + "loss": 2.0189, + "step": 256 + }, + { + "epoch": 0.28164383561643835, + "grad_norm": 2.486750602722168, + "learning_rate": 4.416058394160584e-06, + "loss": 1.9967, + "step": 257 + }, + { + "epoch": 0.2827397260273973, + "grad_norm": 2.9060416221618652, + "learning_rate": 4.434306569343066e-06, + "loss": 1.9775, + "step": 258 + }, + { + "epoch": 0.28383561643835614, + "grad_norm": 2.878208637237549, + "learning_rate": 4.452554744525548e-06, + "loss": 2.0221, + "step": 259 + }, + { + "epoch": 0.28493150684931506, + "grad_norm": 2.448805809020996, + "learning_rate": 4.4708029197080295e-06, + "loss": 2.0133, + "step": 260 + }, + { + "epoch": 0.28493150684931506, + "eval_loss": 2.147843837738037, + "eval_runtime": 42.5243, + "eval_samples_per_second": 23.516, + "eval_steps_per_second": 0.988, + "step": 260 + }, + { + "epoch": 0.286027397260274, + "grad_norm": 2.89329195022583, + "learning_rate": 4.489051094890512e-06, + "loss": 2.0496, + "step": 261 + }, + { + "epoch": 0.28712328767123285, + "grad_norm": 3.102792263031006, + "learning_rate": 4.507299270072993e-06, + "loss": 2.01, + "step": 262 + }, + { + "epoch": 0.2882191780821918, + "grad_norm": 2.3834943771362305, + "learning_rate": 4.525547445255475e-06, + "loss": 2.0022, + "step": 263 + }, + { + "epoch": 0.2893150684931507, + "grad_norm": 2.972414493560791, + "learning_rate": 4.543795620437956e-06, + "loss": 1.9495, + "step": 264 + }, + { + "epoch": 0.29041095890410956, + "grad_norm": 2.6329052448272705, + "learning_rate": 4.562043795620438e-06, + "loss": 2.0506, + "step": 265 + }, + { + "epoch": 0.2915068493150685, + "grad_norm": 2.9160213470458984, + "learning_rate": 4.580291970802921e-06, + "loss": 2.0555, + "step": 266 + }, + { + "epoch": 0.2926027397260274, + "grad_norm": 2.4848756790161133, + "learning_rate": 4.598540145985402e-06, + "loss": 2.008, + "step": 267 + }, + { + "epoch": 0.2936986301369863, + "grad_norm": 2.743487596511841, + "learning_rate": 4.6167883211678835e-06, + "loss": 1.9827, + "step": 268 + }, + { + "epoch": 0.2947945205479452, + "grad_norm": 2.4862096309661865, + "learning_rate": 4.635036496350365e-06, + "loss": 1.9724, + "step": 269 + }, + { + "epoch": 0.2958904109589041, + "grad_norm": 2.4750123023986816, + "learning_rate": 4.653284671532847e-06, + "loss": 2.0203, + "step": 270 + }, + { + "epoch": 0.2958904109589041, + "eval_loss": 2.1395180225372314, + "eval_runtime": 44.0335, + "eval_samples_per_second": 22.71, + "eval_steps_per_second": 0.954, + "step": 270 + }, + { + "epoch": 0.29698630136986304, + "grad_norm": 2.7088334560394287, + "learning_rate": 4.671532846715329e-06, + "loss": 2.0242, + "step": 271 + }, + { + "epoch": 0.2980821917808219, + "grad_norm": 2.636436700820923, + "learning_rate": 4.689781021897811e-06, + "loss": 1.9917, + "step": 272 + }, + { + "epoch": 0.2991780821917808, + "grad_norm": 2.508549451828003, + "learning_rate": 4.708029197080292e-06, + "loss": 2.04, + "step": 273 + }, + { + "epoch": 0.30027397260273975, + "grad_norm": 2.850067615509033, + "learning_rate": 4.726277372262774e-06, + "loss": 2.0139, + "step": 274 + }, + { + "epoch": 0.3013698630136986, + "grad_norm": 2.5066423416137695, + "learning_rate": 4.744525547445255e-06, + "loss": 1.9446, + "step": 275 + }, + { + "epoch": 0.30246575342465754, + "grad_norm": 2.587402105331421, + "learning_rate": 4.7627737226277375e-06, + "loss": 2.0131, + "step": 276 + }, + { + "epoch": 0.30356164383561646, + "grad_norm": 2.87727427482605, + "learning_rate": 4.78102189781022e-06, + "loss": 2.0081, + "step": 277 + }, + { + "epoch": 0.3046575342465753, + "grad_norm": 2.5349059104919434, + "learning_rate": 4.799270072992701e-06, + "loss": 1.9645, + "step": 278 + }, + { + "epoch": 0.30575342465753425, + "grad_norm": 2.42708420753479, + "learning_rate": 4.8175182481751835e-06, + "loss": 1.9908, + "step": 279 + }, + { + "epoch": 0.30684931506849317, + "grad_norm": 2.590723991394043, + "learning_rate": 4.835766423357665e-06, + "loss": 1.9896, + "step": 280 + }, + { + "epoch": 0.30684931506849317, + "eval_loss": 2.1315386295318604, + "eval_runtime": 43.1343, + "eval_samples_per_second": 23.183, + "eval_steps_per_second": 0.974, + "step": 280 + }, + { + "epoch": 0.30794520547945203, + "grad_norm": 2.459399461746216, + "learning_rate": 4.854014598540146e-06, + "loss": 1.9304, + "step": 281 + }, + { + "epoch": 0.30904109589041096, + "grad_norm": 2.6509106159210205, + "learning_rate": 4.872262773722628e-06, + "loss": 1.9486, + "step": 282 + }, + { + "epoch": 0.3101369863013699, + "grad_norm": 2.6039066314697266, + "learning_rate": 4.89051094890511e-06, + "loss": 1.9495, + "step": 283 + }, + { + "epoch": 0.31123287671232874, + "grad_norm": 2.7798705101013184, + "learning_rate": 4.9087591240875915e-06, + "loss": 1.9665, + "step": 284 + }, + { + "epoch": 0.31232876712328766, + "grad_norm": 2.664199113845825, + "learning_rate": 4.927007299270074e-06, + "loss": 1.9911, + "step": 285 + }, + { + "epoch": 0.3134246575342466, + "grad_norm": 2.562411308288574, + "learning_rate": 4.945255474452555e-06, + "loss": 1.9501, + "step": 286 + }, + { + "epoch": 0.31452054794520545, + "grad_norm": 2.2824630737304688, + "learning_rate": 4.963503649635037e-06, + "loss": 2.045, + "step": 287 + }, + { + "epoch": 0.3156164383561644, + "grad_norm": 2.917419910430908, + "learning_rate": 4.981751824817518e-06, + "loss": 2.0304, + "step": 288 + }, + { + "epoch": 0.3167123287671233, + "grad_norm": 2.5161173343658447, + "learning_rate": 5e-06, + "loss": 1.9813, + "step": 289 + }, + { + "epoch": 0.3178082191780822, + "grad_norm": 2.579141139984131, + "learning_rate": 4.997969130787978e-06, + "loss": 2.0006, + "step": 290 + }, + { + "epoch": 0.3178082191780822, + "eval_loss": 2.11863112449646, + "eval_runtime": 43.3566, + "eval_samples_per_second": 23.065, + "eval_steps_per_second": 0.969, + "step": 290 + }, + { + "epoch": 0.3189041095890411, + "grad_norm": 2.4404804706573486, + "learning_rate": 4.995938261575955e-06, + "loss": 1.9112, + "step": 291 + }, + { + "epoch": 0.32, + "grad_norm": 2.40389347076416, + "learning_rate": 4.993907392363932e-06, + "loss": 1.9997, + "step": 292 + }, + { + "epoch": 0.32109589041095893, + "grad_norm": 2.5704402923583984, + "learning_rate": 4.9918765231519095e-06, + "loss": 1.9296, + "step": 293 + }, + { + "epoch": 0.3221917808219178, + "grad_norm": 2.9520113468170166, + "learning_rate": 4.989845653939887e-06, + "loss": 1.9505, + "step": 294 + }, + { + "epoch": 0.3232876712328767, + "grad_norm": 2.968468189239502, + "learning_rate": 4.987814784727864e-06, + "loss": 1.9609, + "step": 295 + }, + { + "epoch": 0.32438356164383564, + "grad_norm": 2.600876569747925, + "learning_rate": 4.985783915515841e-06, + "loss": 1.9428, + "step": 296 + }, + { + "epoch": 0.3254794520547945, + "grad_norm": 2.680389165878296, + "learning_rate": 4.983753046303819e-06, + "loss": 1.9668, + "step": 297 + }, + { + "epoch": 0.3265753424657534, + "grad_norm": 2.400132179260254, + "learning_rate": 4.981722177091796e-06, + "loss": 1.9465, + "step": 298 + }, + { + "epoch": 0.32767123287671235, + "grad_norm": 2.792793035507202, + "learning_rate": 4.979691307879773e-06, + "loss": 1.9883, + "step": 299 + }, + { + "epoch": 0.3287671232876712, + "grad_norm": 2.5185678005218506, + "learning_rate": 4.9776604386677504e-06, + "loss": 1.9648, + "step": 300 + }, + { + "epoch": 0.3287671232876712, + "eval_loss": 2.1139590740203857, + "eval_runtime": 42.6627, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 300 + }, + { + "epoch": 0.32986301369863014, + "grad_norm": 2.424886703491211, + "learning_rate": 4.975629569455728e-06, + "loss": 2.0073, + "step": 301 + }, + { + "epoch": 0.33095890410958906, + "grad_norm": 2.735701084136963, + "learning_rate": 4.973598700243705e-06, + "loss": 1.928, + "step": 302 + }, + { + "epoch": 0.3320547945205479, + "grad_norm": 