Add files using upload-large-folder tool

metallama3_8b/limo/checkpoint-205/chat_template.jinja

@@ -0,0 +1,5 @@
+{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>
+
+'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>
+
+' }}{% endif %}

metallama3_8b/limo/checkpoint-205/config.json

@@ -0,0 +1,29 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": 128009,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 8192,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.55.0",
+  "use_cache": false,
+  "vocab_size": 128256
+}

metallama3_8b/limo/checkpoint-205/model.safetensors.index.json

@@ -0,0 +1,299 @@
+{
+  "metadata": {
+    "total_parameters": 2007565312,
+    "total_size": 32121044992
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00007-of-00007.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00007.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00007.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00007-of-00007.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00007-of-00007.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00007-of-00007.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.norm.weight": "model-00007-of-00007.safetensors"
+  }
+}

metallama3_8b/limo/checkpoint-205/special_tokens_map.json

@@ -0,0 +1,26 @@
+{
+  "additional_special_tokens": [
+    {
+      "content": "<|eom_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    }
+  ],
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}

metallama3_8b/limo/checkpoint-205/trainer_state.json

@@ -0,0 +1,1469 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 205,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.004878048780487805,
+      "grad_norm": 27.79998016357422,
+      "learning_rate": 5e-06,
+      "loss": 1.4179,
+      "step": 1
+    },
+    {
+      "epoch": 0.00975609756097561,
+      "grad_norm": 4.086409091949463,
+      "learning_rate": 4.999997064365715e-06,
+      "loss": 1.1405,
+      "step": 2
+    },
+    {
+      "epoch": 0.014634146341463415,
+      "grad_norm": 4.499151229858398,
+      "learning_rate": 4.999988257469751e-06,
+      "loss": 0.8682,
+      "step": 3
+    },
+    {
+      "epoch": 0.01951219512195122,
+      "grad_norm": 4.555822849273682,
+      "learning_rate": 4.999973579332793e-06,
+      "loss": 0.9961,
+      "step": 4
+    },
+    {
+      "epoch": 0.024390243902439025,
+      "grad_norm": 5.6235246658325195,
+      "learning_rate": 4.999953029989312e-06,
+      "loss": 1.0173,
+      "step": 5
+    },
+    {
+      "epoch": 0.02926829268292683,
+      "grad_norm": 3.9943182468414307,
+      "learning_rate": 4.999926609487568e-06,
+      "loss": 1.1083,
+      "step": 6
+    },
+    {
+      "epoch": 0.03414634146341464,
+      "grad_norm": 5.685941219329834,
+      "learning_rate": 4.9998943178896106e-06,
+      "loss": 1.1109,
+      "step": 7
+    },
+    {
+      "epoch": 0.03902439024390244,
+      "grad_norm": 15.914257049560547,
+      "learning_rate": 4.999856155271276e-06,
+      "loss": 1.821,
+      "step": 8
+    },
+    {
+      "epoch": 0.04390243902439024,
+      "grad_norm": 4.147185325622559,
+      "learning_rate": 4.999812121722191e-06,
+      "loss": 1.0417,
+      "step": 9
+    },
+    {
+      "epoch": 0.04878048780487805,
+      "grad_norm": 11.123332977294922,
+      "learning_rate": 4.999762217345766e-06,
+      "loss": 1.5672,
+      "step": 10
+    },
+    {
+      "epoch": 0.05365853658536585,