2.6131203174591064, + "learning_rate": 4.971567831031681e-06, + "loss": 1.9367, + "step": 303 + }, + { + "epoch": 0.33315068493150685, + "grad_norm": 2.5688297748565674, + "learning_rate": 4.9695369618196595e-06, + "loss": 1.821, + "step": 304 + }, + { + "epoch": 0.33424657534246577, + "grad_norm": 2.7849748134613037, + "learning_rate": 4.967506092607636e-06, + "loss": 1.9135, + "step": 305 + }, + { + "epoch": 0.33534246575342463, + "grad_norm": 2.7912747859954834, + "learning_rate": 4.965475223395614e-06, + "loss": 1.9317, + "step": 306 + }, + { + "epoch": 0.33643835616438356, + "grad_norm": 2.519549608230591, + "learning_rate": 4.9634443541835905e-06, + "loss": 1.9242, + "step": 307 + }, + { + "epoch": 0.3375342465753425, + "grad_norm": 2.721736431121826, + "learning_rate": 4.961413484971569e-06, + "loss": 1.9648, + "step": 308 + }, + { + "epoch": 0.33863013698630134, + "grad_norm": 2.818453550338745, + "learning_rate": 4.959382615759545e-06, + "loss": 1.9478, + "step": 309 + }, + { + "epoch": 0.33972602739726027, + "grad_norm": 2.5729873180389404, + "learning_rate": 4.957351746547523e-06, + "loss": 1.9208, + "step": 310 + }, + { + "epoch": 0.33972602739726027, + "eval_loss": 2.1217634677886963, + "eval_runtime": 42.6616, + "eval_samples_per_second": 23.44, + "eval_steps_per_second": 0.984, + "step": 310 + }, + { + "epoch": 0.3408219178082192, + "grad_norm": 2.9972047805786133, + "learning_rate": 4.9553208773355e-06, + "loss": 2.0687, + "step": 311 + }, + { + "epoch": 0.3419178082191781, + "grad_norm": 2.563190221786499, + "learning_rate": 4.953290008123478e-06, + "loss": 1.9544, + "step": 312 + }, + { + "epoch": 0.343013698630137, + "grad_norm": 2.7203011512756348, + "learning_rate": 4.951259138911454e-06, + "loss": 1.8932, + "step": 313 + }, + { + "epoch": 0.3441095890410959, + "grad_norm": 2.5524137020111084, + "learning_rate": 4.9492282696994315e-06, + "loss": 2.0232, + "step": 314 + }, + { + "epoch": 0.3452054794520548, + "grad_norm": 2.728055715560913, + "learning_rate": 4.947197400487409e-06, + "loss": 1.9536, + "step": 315 + }, + { + "epoch": 0.3463013698630137, + "grad_norm": 3.1736466884613037, + "learning_rate": 4.945166531275386e-06, + "loss": 1.9286, + "step": 316 + }, + { + "epoch": 0.3473972602739726, + "grad_norm": 2.6353843212127686, + "learning_rate": 4.943135662063363e-06, + "loss": 1.944, + "step": 317 + }, + { + "epoch": 0.34849315068493153, + "grad_norm": 2.883052110671997, + "learning_rate": 4.9411047928513406e-06, + "loss": 1.9898, + "step": 318 + }, + { + "epoch": 0.3495890410958904, + "grad_norm": 2.9240245819091797, + "learning_rate": 4.939073923639318e-06, + "loss": 1.9865, + "step": 319 + }, + { + "epoch": 0.3506849315068493, + "grad_norm": 2.8139750957489014, + "learning_rate": 4.937043054427295e-06, + "loss": 1.917, + "step": 320 + }, + { + "epoch": 0.3506849315068493, + "eval_loss": 2.109309196472168, + "eval_runtime": 45.2892, + "eval_samples_per_second": 22.08, + "eval_steps_per_second": 0.927, + "step": 320 + }, + { + "epoch": 0.35178082191780824, + "grad_norm": 3.24280047416687, + "learning_rate": 4.935012185215272e-06, + "loss": 1.9095, + "step": 321 + }, + { + "epoch": 0.3528767123287671, + "grad_norm": 2.430203437805176, + "learning_rate": 4.93298131600325e-06, + "loss": 1.9637, + "step": 322 + }, + { + "epoch": 0.353972602739726, + "grad_norm": 2.7171425819396973, + "learning_rate": 4.930950446791227e-06, + "loss": 1.9466, + "step": 323 + }, + { + "epoch": 0.35506849315068495, + "grad_norm": 2.8162405490875244, + "learning_rate": 4.928919577579204e-06, + "loss": 1.9607, + "step": 324 + }, + { + "epoch": 0.3561643835616438, + "grad_norm": 2.696545362472534, + "learning_rate": 4.9268887083671815e-06, + "loss": 1.9337, + "step": 325 + }, + { + "epoch": 0.35726027397260274, + "grad_norm": 2.6707708835601807, + "learning_rate": 4.924857839155159e-06, + "loss": 1.9331, + "step": 326 + }, + { + "epoch": 0.35835616438356166, + "grad_norm": 2.6986441612243652, + "learning_rate": 4.922826969943136e-06, + "loss": 1.9537, + "step": 327 + }, + { + "epoch": 0.3594520547945205, + "grad_norm": 2.650552272796631, + "learning_rate": 4.920796100731113e-06, + "loss": 1.8721, + "step": 328 + }, + { + "epoch": 0.36054794520547945, + "grad_norm": 2.3543972969055176, + "learning_rate": 4.918765231519091e-06, + "loss": 1.9413, + "step": 329 + }, + { + "epoch": 0.36164383561643837, + "grad_norm": 2.9600696563720703, + "learning_rate": 4.916734362307068e-06, + "loss": 1.8884, + "step": 330 + }, + { + "epoch": 0.36164383561643837, + "eval_loss": 2.098888874053955, + "eval_runtime": 45.7103, + "eval_samples_per_second": 21.877, + "eval_steps_per_second": 0.919, + "step": 330 + }, + { + "epoch": 0.36273972602739724, + "grad_norm": 2.371508836746216, + "learning_rate": 4.914703493095045e-06, + "loss": 1.8948, + "step": 331 + }, + { + "epoch": 0.36383561643835616, + "grad_norm": 2.393888473510742, + "learning_rate": 4.912672623883022e-06, + "loss": 1.8886, + "step": 332 + }, + { + "epoch": 0.3649315068493151, + "grad_norm": 2.9001216888427734, + "learning_rate": 4.910641754671e-06, + "loss": 1.9235, + "step": 333 + }, + { + "epoch": 0.36602739726027395, + "grad_norm": 2.282672166824341, + "learning_rate": 4.908610885458977e-06, + "loss": 1.9157, + "step": 334 + }, + { + "epoch": 0.36712328767123287, + "grad_norm": 2.6463375091552734, + "learning_rate": 4.906580016246954e-06, + "loss": 1.9656, + "step": 335 + }, + { + "epoch": 0.3682191780821918, + "grad_norm": 2.5853428840637207, + "learning_rate": 4.9045491470349315e-06, + "loss": 1.9431, + "step": 336 + }, + { + "epoch": 0.3693150684931507, + "grad_norm": 2.53450870513916, + "learning_rate": 4.902518277822909e-06, + "loss": 1.9451, + "step": 337 + }, + { + "epoch": 0.3704109589041096, + "grad_norm": 2.6467676162719727, + "learning_rate": 4.900487408610886e-06, + "loss": 1.9503, + "step": 338 + }, + { + "epoch": 0.3715068493150685, + "grad_norm": 2.579293727874756, + "learning_rate": 4.898456539398863e-06, + "loss": 1.9637, + "step": 339 + }, + { + "epoch": 0.3726027397260274, + "grad_norm": 2.525853157043457, + "learning_rate": 4.89642567018684e-06, + "loss": 2.0277, + "step": 340 + }, + { + "epoch": 0.3726027397260274, + "eval_loss": 2.0921695232391357, + "eval_runtime": 44.072, + "eval_samples_per_second": 22.69, + "eval_steps_per_second": 0.953, + "step": 340 + }, + { + "epoch": 0.3736986301369863, + "grad_norm": 2.587911367416382, + "learning_rate": 4.894394800974818e-06, + "loss": 1.8933, + "step": 341 + }, + { + "epoch": 0.3747945205479452, + "grad_norm": 2.576068878173828, + "learning_rate": 4.892363931762794e-06, + "loss": 1.9671, + "step": 342 + }, + { + "epoch": 0.37589041095890413, + "grad_norm": 2.5170862674713135, + "learning_rate": 4.8903330625507725e-06, + "loss": 1.8972, + "step": 343 + }, + { + "epoch": 0.376986301369863, + "grad_norm": 2.6590752601623535, + "learning_rate": 4.888302193338749e-06, + "loss": 1.8702, + "step": 344 + }, + { + "epoch": 0.3780821917808219, + "grad_norm": 2.395724296569824, + "learning_rate": 4.886271324126727e-06, + "loss": 1.9806, + "step": 345 + }, + { + "epoch": 0.37917808219178084, + "grad_norm": 2.3438048362731934, + "learning_rate": 4.8842404549147034e-06, + "loss": 1.9556, + "step": 346 + }, + { + "epoch": 0.3802739726027397, + "grad_norm": 2.326134204864502, + "learning_rate": 4.8822095857026816e-06, + "loss": 1.9184, + "step": 347 + }, + { + "epoch": 0.38136986301369863, + "grad_norm": 2.5018043518066406, + "learning_rate": 4.880178716490658e-06, + "loss": 1.87, + "step": 348 + }, + { + "epoch": 0.38246575342465755, + "grad_norm": 2.487388849258423, + "learning_rate": 4.878147847278636e-06, + "loss": 1.8899, + "step": 349 + }, + { + "epoch": 0.3835616438356164, + "grad_norm": 2.30224347114563, + "learning_rate": 4.8761169780666125e-06, + "loss": 1.8625, + "step": 350 + }, + { + "epoch": 0.3835616438356164, + "eval_loss": 2.0868148803710938, + "eval_runtime": 45.2075, + "eval_samples_per_second": 22.12, + "eval_steps_per_second": 0.929, + "step": 350 + }, + { + "epoch": 0.38465753424657534, + "grad_norm": 2.497983694076538, + "learning_rate": 4.874086108854591e-06, + "loss": 1.8792, + "step": 351 + }, + { + "epoch": 0.38575342465753426, + "grad_norm": 2.3827388286590576, + "learning_rate": 4.872055239642567e-06, + "loss": 1.8522, + "step": 352 + }, + { + "epoch": 0.3868493150684931, + "grad_norm": 2.450486660003662, + "learning_rate": 4.870024370430544e-06, + "loss": 1.9169, + "step": 353 + }, + { + "epoch": 0.38794520547945205, + "grad_norm": 2.4582815170288086, + "learning_rate": 4.867993501218522e-06, + "loss": 1.9417, + "step": 354 + }, + { + "epoch": 0.38904109589041097, + "grad_norm": 2.383026123046875, + "learning_rate": 4.865962632006499e-06, + "loss": 1.8785, + "step": 355 + }, + { + "epoch": 0.39013698630136984, + "grad_norm": 2.528388261795044, + "learning_rate": 4.863931762794476e-06, + "loss": 1.9112, + "step": 356 + }, + { + "epoch": 0.39123287671232876, + "grad_norm": 2.602412462234497, + "learning_rate": 4.8619008935824535e-06, + "loss": 1.9408, + "step": 357 + }, + { + "epoch": 0.3923287671232877, + "grad_norm": 2.9166955947875977, + "learning_rate": 4.859870024370431e-06, + "loss": 1.9523, + "step": 358 + }, + { + "epoch": 0.3934246575342466, + "grad_norm": 2.437335252761841, + "learning_rate": 4.857839155158408e-06, + "loss": 1.9649, + "step": 359 + }, + { + "epoch": 0.39452054794520547, + "grad_norm": 2.6596179008483887, + "learning_rate": 4.855808285946385e-06, + "loss": 1.9227, + "step": 360 + }, + { + "epoch": 0.39452054794520547, + "eval_loss": 2.0740489959716797, + "eval_runtime": 49.1408, + "eval_samples_per_second": 20.35, + "eval_steps_per_second": 0.855, + "step": 360 + }, + { + "epoch": 0.3956164383561644, + "grad_norm": 2.3270347118377686, + "learning_rate": 4.8537774167343626e-06, + "loss": 1.9159, + "step": 361 + }, + { + "epoch": 0.3967123287671233, + "grad_norm": 2.531248092651367, + "learning_rate": 4.85174654752234e-06, + "loss": 1.9132, + "step": 362 + }, + { + "epoch": 0.3978082191780822, + "grad_norm": 2.6526854038238525, + "learning_rate": 4.849715678310317e-06, + "loss": 1.9169, + "step": 363 + }, + { + "epoch": 0.3989041095890411, + "grad_norm": 2.399012565612793, + "learning_rate": 4.847684809098294e-06, + "loss": 1.903, + "step": 364 + }, + { + "epoch": 0.4, + "grad_norm": 3.561117649078369, + "learning_rate": 4.845653939886272e-06, + "loss": 1.921, + "step": 365 + }, + { + "epoch": 0.4010958904109589, + "grad_norm": 2.383044481277466, + "learning_rate": 4.843623070674249e-06, + "loss": 1.9079, + "step": 366 + }, + { + "epoch": 0.4021917808219178, + "grad_norm": 3.1530423164367676, + "learning_rate": 4.841592201462226e-06, + "loss": 1.9511, + "step": 367 + }, + { + "epoch": 0.40328767123287673, + "grad_norm": 2.2790896892547607, + "learning_rate": 4.8395613322502035e-06, + "loss": 1.8534, + "step": 368 + }, + { + "epoch": 0.4043835616438356, + "grad_norm": 3.524251937866211, + "learning_rate": 4.837530463038181e-06, + "loss": 1.8941, + "step": 369 + }, + { + "epoch": 0.4054794520547945, + "grad_norm": 2.481882333755493, + "learning_rate": 4.835499593826158e-06, + "loss": 1.9341, + "step": 370 + }, + { + "epoch": 0.4054794520547945, + "eval_loss": 2.0755298137664795, + "eval_runtime": 47.3237, + "eval_samples_per_second": 21.131, + "eval_steps_per_second": 0.888, + "step": 370 + }, + { + "epoch": 0.40657534246575344, + "grad_norm": 3.404475212097168, + "learning_rate": 4.833468724614135e-06, + "loss": 1.9039, + "step": 371 + }, + { + "epoch": 0.4076712328767123, + "grad_norm": 2.4754388332366943, + "learning_rate": 4.831437855402113e-06, + "loss": 1.8045, + "step": 372 + }, + { + "epoch": 0.40876712328767123, + "grad_norm": 2.8027169704437256, + "learning_rate": 4.82940698619009e-06, + "loss": 1.8906, + "step": 373 + }, + { + "epoch": 0.40986301369863015, + "grad_norm": 3.158459424972534, + "learning_rate": 4.827376116978067e-06, + "loss": 1.9376, + "step": 374 + }, + { + "epoch": 0.410958904109589, + "grad_norm": 2.2808542251586914, + "learning_rate": 4.8253452477660444e-06, + "loss": 1.9159, + "step": 375 + }, + { + "epoch": 0.41205479452054794, + "grad_norm": 3.2421393394470215, + "learning_rate": 4.823314378554022e-06, + "loss": 1.9201, + "step": 376 + }, + { + "epoch": 0.41315068493150686, + "grad_norm": 2.7290635108947754, + "learning_rate": 4.821283509341998e-06, + "loss": 1.8812, + "step": 377 + }, + { + "epoch": 0.41424657534246573, + "grad_norm": 2.661762237548828, + "learning_rate": 4.819252640129976e-06, + "loss": 1.9024, + "step": 378 + }, + { + "epoch": 0.41534246575342465, + "grad_norm": 3.1267426013946533, + "learning_rate": 4.817221770917953e-06, + "loss": 1.8726, + "step": 379 + }, + { + "epoch": 0.41643835616438357, + "grad_norm": 2.4203412532806396, + "learning_rate": 4.815190901705931e-06, + "loss": 1.9071, + "step": 380 + }, + { + "epoch": 0.41643835616438357, + "eval_loss": 2.0662670135498047, + "eval_runtime": 44.15, + "eval_samples_per_second": 22.65, + "eval_steps_per_second": 0.951, + "step": 380 + }, + { + "epoch": 0.41753424657534244, + "grad_norm": 3.370844602584839, + "learning_rate": 4.813160032493907e-06, + "loss": 1.8899, + "step": 381 + }, + { + "epoch": 0.41863013698630136, + "grad_norm": 2.273742437362671, + "learning_rate": 4.811129163281885e-06, + "loss": 1.9222, + "step": 382 + }, + { + "epoch": 0.4197260273972603, + "grad_norm": 2.5339367389678955, + "learning_rate": 4.809098294069862e-06, + "loss": 1.9098, + "step": 383 + }, + { + "epoch": 0.4208219178082192, + "grad_norm": 2.6305623054504395, + "learning_rate": 4.80706742485784e-06, + "loss": 1.959, + "step": 384 + }, + { + "epoch": 0.42191780821917807, + "grad_norm": 2.512516498565674, + "learning_rate": 4.805036555645816e-06, + "loss": 1.874, + "step": 385 + }, + { + "epoch": 0.423013698630137, + "grad_norm": 2.622652530670166, + "learning_rate": 4.8030056864337945e-06, + "loss": 1.9404, + "step": 386 + }, + { + "epoch": 0.4241095890410959, + "grad_norm": 2.4640324115753174, + "learning_rate": 4.800974817221771e-06, + "loss": 1.8976, + "step": 387 + }, + { + "epoch": 0.4252054794520548, + "grad_norm": 2.490764856338501, + "learning_rate": 4.798943948009749e-06, + "loss": 1.9041, + "step": 388 + }, + { + "epoch": 0.4263013698630137, + "grad_norm": 2.2342710494995117, + "learning_rate": 4.7969130787977254e-06, + "loss": 1.8972, + "step": 389 + }, + { + "epoch": 0.4273972602739726, + "grad_norm": 2.4572854042053223, + "learning_rate": 4.794882209585703e-06, + "loss": 1.933, + "step": 390 + }, + { + "epoch": 0.4273972602739726, + "eval_loss": 2.0589487552642822, + "eval_runtime": 44.517, + "eval_samples_per_second": 22.463, + "eval_steps_per_second": 0.943, + "step": 390 + }, + { + "epoch": 0.4284931506849315, + "grad_norm": 2.545597553253174, + "learning_rate": 4.79285134037368e-06, + "loss": 1.8734, + "step": 391 + }, + { + "epoch": 0.4295890410958904, + "grad_norm": 2.1523783206939697, + "learning_rate": 4.790820471161657e-06, + "loss": 1.8738, + "step": 392 + }, + { + "epoch": 0.43068493150684933, + "grad_norm": 2.5466248989105225, + "learning_rate": 4.7887896019496345e-06, + "loss": 1.8835, + "step": 393 + }, + { + "epoch": 0.4317808219178082, + "grad_norm": 2.48857045173645, + "learning_rate": 4.786758732737612e-06, + "loss": 1.9364, + "step": 394 + }, + { + "epoch": 0.4328767123287671, + "grad_norm": 2.3035526275634766, + "learning_rate": 4.784727863525589e-06, + "loss": 1.8482, + "step": 395 + }, + { + "epoch": 0.43397260273972604, + "grad_norm": 2.565887451171875, + "learning_rate": 4.782696994313566e-06, + "loss": 1.8953, + "step": 396 + }, + { + "epoch": 