+      "grad_norm": 2.842331886291504,
+      "learning_rate": 4.999706442259205e-06,
+      "loss": 0.7297,
+      "step": 11
+    },
+    {
+      "epoch": 0.05853658536585366,
+      "grad_norm": 37.685062408447266,
+      "learning_rate": 4.999644796593492e-06,
+      "loss": 0.9112,
+      "step": 12
+    },
+    {
+      "epoch": 0.06341463414634146,
+      "grad_norm": 11.214252471923828,
+      "learning_rate": 4.999577280493407e-06,
+      "loss": 0.7854,
+      "step": 13
+    },
+    {
+      "epoch": 0.06829268292682927,
+      "grad_norm": 5.10387659072876,
+      "learning_rate": 4.99950389411751e-06,
+      "loss": 1.1317,
+      "step": 14
+    },
+    {
+      "epoch": 0.07317073170731707,
+      "grad_norm": 3.685403347015381,
+      "learning_rate": 4.999424637638148e-06,
+      "loss": 0.7864,
+      "step": 15
+    },
+    {
+      "epoch": 0.07804878048780488,
+      "grad_norm": 2.9567184448242188,
+      "learning_rate": 4.999339511241458e-06,
+      "loss": 0.8494,
+      "step": 16
+    },
+    {
+      "epoch": 0.08292682926829269,
+      "grad_norm": 11.396956443786621,
+      "learning_rate": 4.9992485151273584e-06,
+      "loss": 1.2189,
+      "step": 17
+    },
+    {
+      "epoch": 0.08780487804878048,
+      "grad_norm": 7.007385730743408,
+      "learning_rate": 4.999151649509554e-06,
+      "loss": 1.0532,
+      "step": 18
+    },
+    {
+      "epoch": 0.09268292682926829,
+      "grad_norm": 3.4347329139709473,
+      "learning_rate": 4.9990489146155356e-06,
+      "loss": 1.088,
+      "step": 19
+    },
+    {
+      "epoch": 0.0975609756097561,
+      "grad_norm": 3.1865031719207764,
+      "learning_rate": 4.9989403106865765e-06,
+      "loss": 1.0414,
+      "step": 20
+    },
+    {
+      "epoch": 0.1024390243902439,
+      "grad_norm": 3.4605791568756104,
+      "learning_rate": 4.9988258379777334e-06,
+      "loss": 0.8878,
+      "step": 21
+    },
+    {
+      "epoch": 0.1073170731707317,
+      "grad_norm": 2.860478639602661,
+      "learning_rate": 4.998705496757846e-06,
+      "loss": 0.9151,
+      "step": 22
+    },
+    {
+      "epoch": 0.11219512195121951,
+      "grad_norm": 9.101946830749512,
+      "learning_rate": 4.998579287309538e-06,
+      "loss": 1.4304,
+      "step": 23
+    },
+    {
+      "epoch": 0.11707317073170732,
+      "grad_norm": 24.21122169494629,
+      "learning_rate": 4.998447209929211e-06,
+      "loss": 1.0858,
+      "step": 24
+    },
+    {
+      "epoch": 0.12195121951219512,
+      "grad_norm": 3.286980152130127,
+      "learning_rate": 4.998309264927053e-06,
+      "loss": 0.6571,
+      "step": 25
+    },
+    {
+      "epoch": 0.12682926829268293,
+      "grad_norm": 4.0232062339782715,
+      "learning_rate": 4.998165452627025e-06,
+      "loss": 0.8493,
+      "step": 26
+    },
+    {
+      "epoch": 0.13170731707317074,
+      "grad_norm": 3.7688663005828857,
+      "learning_rate": 4.998015773366874e-06,
+      "loss": 0.9224,
+      "step": 27
+    },
+    {
+      "epoch": 0.13658536585365855,
+      "grad_norm": 2.9382026195526123,
+      "learning_rate": 4.997860227498122e-06,
+      "loss": 0.7588,
+      "step": 28
+    },
+    {
+      "epoch": 0.14146341463414633,
+      "grad_norm": 4.327457904815674,
+      "learning_rate": 4.99769881538607e-06,
+      "loss": 1.1817,
+      "step": 29
+    },
+    {
+      "epoch": 0.14634146341463414,
+      "grad_norm": 3.47487735748291,
+      "learning_rate": 4.997531537409794e-06,
+      "loss": 1.0737,
+      "step": 30
+    },
+    {
+      "epoch": 0.15121951219512195,
+      "grad_norm": 3.0616214275360107,
+      "learning_rate": 4.99735839396215e-06,
+      "loss": 0.7899,
+      "step": 31
+    },
+    {
+      "epoch": 0.15609756097560976,
+      "grad_norm": 3.065070152282715,
+      "learning_rate": 4.9971793854497655e-06,
+      "loss": 0.7745,
+      "step": 32
+    },
+    {
+      "epoch": 0.16097560975609757,
+      "grad_norm": 3.5202279090881348,
+      "learning_rate": 4.996994512293042e-06,
+      "loss": 0.984,
+      "step": 33
+    },
+    {
+      "epoch": 0.16585365853658537,
+      "grad_norm": 3.421769142150879,
+      "learning_rate": 4.996803774926157e-06,
+      "loss": 0.8235,
+      "step": 34
+    },
+    {
+      "epoch": 0.17073170731707318,
+      "grad_norm": 4.6582207679748535,
+      "learning_rate": 4.996607173797059e-06,
+      "loss": 1.3227,
+      "step": 35
+    },
+    {
+      "epoch": 0.17560975609756097,
+      "grad_norm": 2.9829282760620117,
+      "learning_rate": 4.996404709367466e-06,
+      "loss": 0.8854,
+      "step": 36
+    },
+    {
+      "epoch": 0.18048780487804877,
+      "grad_norm": 2.5982632637023926,
+      "learning_rate": 4.996196382112868e-06,
+      "loss": 0.6786,
+      "step": 37
+    },
+    {
+      "epoch": 0.18536585365853658,
+      "grad_norm": 2.9807393550872803,
+      "learning_rate": 4.9959821925225235e-06,
+      "loss": 0.9344,
+      "step": 38
+    },
+    {
+      "epoch": 0.1902439024390244,
+      "grad_norm": 2.7364351749420166,
+      "learning_rate": 4.995762141099456e-06,
+      "loss": 0.814,
+      "step": 39
+    },
+    {
+      "epoch": 0.1951219512195122,
+      "grad_norm": 3.4324638843536377,
+      "learning_rate": 4.995536228360461e-06,
+      "loss": 1.0276,
+      "step": 40
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 2.911834716796875,
+      "learning_rate": 4.995304454836095e-06,
+      "loss": 0.9291,
+      "step": 41
+    },
+    {
+      "epoch": 0.2048780487804878,
+      "grad_norm": 3.0294723510742188,
+      "learning_rate": 4.9950668210706795e-06,
+      "loss": 0.8145,
+      "step": 42
+    },
+    {
+      "epoch": 0.2097560975609756,
+      "grad_norm": 4.681829452514648,
+      "learning_rate": 4.994823327622299e-06,
+      "loss": 0.8779,
+      "step": 43
+    },
+    {
+      "epoch": 0.2146341463414634,
+      "grad_norm": 3.643914222717285,
+      "learning_rate": 4.9945739750628e-06,
+      "loss": 0.8196,
+      "step": 44
+    },
+    {
+      "epoch": 0.21951219512195122,
+      "grad_norm": 2.7542076110839844,
+      "learning_rate": 4.994318763977789e-06,
+      "loss": 0.8443,
+      "step": 45
+    },
+    {
+      "epoch": 0.22439024390243903,
+      "grad_norm": 6.873605728149414,
+      "learning_rate": 4.994057694966632e-06,
+      "loss": 1.0328,
+      "step": 46
+    },
+    {
+      "epoch": 0.22926829268292684,
+      "grad_norm": 3.11810040473938,
+      "learning_rate": 4.993790768642449e-06,
+      "loss": 1.0673,
+      "step": 47
+    },
+    {
+      "epoch": 0.23414634146341465,
+      "grad_norm": 4.360548496246338,
+      "learning_rate": 4.99351798563212e-06,
+      "loss": 1.3198,
+      "step": 48
+    },
+    {
+      "epoch": 0.23902439024390243,
+      "grad_norm": 2.6894314289093018,
+      "learning_rate": 4.993239346576278e-06,
+      "loss": 0.8743,
+      "step": 49
+    },
+    {