0.4350684931506849, + "grad_norm": 2.4048995971679688, + "learning_rate": 4.780666125101544e-06, + "loss": 1.8366, + "step": 397 + }, + { + "epoch": 0.43616438356164383, + "grad_norm": 2.3187754154205322, + "learning_rate": 4.778635255889521e-06, + "loss": 1.8625, + "step": 398 + }, + { + "epoch": 0.43726027397260275, + "grad_norm": 2.5419323444366455, + "learning_rate": 4.776604386677498e-06, + "loss": 1.8768, + "step": 399 + }, + { + "epoch": 0.4383561643835616, + "grad_norm": 2.5917835235595703, + "learning_rate": 4.7745735174654755e-06, + "loss": 1.9355, + "step": 400 + }, + { + "epoch": 0.4383561643835616, + "eval_loss": 2.055696725845337, + "eval_runtime": 45.8295, + "eval_samples_per_second": 21.82, + "eval_steps_per_second": 0.916, + "step": 400 + }, + { + "epoch": 0.43945205479452054, + "grad_norm": 2.293874740600586, + "learning_rate": 4.772542648253453e-06, + "loss": 1.9239, + "step": 401 + }, + { + "epoch": 0.44054794520547946, + "grad_norm": 2.388672113418579, + "learning_rate": 4.77051177904143e-06, + "loss": 1.8583, + "step": 402 + }, + { + "epoch": 0.44164383561643833, + "grad_norm": 2.455951452255249, + "learning_rate": 4.768480909829407e-06, + "loss": 1.9161, + "step": 403 + }, + { + "epoch": 0.44273972602739725, + "grad_norm": 2.553738832473755, + "learning_rate": 4.766450040617385e-06, + "loss": 1.877, + "step": 404 + }, + { + "epoch": 0.4438356164383562, + "grad_norm": 2.6842715740203857, + "learning_rate": 4.764419171405362e-06, + "loss": 1.9279, + "step": 405 + }, + { + "epoch": 0.4449315068493151, + "grad_norm": 2.5113956928253174, + "learning_rate": 4.762388302193339e-06, + "loss": 1.8994, + "step": 406 + }, + { + "epoch": 0.44602739726027396, + "grad_norm": 2.3073933124542236, + "learning_rate": 4.760357432981316e-06, + "loss": 1.8861, + "step": 407 + }, + { + "epoch": 0.4471232876712329, + "grad_norm": 2.4669084548950195, + "learning_rate": 4.758326563769294e-06, + "loss": 1.8907, + "step": 408 + }, + { + "epoch": 0.4482191780821918, + "grad_norm": 2.659658908843994, + "learning_rate": 4.756295694557271e-06, + "loss": 1.9004, + "step": 409 + }, + { + "epoch": 0.44931506849315067, + "grad_norm": 2.505967378616333, + "learning_rate": 4.754264825345248e-06, + "loss": 1.8365, + "step": 410 + }, + { + "epoch": 0.44931506849315067, + "eval_loss": 2.0524449348449707, + "eval_runtime": 41.8597, + "eval_samples_per_second": 23.889, + "eval_steps_per_second": 1.003, + "step": 410 + }, + { + "epoch": 0.4504109589041096, + "grad_norm": 2.4327261447906494, + "learning_rate": 4.7522339561332255e-06, + "loss": 1.8566, + "step": 411 + }, + { + "epoch": 0.4515068493150685, + "grad_norm": 2.727325916290283, + "learning_rate": 4.750203086921203e-06, + "loss": 1.8088, + "step": 412 + }, + { + "epoch": 0.4526027397260274, + "grad_norm": 2.286327838897705, + "learning_rate": 4.74817221770918e-06, + "loss": 1.9004, + "step": 413 + }, + { + "epoch": 0.4536986301369863, + "grad_norm": 2.795856237411499, + "learning_rate": 4.7461413484971565e-06, + "loss": 1.9413, + "step": 414 + }, + { + "epoch": 0.4547945205479452, + "grad_norm": 2.4858641624450684, + "learning_rate": 4.744110479285135e-06, + "loss": 1.8169, + "step": 415 + }, + { + "epoch": 0.4558904109589041, + "grad_norm": 2.5095131397247314, + "learning_rate": 4.742079610073111e-06, + "loss": 1.8411, + "step": 416 + }, + { + "epoch": 0.456986301369863, + "grad_norm": 2.658841609954834, + "learning_rate": 4.740048740861089e-06, + "loss": 1.8675, + "step": 417 + }, + { + "epoch": 0.45808219178082193, + "grad_norm": 2.243589162826538, + "learning_rate": 4.738017871649066e-06, + "loss": 1.8749, + "step": 418 + }, + { + "epoch": 0.4591780821917808, + "grad_norm": 3.090933322906494, + "learning_rate": 4.735987002437044e-06, + "loss": 1.9215, + "step": 419 + }, + { + "epoch": 0.4602739726027397, + "grad_norm": 2.543403387069702, + "learning_rate": 4.73395613322502e-06, + "loss": 1.9416, + "step": 420 + }, + { + "epoch": 0.4602739726027397, + "eval_loss": 2.0476129055023193, + "eval_runtime": 42.3121, + "eval_samples_per_second": 23.634, + "eval_steps_per_second": 0.993, + "step": 420 + }, + { + "epoch": 0.46136986301369864, + "grad_norm": 2.544772148132324, + "learning_rate": 4.731925264012998e-06, + "loss": 1.9062, + "step": 421 + }, + { + "epoch": 0.4624657534246575, + "grad_norm": 2.6936161518096924, + "learning_rate": 4.729894394800975e-06, + "loss": 1.8338, + "step": 422 + }, + { + "epoch": 0.46356164383561643, + "grad_norm": 2.5742850303649902, + "learning_rate": 4.727863525588953e-06, + "loss": 1.8066, + "step": 423 + }, + { + "epoch": 0.46465753424657535, + "grad_norm": 2.4559762477874756, + "learning_rate": 4.725832656376929e-06, + "loss": 1.852, + "step": 424 + }, + { + "epoch": 0.4657534246575342, + "grad_norm": 2.650364637374878, + "learning_rate": 4.723801787164907e-06, + "loss": 1.901, + "step": 425 + }, + { + "epoch": 0.46684931506849314, + "grad_norm": 2.4269678592681885, + "learning_rate": 4.721770917952885e-06, + "loss": 1.838, + "step": 426 + }, + { + "epoch": 0.46794520547945206, + "grad_norm": 2.61296010017395, + "learning_rate": 4.719740048740861e-06, + "loss": 1.8712, + "step": 427 + }, + { + "epoch": 0.469041095890411, + "grad_norm": 2.509075164794922, + "learning_rate": 4.717709179528839e-06, + "loss": 1.8918, + "step": 428 + }, + { + "epoch": 0.47013698630136985, + "grad_norm": 2.422539234161377, + "learning_rate": 4.715678310316816e-06, + "loss": 1.8756, + "step": 429 + }, + { + "epoch": 0.4712328767123288, + "grad_norm": 2.7383527755737305, + "learning_rate": 4.713647441104794e-06, + "loss": 1.8492, + "step": 430 + }, + { + "epoch": 0.4712328767123288, + "eval_loss": 2.0373315811157227, + "eval_runtime": 42.8812, + "eval_samples_per_second": 23.32, + "eval_steps_per_second": 0.979, + "step": 430 + }, + { + "epoch": 0.4723287671232877, + "grad_norm": 2.5969882011413574, + "learning_rate": 4.71161657189277e-06, + "loss": 1.8399, + "step": 431 + }, + { + "epoch": 0.47342465753424656, + "grad_norm": 2.4129638671875, + "learning_rate": 4.709585702680748e-06, + "loss": 1.9084, + "step": 432 + }, + { + "epoch": 0.4745205479452055, + "grad_norm": 2.526293992996216, + "learning_rate": 4.707554833468725e-06, + "loss": 1.9568, + "step": 433 + }, + { + "epoch": 0.4756164383561644, + "grad_norm": 2.8506112098693848, + "learning_rate": 4.705523964256703e-06, + "loss": 1.8468, + "step": 434 + }, + { + "epoch": 0.4767123287671233, + "grad_norm": 2.497939109802246, + "learning_rate": 4.703493095044679e-06, + "loss": 1.866, + "step": 435 + }, + { + "epoch": 0.4778082191780822, + "grad_norm": 2.702066659927368, + "learning_rate": 4.701462225832657e-06, + "loss": 1.8046, + "step": 436 + }, + { + "epoch": 0.4789041095890411, + "grad_norm": 2.7489166259765625, + "learning_rate": 4.699431356620634e-06, + "loss": 1.9353, + "step": 437 + }, + { + "epoch": 0.48, + "grad_norm": 2.3662068843841553, + "learning_rate": 4.697400487408612e-06, + "loss": 1.8865, + "step": 438 + }, + { + "epoch": 0.4810958904109589, + "grad_norm": 2.72012996673584, + "learning_rate": 4.695369618196588e-06, + "loss": 1.878, + "step": 439 + }, + { + "epoch": 0.4821917808219178, + "grad_norm": 2.6107311248779297, + "learning_rate": 4.693338748984566e-06, + "loss": 1.8591, + "step": 440 + }, + { + "epoch": 0.4821917808219178, + "eval_loss": 2.0337164402008057, + "eval_runtime": 41.7046, + "eval_samples_per_second": 23.978, + "eval_steps_per_second": 1.007, + "step": 440 + }, + { + "epoch": 0.4832876712328767, + "grad_norm": 2.6732606887817383, + "learning_rate": 4.691307879772543e-06, + "loss": 1.8257, + "step": 441 + }, + { + "epoch": 0.4843835616438356, + "grad_norm": 2.579104423522949, + "learning_rate": 4.68927701056052e-06, + "loss": 1.911, + "step": 442 + }, + { + "epoch": 0.48547945205479454, + "grad_norm": 2.550381898880005, + "learning_rate": 4.6872461413484975e-06, + "loss": 1.8852, + "step": 443 + }, + { + "epoch": 0.4865753424657534, + "grad_norm": 2.467207193374634, + "learning_rate": 4.685215272136475e-06, + "loss": 1.8739, + "step": 444 + }, + { + "epoch": 0.4876712328767123, + "grad_norm": 2.493243932723999, + "learning_rate": 4.683184402924452e-06, + "loss": 1.867, + "step": 445 + }, + { + "epoch": 0.48876712328767125, + "grad_norm": 2.6451704502105713, + "learning_rate": 4.681153533712429e-06, + "loss": 1.9383, + "step": 446 + }, + { + "epoch": 0.4898630136986301, + "grad_norm": 2.5728414058685303, + "learning_rate": 4.679122664500407e-06, + "loss": 1.833, + "step": 447 + }, + { + "epoch": 0.49095890410958903, + "grad_norm": 2.4008536338806152, + "learning_rate": 4.677091795288384e-06, + "loss": 1.8092, + "step": 448 + }, + { + "epoch": 0.49205479452054796, + "grad_norm": 2.524291753768921, + "learning_rate": 4.675060926076361e-06, + "loss": 1.912, + "step": 449 + }, + { + "epoch": 0.4931506849315068, + "grad_norm": 2.4657843112945557, + "learning_rate": 4.673030056864338e-06, + "loss": 1.8057, + "step": 450 + }, + { + "epoch": 0.4931506849315068, + "eval_loss": 2.0297820568084717, + "eval_runtime": 45.3047, + "eval_samples_per_second": 22.073, + "eval_steps_per_second": 0.927, + "step": 450 + }, + { + "epoch": 0.49424657534246574, + "grad_norm": 2.371495008468628, + "learning_rate": 4.670999187652316e-06, + "loss": 1.8279, + "step": 451 + }, + { + "epoch": 0.49534246575342467, + "grad_norm": 2.367461919784546, + "learning_rate": 4.668968318440293e-06, + "loss": 1.821, + "step": 452 + }, + { + "epoch": 0.4964383561643836, + "grad_norm": 2.499398708343506, + "learning_rate": 4.66693744922827e-06, + "loss": 1.8923, + "step": 453 + }, + { + "epoch": 0.49753424657534245, + "grad_norm": 2.2641563415527344, + "learning_rate": 4.6649065800162475e-06, + "loss": 1.7918, + "step": 454 + }, + { + "epoch": 0.4986301369863014, + "grad_norm": 2.6121327877044678, + "learning_rate": 4.662875710804225e-06, + "loss": 1.8706, + "step": 455 + }, + { + "epoch": 0.4997260273972603, + "grad_norm": 2.25490403175354, + "learning_rate": 4.660844841592202e-06, + "loss": 1.79, + "step": 456 + }, + { + "epoch": 0.5008219178082192, + "grad_norm": 3.2300243377685547, + "learning_rate": 4.658813972380179e-06, + "loss": 1.9029, + "step": 457 + }, + { + "epoch": 0.5019178082191781, + "grad_norm": 2.537505626678467, + "learning_rate": 4.656783103168157e-06, + "loss": 1.7776, + "step": 458 + }, + { + "epoch": 0.503013698630137, + "grad_norm": 3.5579898357391357, + "learning_rate": 4.654752233956134e-06, + "loss": 1.8605, + "step": 459 + }, + { + "epoch": 0.5041095890410959, + "grad_norm": 2.3642566204071045, + "learning_rate": 4.652721364744111e-06, + "loss": 1.8437, + "step": 460 + }, + { + "epoch": 0.5041095890410959, + "eval_loss": 2.030970811843872, + "eval_runtime": 45.663, + "eval_samples_per_second": 21.9, + "eval_steps_per_second": 0.92, + "step": 460 + }, + { + "epoch": 0.5052054794520547, + "grad_norm": 2.901538133621216, + "learning_rate": 4.6506904955320885e-06, + "loss": 1.8993, + "step": 461 + }, + { + "epoch": 0.5063013698630137, + "grad_norm": 2.3952205181121826, + "learning_rate": 4.648659626320066e-06, + "loss": 1.8835, + "step": 462 + }, + { + "epoch": 0.5073972602739726, + "grad_norm": 3.1590476036071777, + "learning_rate": 4.646628757108043e-06, + "loss": 1.8521, + "step": 463 + }, + { + "epoch": 0.5084931506849315, + "grad_norm": 2.626070976257324, + "learning_rate": 4.6445978878960194e-06, + "loss": 1.8176, + "step": 464 + }, + { + "epoch": 0.5095890410958904, + "grad_norm": 4.6324286460876465, + "learning_rate": 4.6425670186839976e-06, + "loss": 1.8601, + "step": 465 + }, + { + "epoch": 0.5106849315068493, + "grad_norm": 3.3459413051605225, + "learning_rate": 4.640536149471974e-06, + "loss": 1.8909, + "step": 466 + }, + { + "epoch": 0.5117808219178083, + "grad_norm": 2.9864730834960938, + "learning_rate": 4.638505280259952e-06, + "loss": 1.8921, + "step": 467 + }, + { + "epoch": 0.5128767123287671, + "grad_norm": 2.6117241382598877, + "learning_rate": 4.6364744110479285e-06, + "loss": 1.8533, + "step": 468 + }, + { + "epoch": 0.513972602739726, + "grad_norm": 4.3972249031066895, + "learning_rate": 4.634443541835907e-06, + "loss": 1.9038, + "step": 469 + }, + { + "epoch": 0.5150684931506849, + "grad_norm": 2.7811403274536133, + "learning_rate": 4.632412672623883e-06, + "loss": 1.8459, + "step": 470 + }, + { + "epoch": 0.5150684931506849, + "eval_loss": 2.0335161685943604, + "eval_runtime": 43.3492, + "eval_samples_per_second": 23.068, + "eval_steps_per_second": 0.969, + "step": 470 + }, + { + "epoch": 0.5161643835616438, + "grad_norm": 7.532620906829834, + "learning_rate": 4.630381803411861e-06, + "loss": 1.7474, + "step": 471 + }, + { + "epoch": 0.5172602739726028, + "grad_norm": 6.09933614730835, + "learning_rate": 4.628350934199838e-06, + "loss": 1.8457, + "step": 472 + }, + { + "epoch": 0.5183561643835617, + "grad_norm": 3.9755234718322754, + "learning_rate": 4.626320064987816e-06, + "loss": 1.8968, + "step": 473 + }, + { + "epoch": 0.5194520547945205, + "grad_norm": 2.8951804637908936, + "learning_rate": 4.624289195775792e-06, + "loss": 1.8552, + "step": 474 + }, + { + "epoch": 0.5205479452054794, + "grad_norm": 4.744227409362793, + "learning_rate": 4.62225832656377e-06, + "loss": 1.8469, + "step": 475 + }, + { + "epoch": 0.5216438356164383, + "grad_norm": 2.669635772705078, + "learning_rate": 4.620227457351747e-06, + "loss": 1.8517, + "step": 476 + }, + { + "epoch": 0.5227397260273973, + "grad_norm": 3.0501551628112793, + "learning_rate": 4.618196588139724e-06, + "loss": 1.8662, + "step": 477 + }, + { + "epoch": 0.5238356164383562, + "grad_norm": 3.5298409461975098, + "learning_rate": 4.616165718927701e-06, + "loss": 1.8304, + "step": 478 + }, + { + "epoch": 0.5249315068493151, + "grad_norm": 2.763378143310547, + "learning_rate": 4.6141348497156786e-06, + "loss": 1.9114, + "step": 479 + }, + { + "epoch": 0.5260273972602739, + "grad_norm": 2.618973731994629, + "learning_rate": 4.612103980503656e-06, + "loss": 1.8871, + "step": 480 + }, + { + "epoch": 0.5260273972602739, + "eval_loss": 2.021653652191162, + "eval_runtime": 45.2748, + "eval_samples_per_second": 22.087, + "eval_steps_per_second": 0.928, + "step": 480 + }, + { + "epoch": 0.5271232876712328, + "grad_norm": 2.825256586074829, + "learning_rate": 4.610073111291633e-06, + "loss": 1.8559, + "step": 481 + }, + { + "epoch": 0.5282191780821918, + "grad_norm": 2.795807123184204, + "learning_rate": 4.60804224207961e-06, + "loss": 1.8441, + "step": 482 + }, + { + "epoch": 0.5293150684931507, + "grad_norm": 2.6094915866851807, + "learning_rate": 4.606011372867588e-06, + "loss": 1.8672, + "step": 483 + }, + { + "epoch": 0.5304109589041096, + "grad_norm": 2.6316585540771484, + "learning_rate": 4.603980503655565e-06, + "loss": 1.8473, + "step": 484 + }, + { + "epoch": 0.5315068493150685, + "grad_norm": 2.8802528381347656, + "learning_rate": 4.601949634443542e-06, + "loss": 1.7769, + "step": 485 + }, + { + "epoch": 0.5326027397260275, + "grad_norm": 2.5248327255249023, + "learning_rate": 4.5999187652315195e-06, + "loss": 1.7855, + "step": 486 + }, + { + "epoch": 0.5336986301369863, + "grad_norm": 2.6327872276306152, + "learning_rate": 4.597887896019497e-06, + "loss": 1.8439, + "step": 487 + }, + { + "epoch": 0.5347945205479452, + "grad_norm": 2.6007511615753174, + "learning_rate": 4.595857026807474e-06, + "loss": 1.8408, + "step": 488 + }, + { + "epoch": 0.5358904109589041, + "grad_norm": 3.1975581645965576, + "learning_rate": 4.593826157595451e-06, + "loss": 1.8136, + "step": 489 + }, + { + "epoch": 0.536986301369863, + "grad_norm": 2.620628833770752, + "learning_rate": 4.591795288383429e-06, + "loss": 1.893, + "step": 490 + }, + { + "epoch": 