+      "epoch": 0.24390243902439024,
+      "grad_norm": 3.2640421390533447,
+      "learning_rate": 4.99295485212931e-06,
+      "loss": 1.109,
+      "step": 50
+    },
+    {
+      "epoch": 0.24878048780487805,
+      "grad_norm": 3.1565866470336914,
+      "learning_rate": 4.992664502959351e-06,
+      "loss": 0.9291,
+      "step": 51
+    },
+    {
+      "epoch": 0.25365853658536586,
+      "grad_norm": 3.4829447269439697,
+      "learning_rate": 4.99236829974829e-06,
+      "loss": 0.8159,
+      "step": 52
+    },
+    {
+      "epoch": 0.25853658536585367,
+      "grad_norm": 2.7535626888275146,
+      "learning_rate": 4.992066243191762e-06,
+      "loss": 1.0359,
+      "step": 53
+    },
+    {
+      "epoch": 0.2634146341463415,
+      "grad_norm": 2.482935905456543,
+      "learning_rate": 4.991758333999148e-06,
+      "loss": 0.8091,
+      "step": 54
+    },
+    {
+      "epoch": 0.2682926829268293,
+      "grad_norm": 2.917445659637451,
+      "learning_rate": 4.991444572893575e-06,
+      "loss": 0.6925,
+      "step": 55
+    },
+    {
+      "epoch": 0.2731707317073171,
+      "grad_norm": 2.9802236557006836,
+      "learning_rate": 4.991124960611916e-06,
+      "loss": 0.6329,
+      "step": 56
+    },
+    {
+      "epoch": 0.2780487804878049,
+      "grad_norm": 2.9677224159240723,
+      "learning_rate": 4.99079949790478e-06,
+      "loss": 0.8069,
+      "step": 57
+    },
+    {
+      "epoch": 0.28292682926829266,
+      "grad_norm": 2.8304293155670166,
+      "learning_rate": 4.99046818553652e-06,
+      "loss": 0.8682,
+      "step": 58
+    },
+    {
+      "epoch": 0.28780487804878047,
+      "grad_norm": 5.253443717956543,
+      "learning_rate": 4.9901310242852246e-06,
+      "loss": 1.1069,
+      "step": 59
+    },
+    {
+      "epoch": 0.2926829268292683,
+      "grad_norm": 3.686016082763672,
+      "learning_rate": 4.9897880149427206e-06,
+      "loss": 0.9465,
+      "step": 60
+    },
+    {
+      "epoch": 0.2975609756097561,
+      "grad_norm": 3.6372263431549072,
+      "learning_rate": 4.989439158314566e-06,
+      "loss": 0.9738,
+      "step": 61
+    },
+    {
+      "epoch": 0.3024390243902439,
+      "grad_norm": 3.0756819248199463,
+      "learning_rate": 4.989084455220056e-06,
+      "loss": 0.6417,
+      "step": 62
+    },
+    {
+      "epoch": 0.3073170731707317,
+      "grad_norm": 3.379222869873047,
+      "learning_rate": 4.988723906492212e-06,
+      "loss": 1.0092,
+      "step": 63
+    },
+    {
+      "epoch": 0.3121951219512195,
+      "grad_norm": 3.4571032524108887,
+      "learning_rate": 4.988357512977785e-06,
+      "loss": 0.6691,
+      "step": 64
+    },
+    {
+      "epoch": 0.3170731707317073,
+      "grad_norm": 3.1982104778289795,
+      "learning_rate": 4.987985275537252e-06,
+      "loss": 0.6651,
+      "step": 65
+    },
+    {
+      "epoch": 0.32195121951219513,
+      "grad_norm": 2.9723124504089355,
+      "learning_rate": 4.9876071950448185e-06,
+      "loss": 0.9227,
+      "step": 66
+    },
+    {
+      "epoch": 0.32682926829268294,
+      "grad_norm": 2.5521399974823,
+      "learning_rate": 4.987223272388407e-06,
+      "loss": 0.6664,
+      "step": 67
+    },
+    {
+      "epoch": 0.33170731707317075,
+      "grad_norm": 2.8934121131896973,
+      "learning_rate": 4.986833508469663e-06,
+      "loss": 0.997,
+      "step": 68
+    },
+    {
+      "epoch": 0.33658536585365856,
+      "grad_norm": 4.7546586990356445,