0.536986301369863, + "eval_loss": 2.018796920776367, + "eval_runtime": 43.9241, + "eval_samples_per_second": 22.767, + "eval_steps_per_second": 0.956, + "step": 490 + }, + { + "epoch": 0.538082191780822, + "grad_norm": 2.4748106002807617, + "learning_rate": 4.589764419171406e-06, + "loss": 1.8625, + "step": 491 + }, + { + "epoch": 0.5391780821917809, + "grad_norm": 4.04419469833374, + "learning_rate": 4.587733549959383e-06, + "loss": 1.8184, + "step": 492 + }, + { + "epoch": 0.5402739726027397, + "grad_norm": 2.2681894302368164, + "learning_rate": 4.5857026807473604e-06, + "loss": 1.7962, + "step": 493 + }, + { + "epoch": 0.5413698630136986, + "grad_norm": 2.564131498336792, + "learning_rate": 4.583671811535338e-06, + "loss": 1.783, + "step": 494 + }, + { + "epoch": 0.5424657534246575, + "grad_norm": 3.393205165863037, + "learning_rate": 4.581640942323315e-06, + "loss": 1.8806, + "step": 495 + }, + { + "epoch": 0.5435616438356164, + "grad_norm": 2.496661901473999, + "learning_rate": 4.579610073111292e-06, + "loss": 1.8838, + "step": 496 + }, + { + "epoch": 0.5446575342465754, + "grad_norm": 2.265989065170288, + "learning_rate": 4.5775792038992695e-06, + "loss": 1.8752, + "step": 497 + }, + { + "epoch": 0.5457534246575343, + "grad_norm": 2.88940691947937, + "learning_rate": 4.575548334687247e-06, + "loss": 1.9097, + "step": 498 + }, + { + "epoch": 0.5468493150684931, + "grad_norm": 2.9301936626434326, + "learning_rate": 4.573517465475224e-06, + "loss": 1.7972, + "step": 499 + }, + { + "epoch": 0.547945205479452, + "grad_norm": 2.321361780166626, + "learning_rate": 4.571486596263201e-06, + "loss": 1.8514, + "step": 500 + }, + { + "epoch": 0.547945205479452, + "eval_loss": 2.0164716243743896, + "eval_runtime": 45.8538, + "eval_samples_per_second": 21.808, + "eval_steps_per_second": 0.916, + "step": 500 + }, + { + "epoch": 0.549041095890411, + "grad_norm": 2.5818793773651123, + "learning_rate": 4.569455727051178e-06, + "loss": 1.85, + "step": 501 + }, + { + "epoch": 0.5501369863013699, + "grad_norm": 3.0734691619873047, + "learning_rate": 4.567424857839156e-06, + "loss": 1.8682, + "step": 502 + }, + { + "epoch": 0.5512328767123288, + "grad_norm": 2.496253252029419, + "learning_rate": 4.565393988627132e-06, + "loss": 1.9007, + "step": 503 + }, + { + "epoch": 0.5523287671232877, + "grad_norm": 2.360891342163086, + "learning_rate": 4.5633631194151105e-06, + "loss": 1.8262, + "step": 504 + }, + { + "epoch": 0.5534246575342465, + "grad_norm": 2.5731117725372314, + "learning_rate": 4.561332250203087e-06, + "loss": 1.8369, + "step": 505 + }, + { + "epoch": 0.5545205479452054, + "grad_norm": 2.700558662414551, + "learning_rate": 4.559301380991065e-06, + "loss": 1.7825, + "step": 506 + }, + { + "epoch": 0.5556164383561644, + "grad_norm": 2.561306953430176, + "learning_rate": 4.5572705117790414e-06, + "loss": 1.8174, + "step": 507 + }, + { + "epoch": 0.5567123287671233, + "grad_norm": 2.618889570236206, + "learning_rate": 4.5552396425670196e-06, + "loss": 1.8812, + "step": 508 + }, + { + "epoch": 0.5578082191780822, + "grad_norm": 2.4671308994293213, + "learning_rate": 4.553208773354996e-06, + "loss": 1.7959, + "step": 509 + }, + { + "epoch": 0.5589041095890411, + "grad_norm": 2.4571878910064697, + "learning_rate": 4.551177904142974e-06, + "loss": 1.8943, + "step": 510 + }, + { + "epoch": 0.5589041095890411, + "eval_loss": 2.010777711868286, + "eval_runtime": 43.7424, + "eval_samples_per_second": 22.861, + "eval_steps_per_second": 0.96, + "step": 510 + }, + { + "epoch": 0.56, + "grad_norm": 2.4786667823791504, + "learning_rate": 4.5491470349309505e-06, + "loss": 1.8147, + "step": 511 + }, + { + "epoch": 0.5610958904109589, + "grad_norm": 2.2955000400543213, + "learning_rate": 4.547116165718929e-06, + "loss": 1.8547, + "step": 512 + }, + { + "epoch": 0.5621917808219178, + "grad_norm": 2.3873913288116455, + "learning_rate": 4.545085296506905e-06, + "loss": 1.8156, + "step": 513 + }, + { + "epoch": 0.5632876712328767, + "grad_norm": 2.2449734210968018, + "learning_rate": 4.543054427294882e-06, + "loss": 1.8053, + "step": 514 + }, + { + "epoch": 0.5643835616438356, + "grad_norm": 2.3969857692718506, + "learning_rate": 4.54102355808286e-06, + "loss": 1.8085, + "step": 515 + }, + { + "epoch": 0.5654794520547946, + "grad_norm": 2.3446357250213623, + "learning_rate": 4.538992688870837e-06, + "loss": 1.8355, + "step": 516 + }, + { + "epoch": 0.5665753424657535, + "grad_norm": 2.408762216567993, + "learning_rate": 4.536961819658814e-06, + "loss": 1.8319, + "step": 517 + }, + { + "epoch": 0.5676712328767123, + "grad_norm": 2.362962245941162, + "learning_rate": 4.5349309504467915e-06, + "loss": 1.8422, + "step": 518 + }, + { + "epoch": 0.5687671232876712, + "grad_norm": 2.402984857559204, + "learning_rate": 4.532900081234769e-06, + "loss": 1.8125, + "step": 519 + }, + { + "epoch": 0.5698630136986301, + "grad_norm": 2.3964452743530273, + "learning_rate": 4.530869212022746e-06, + "loss": 1.8417, + "step": 520 + }, + { + "epoch": 0.5698630136986301, + "eval_loss": 2.006908416748047, + "eval_runtime": 44.2701, + "eval_samples_per_second": 22.589, + "eval_steps_per_second": 0.949, + "step": 520 + }, + { + "epoch": 0.570958904109589, + "grad_norm": 2.595233201980591, + "learning_rate": 4.528838342810723e-06, + "loss": 1.8275, + "step": 521 + }, + { + "epoch": 0.572054794520548, + "grad_norm": 2.326117753982544, + "learning_rate": 4.526807473598701e-06, + "loss": 1.8405, + "step": 522 + }, + { + "epoch": 0.5731506849315069, + "grad_norm": 2.4904465675354004, + "learning_rate": 4.524776604386678e-06, + "loss": 1.8172, + "step": 523 + }, + { + "epoch": 0.5742465753424657, + "grad_norm": 2.4591610431671143, + "learning_rate": 4.522745735174655e-06, + "loss": 1.876, + "step": 524 + }, + { + "epoch": 0.5753424657534246, + "grad_norm": 2.319800853729248, + "learning_rate": 4.520714865962632e-06, + "loss": 1.8294, + "step": 525 + }, + { + "epoch": 0.5764383561643835, + "grad_norm": 2.28790020942688, + "learning_rate": 4.51868399675061e-06, + "loss": 1.8537, + "step": 526 + }, + { + "epoch": 0.5775342465753425, + "grad_norm": 2.284822702407837, + "learning_rate": 4.516653127538587e-06, + "loss": 1.8087, + "step": 527 + }, + { + "epoch": 0.5786301369863014, + "grad_norm": 2.273820638656616, + "learning_rate": 4.514622258326564e-06, + "loss": 1.7992, + "step": 528 + }, + { + "epoch": 0.5797260273972603, + "grad_norm": 2.2025654315948486, + "learning_rate": 4.5125913891145415e-06, + "loss": 1.8248, + "step": 529 + }, + { + "epoch": 0.5808219178082191, + "grad_norm": 2.3218843936920166, + "learning_rate": 4.510560519902519e-06, + "loss": 1.8647, + "step": 530 + }, + { + "epoch": 0.5808219178082191, + "eval_loss": 2.0048956871032715, + "eval_runtime": 44.5972, + "eval_samples_per_second": 22.423, + "eval_steps_per_second": 0.942, + "step": 530 + }, + { + "epoch": 0.581917808219178, + "grad_norm": 2.515766143798828, + "learning_rate": 4.508529650690496e-06, + "loss": 1.7673, + "step": 531 + }, + { + "epoch": 0.583013698630137, + "grad_norm": 2.4376487731933594, + "learning_rate": 4.506498781478473e-06, + "loss": 1.7755, + "step": 532 + }, + { + "epoch": 0.5841095890410959, + "grad_norm": 2.2801272869110107, + "learning_rate": 4.504467912266451e-06, + "loss": 1.8273, + "step": 533 + }, + { + "epoch": 0.5852054794520548, + "grad_norm": 2.4707891941070557, + "learning_rate": 4.502437043054428e-06, + "loss": 1.8971, + "step": 534 + }, + { + "epoch": 0.5863013698630137, + "grad_norm": 2.301301956176758, + "learning_rate": 4.500406173842405e-06, + "loss": 1.8345, + "step": 535 + }, + { + "epoch": 0.5873972602739727, + "grad_norm": 2.2317373752593994, + "learning_rate": 4.4983753046303824e-06, + "loss": 1.8248, + "step": 536 + }, + { + "epoch": 0.5884931506849315, + "grad_norm": 2.447373867034912, + "learning_rate": 