+      "learning_rate": 4.98643790420395e-06,
+      "loss": 0.8551,
+      "step": 69
+    },
+    {
+      "epoch": 0.34146341463414637,
+      "grad_norm": 3.091616153717041,
+      "learning_rate": 4.986036460520348e-06,
+      "loss": 0.8874,
+      "step": 70
+    },
+    {
+      "epoch": 0.3463414634146341,
+      "grad_norm": 4.1724677085876465,
+      "learning_rate": 4.98562917836165e-06,
+      "loss": 1.1393,
+      "step": 71
+    },
+    {
+      "epoch": 0.35121951219512193,
+      "grad_norm": 2.6568572521209717,
+      "learning_rate": 4.985216058684362e-06,
+      "loss": 0.6379,
+      "step": 72
+    },
+    {
+      "epoch": 0.35609756097560974,
+      "grad_norm": 2.396416187286377,
+      "learning_rate": 4.984797102458697e-06,
+      "loss": 1.0292,
+      "step": 73
+    },
+    {
+      "epoch": 0.36097560975609755,
+      "grad_norm": 3.0667319297790527,
+      "learning_rate": 4.984372310668579e-06,
+      "loss": 0.7048,
+      "step": 74
+    },
+    {
+      "epoch": 0.36585365853658536,
+      "grad_norm": 2.4820518493652344,
+      "learning_rate": 4.983941684311633e-06,
+      "loss": 1.2353,
+      "step": 75
+    },
+    {
+      "epoch": 0.37073170731707317,
+      "grad_norm": 4.062836647033691,
+      "learning_rate": 4.983505224399188e-06,
+      "loss": 0.8933,
+      "step": 76
+    },
+    {
+      "epoch": 0.375609756097561,
+      "grad_norm": 2.4480767250061035,
+      "learning_rate": 4.983062931956275e-06,
+      "loss": 0.8221,
+      "step": 77
+    },
+    {
+      "epoch": 0.3804878048780488,
+      "grad_norm": 3.134138822555542,
+      "learning_rate": 4.9826148080216195e-06,
+      "loss": 0.8899,
+      "step": 78
+    },
+    {
+      "epoch": 0.3853658536585366,
+      "grad_norm": 2.8165836334228516,
+      "learning_rate": 4.9821608536476445e-06,
+      "loss": 1.2451,
+      "step": 79
+    },
+    {
+      "epoch": 0.3902439024390244,
+      "grad_norm": 3.734433650970459,
+      "learning_rate": 4.981701069900465e-06,
+      "loss": 0.8536,
+      "step": 80
+    },
+    {
+      "epoch": 0.3951219512195122,
+      "grad_norm": 2.853421449661255,
+      "learning_rate": 4.9812354578598876e-06,
+      "loss": 0.7857,
+      "step": 81
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 2.541687250137329,
+      "learning_rate": 4.980764018619405e-06,
+      "loss": 0.8332,
+      "step": 82
+    },
+    {
+      "epoch": 0.40487804878048783,
+      "grad_norm": 4.405911445617676,
+      "learning_rate": 4.980286753286196e-06,
+      "loss": 0.9927,
+      "step": 83
+    },
+    {
+      "epoch": 0.4097560975609756,
+      "grad_norm": 3.3034985065460205,
+      "learning_rate": 4.97980366298112e-06,
+      "loss": 0.8161,
+      "step": 84
+    },
+    {
+      "epoch": 0.4146341463414634,
+      "grad_norm": 2.6678085327148438,
+      "learning_rate": 4.97931474883872e-06,
+      "loss": 0.8017,
+      "step": 85
+    },
+    {
+      "epoch": 0.4195121951219512,
+      "grad_norm": 2.58524227142334,
+      "learning_rate": 4.978820012007213e-06,
+      "loss": 0.8811,
+      "step": 86
+    },
+    {
+      "epoch": 0.424390243902439,
+      "grad_norm": 2.482597827911377,
+      "learning_rate": 4.978319453648495e-06,
+      "loss": 0.9461,
+      "step": 87
+    },
+    {
+      "epoch": 0.4292682926829268,
+      "grad_norm": 2.5731301307678223,
+      "learning_rate": 4.977813074938128e-06,
+      "loss": 0.8835,
+      