4.49634443541836e-06, + "loss": 1.8148, + "step": 537 + }, + { + "epoch": 0.5895890410958904, + "grad_norm": 2.337245225906372, + "learning_rate": 4.494313566206336e-06, + "loss": 1.8811, + "step": 538 + }, + { + "epoch": 0.5906849315068493, + "grad_norm": 2.3717031478881836, + "learning_rate": 4.492282696994314e-06, + "loss": 1.8311, + "step": 539 + }, + { + "epoch": 0.5917808219178082, + "grad_norm": 2.287677526473999, + "learning_rate": 4.490251827782291e-06, + "loss": 1.7728, + "step": 540 + }, + { + "epoch": 0.5917808219178082, + "eval_loss": 2.004889965057373, + "eval_runtime": 43.8132, + "eval_samples_per_second": 22.824, + "eval_steps_per_second": 0.959, + "step": 540 + }, + { + "epoch": 0.5928767123287672, + "grad_norm": 2.441445827484131, + "learning_rate": 4.488220958570269e-06, + "loss": 1.7927, + "step": 541 + }, + { + "epoch": 0.5939726027397261, + "grad_norm": 2.3191535472869873, + "learning_rate": 4.486190089358245e-06, + "loss": 1.759, + "step": 542 + }, + { + "epoch": 0.5950684931506849, + "grad_norm": 2.3118820190429688, + "learning_rate": 4.484159220146223e-06, + "loss": 1.8509, + "step": 543 + }, + { + "epoch": 0.5961643835616438, + "grad_norm": 2.4711859226226807, + "learning_rate": 4.4821283509342e-06, + "loss": 1.8402, + "step": 544 + }, + { + "epoch": 0.5972602739726027, + "grad_norm": 2.3142173290252686, + "learning_rate": 4.480097481722178e-06, + "loss": 1.8238, + "step": 545 + }, + { + "epoch": 0.5983561643835617, + "grad_norm": 2.2665774822235107, + "learning_rate": 4.478066612510154e-06, + "loss": 1.7826, + "step": 546 + }, + { + "epoch": 0.5994520547945206, + "grad_norm": 2.1692490577697754, + "learning_rate": 4.4760357432981325e-06, + "loss": 1.8138, + "step": 547 + }, + { + "epoch": 0.6005479452054795, + "grad_norm": 2.412783145904541, + "learning_rate": 4.474004874086109e-06, + "loss": 1.9421, + "step": 548 + }, + { + "epoch": 0.6016438356164383, + "grad_norm": 2.3692686557769775, + "learning_rate": 4.471974004874087e-06, + "loss": 1.7722, + "step": 549 + }, + { + "epoch": 0.6027397260273972, + "grad_norm": 2.3224151134490967, + "learning_rate": 4.4699431356620635e-06, + "loss": 1.8753, + "step": 550 + }, + { + "epoch": 0.6027397260273972, + "eval_loss": 2.0016305446624756, + "eval_runtime": 45.564, + "eval_samples_per_second": 21.947, + "eval_steps_per_second": 0.922, + "step": 550 + }, + { + "epoch": 0.6038356164383561, + "grad_norm": 2.51043438911438, + "learning_rate": 4.467912266450041e-06, + "loss": 1.8325, + "step": 551 + }, + { + "epoch": 0.6049315068493151, + "grad_norm": 2.4568657875061035, + "learning_rate": 4.465881397238018e-06, + "loss": 1.8081, + "step": 552 + }, + { + "epoch": 0.606027397260274, + "grad_norm": 2.3279001712799072, + "learning_rate": 4.463850528025995e-06, + "loss": 1.8439, + "step": 553 + }, + { + "epoch": 0.6071232876712329, + "grad_norm": 2.6041171550750732, + "learning_rate": 4.4618196588139726e-06, + "loss": 1.8623, + "step": 554 + }, + { + "epoch": 0.6082191780821918, + "grad_norm": 2.3780531883239746, + "learning_rate": 4.45978878960195e-06, + "loss": 1.8269, + "step": 555 + }, + { + "epoch": 0.6093150684931506, + "grad_norm": 2.511500597000122, + "learning_rate": 4.457757920389927e-06, + "loss": 1.7749, + "step": 556 + }, + { + "epoch": 0.6104109589041096, + "grad_norm": 2.707789659500122, + "learning_rate": 4.455727051177904e-06, + "loss": 1.8157, + "step": 557 + }, + { + "epoch": 0.6115068493150685, + "grad_norm": 2.5495495796203613, + "learning_rate": 4.453696181965882e-06, + "loss": 1.8534, + "step": 558 + }, + { + "epoch": 0.6126027397260274, + "grad_norm": 2.255876302719116, + "learning_rate": 4.451665312753859e-06, + "loss": 1.8377, + "step": 559 + }, + { + "epoch": 0.6136986301369863, + "grad_norm": 2.415097951889038, + "learning_rate": 4.449634443541836e-06, + "loss": 1.8764, + "step": 560 + }, + { + "epoch": 0.6136986301369863, + "eval_loss": 1.9943474531173706, + "eval_runtime": 42.2985, + "eval_samples_per_second": 23.642, + "eval_steps_per_second": 0.993, + "step": 560 + }, + { + "epoch": 0.6147945205479453, + "grad_norm": 2.2604713439941406, + "learning_rate": 4.4476035743298135e-06, + "loss": 1.7898, + "step": 561 + }, + { + "epoch": 0.6158904109589041, + "grad_norm": 2.5643842220306396, + "learning_rate": 4.445572705117791e-06, + "loss": 1.7953, + "step": 562 + }, + { + "epoch": 0.616986301369863, + "grad_norm": 2.412548780441284, + "learning_rate": 4.443541835905768e-06, + "loss": 1.8611, + "step": 563 + }, + { + "epoch": 0.6180821917808219, + "grad_norm": 2.5423877239227295, + "learning_rate": 4.441510966693745e-06, + "loss": 1.8049, + "step": 564 + }, + { + "epoch": 0.6191780821917808, + "grad_norm": 2.4986631870269775, + "learning_rate": 4.439480097481723e-06, + "loss": 1.8759, + "step": 565 + }, + { + "epoch": 0.6202739726027398, + "grad_norm": 2.8349273204803467, + "learning_rate": 4.4374492282697e-06, + "loss": 1.7416, + "step": 566 + }, + { + "epoch": 0.6213698630136987, + "grad_norm": 2.452017068862915, + "learning_rate": 4.435418359057677e-06, + "loss": 1.8268, + "step": 567 + }, + { + "epoch": 0.6224657534246575, + "grad_norm": 2.6425273418426514, + "learning_rate": 4.433387489845654e-06, + "loss": 1.8334, + "step": 568 + }, + { + "epoch": 0.6235616438356164, + "grad_norm": 2.328740358352661, + "learning_rate": 4.431356620633632e-06, + "loss": 1.7931, + "step": 569 + }, + { + "epoch": 0.6246575342465753, + "grad_norm": 2.262209415435791, + "learning_rate": 4.429325751421609e-06, + "loss": 1.8151, + "step": 570 + }, + { + "epoch": 0.6246575342465753, + "eval_loss": 1.9917123317718506, + "eval_runtime": 42.4309, + "eval_samples_per_second": 23.568, + "eval_steps_per_second": 0.99, + "step": 570 + }, + { + "epoch": 0.6257534246575343, + "grad_norm": 2.2984049320220947, + "learning_rate": 4.427294882209586e-06, + "loss": 1.8187, + "step": 571 + }, + { + "epoch": 0.6268493150684932, + "grad_norm": 2.3910019397735596, + "learning_rate": 4.4252640129975635e-06, + "loss": 1.7512, + "step": 572 + }, + { + "epoch": 0.6279452054794521, + "grad_norm": 2.4494471549987793, + "learning_rate": 4.423233143785541e-06, + "loss": 1.8202, + "step": 573 + }, + { + "epoch": 0.6290410958904109, + "grad_norm": 2.3857171535491943, + "learning_rate": 4.421202274573518e-06, + "loss": 1.8279, + "step": 574 + }, + { + "epoch": 0.6301369863013698, + "grad_norm": 2.4526500701904297, + "learning_rate": 4.4191714053614945e-06, + "loss": 1.7136, + "step": 575 + }, + { + "epoch": 0.6312328767123287, + "grad_norm": 2.2216501235961914, + "learning_rate": 4.417140536149473e-06, + "loss": 1.7574, + "step": 576 + }, + { + "epoch": 0.6323287671232877, + "grad_norm": 2.7639758586883545, + "learning_rate": 4.415109666937449e-06, + "loss": 1.8403, + "step": 577 + }, + { + "epoch": 0.6334246575342466, + "grad_norm": 2.3177521228790283, + "learning_rate": 4.413078797725427e-06, + "loss": 1.7898, + "step": 578 + }, + { + "epoch": 0.6345205479452055, + "grad_norm": 2.527902364730835, + "learning_rate": 4.411047928513404e-06, + "loss": 1.8184, + "step": 579 + }, + { + "epoch": 0.6356164383561644, + "grad_norm": 2.34061336517334, + "learning_rate": 4.409017059301382e-06, + "loss": 1.811, + "step": 580 + }, + { + "epoch": 0.6356164383561644, + "eval_loss": 1.9882216453552246, + "eval_runtime": 44.6687, + "eval_samples_per_second": 22.387, + "eval_steps_per_second": 0.94, + "step": 580 + }, + { + "epoch": 0.6367123287671232, + "grad_norm": 2.6462934017181396, + "learning_rate": 4.406986190089358e-06, + "loss": 1.8501, + "step": 581 + }, + { + "epoch": 0.6378082191780822, + "grad_norm": 2.2725913524627686, + "learning_rate": 4.404955320877336e-06, + "loss": 1.8262, + "step": 582 + }, + { + "epoch": 0.6389041095890411, + "grad_norm": 2.390084981918335, + "learning_rate": 4.402924451665313e-06, + "loss": 1.7842, + "step": 583 + }, + { + "epoch": 0.64, + "grad_norm": 2.314305305480957, + "learning_rate": 4.400893582453291e-06, + "loss": 1.7692, + "step": 584 + }, + { + "epoch": 0.6410958904109589, + "grad_norm": 2.2828257083892822, + "learning_rate": 4.398862713241267e-06, + "loss": 1.8031, + "step": 585 + }, + { + "epoch": 0.6421917808219179, + "grad_norm": 2.3493599891662598, + "learning_rate": 4.396831844029245e-06, + "loss": 1.8183, + "step": 586 + }, + { + "epoch": 0.6432876712328767, + "grad_norm": 2.423365831375122, + "learning_rate": 4.394800974817222e-06, + "loss": 1.8012, + "step": 587 + }, + { + "epoch": 0.6443835616438356, + "grad_norm": 2.438015937805176, + "learning_rate": 4.392770105605199e-06, + "loss": 1.8217, + "step": 588 + }, + { + "epoch": 0.6454794520547945, + "grad_norm": 2.357172727584839, + "learning_rate": 4.390739236393176e-06, + "loss": 1.8382, + "step": 589 + }, + { + "epoch": 0.6465753424657534, + "grad_norm": 2.3383474349975586, + "learning_rate": 4.388708367181154e-06, + "loss": 1.841, + "step": 590 + }, + { + "epoch": 0.6465753424657534, + "eval_loss": 1.9878787994384766, + "eval_runtime": 44.0694, + "eval_samples_per_second": 22.691, + "eval_steps_per_second": 0.953, + "step": 590 + }, + { + "epoch": 0.6476712328767124, + "grad_norm": 2.459385633468628, + "learning_rate": 4.386677497969131e-06, + "loss": 1.8377, + "step": 591 + }, + { + "epoch": 0.6487671232876713, + "grad_norm": 2.356788158416748, + "learning_rate": 4.384646628757108e-06, + "loss": 1.8718, + "step": 592 + }, + { + "epoch": 0.6498630136986301, + "grad_norm": 2.3217222690582275, + "learning_rate": 4.3826157595450855e-06, + "loss": 1.8073, + "step": 593 + }, + { + "epoch": 0.650958904109589, + "grad_norm": 2.3134331703186035, + "learning_rate": 4.380584890333063e-06, + "loss": 1.7675, + "step": 594 + }, + { + "epoch": 0.6520547945205479, + "grad_norm": 2.3273653984069824, + "learning_rate": 4.37855402112104e-06, + "loss": 1.7169, + "step": 595 + }, + { + "epoch": 0.6531506849315069, + "grad_norm": 2.611034393310547, + "learning_rate": 4.376523151909017e-06, + "loss": 1.8702, + "step": 596 + }, + { + "epoch": 0.6542465753424658, + "grad_norm": 2.2077300548553467, + "learning_rate": 4.3744922826969946e-06, + "loss": 1.7951, + "step": 597 + }, + { + "epoch": 0.6553424657534247, + "grad_norm": 2.3393311500549316, + "learning_rate": 4.372461413484972e-06, + "loss": 1.8136, + "step": 598 + }, + { + "epoch": 0.6564383561643835, + "grad_norm": 2.3306660652160645, + "learning_rate": 4.370430544272949e-06, + "loss": 1.8372, + "step": 599 + }, + { + "epoch": 0.6575342465753424, + "grad_norm": 2.4440250396728516, + "learning_rate": 4.368399675060926e-06, + "loss": 1.844, + "step": 600 + }, + { + "epoch": 0.6575342465753424, + "eval_loss": 1.98439621925354, + "eval_runtime": 44.4595, + "eval_samples_per_second": 22.492, + "eval_steps_per_second": 0.945, + "step": 600 + } + ], + "logging_steps": 1, + "max_steps": 2736, + "num_input_tokens_seen": 0, + "num_train_epochs": 3, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.870385645617152e+17, + "train_batch_size": 3, + "trial_name": null, + "trial_params": null +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/zero_to_fp32.py b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-600/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/config.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/config.json new file mode 100644 index 0000000000000000000000000000000000000000..fb0f9a53a507ffcdb60410deb71d60ef801bf350 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/config.json @@ -0,0 +1,36 @@ +{ + "_name_or_path": "meta-llama/Llama-3.2-3B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 3072, + "initializer_range": 0.02, + "intermediate_size": 8192, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 24, + "num_hidden_layers": 28, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 32.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": true, + "torch_dtype": "float16", + "transformers_version": "4.45.1", + "use_cache": true, + "vocab_size": 128256 +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/latest b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/latest new file mode 100644 index 0000000000000000000000000000000000000000..57729c0be88118cbd582c8c68b4149cee821f0b4 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/latest @@ -0,0 +1 @@ +global_step800 \ No newline at end of file diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/model.safetensors.index.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/model.safetensors.index.json new file mode 100644 index 0000000000000000000000000000000000000000..ed64de846d720b9a7859dc20575fea8e8ca51940 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/model.safetensors.index.json @@ -0,0 +1,262 @@ +{ + "metadata": { + "total_size": 7213504512 + }, + "weight_map": { + "lm_head.weight": "model-00002-of-00002.safetensors", + "model.embed_tokens.weight": "model-00001-of-00002.safetensors", + "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.21.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.24.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.25.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.26.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.input_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.down_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.gate_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.mlp.up_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.post_attention_layernorm.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.k_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.o_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.q_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.27.self_attn.v_proj.weight": "model-00002-of-00002.safetensors", + "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors", + "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors", + "model.norm.weight": "model-00002-of-00002.safetensors" + } +} diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/rng_state_0.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..10b2f5e4f6c9f32023941f5b8cf39645a86289fd --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2ccffad8722b9db65bab96085aca5b7293a1ef8d534c7b8817b948837c87f6b8 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/tokenizer.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..309789d5ccd06f418fbcfb7472a6a9affda68101 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d422bc7d90212ffb8070a743cdbe8ff2dfc5634b1fce8f2e3fa671ee441989eb +size 17210374 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/training_args.bin b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..1d4ddd41da4a46eae8dee0091f03595134907048 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-800/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:aeceb6dab6c8441ce9b39e4c6cb0939317837193ead4ec8bebf841770b625bcf +size 6011 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-900/rng_state_2.pth b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-900/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..137d87c143da2b66811890c8cf1e8b013ef6eacf --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-900/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:62c45c6121ed4cfe9395ee5153d26d376e4f02f8dfa1bbceafb8a5416efcdd14 +size 16567 diff --git a/babylm_shuffle_local10_10M_seed0/runs/checkpoint-900/special_tokens_map.json b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-900/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..6949c5975ee0e961ef61cf31010dce04df0a03f8 --- /dev/null +++ b/babylm_shuffle_local10_10M_seed0/runs/checkpoint-900/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "[PAD]", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +}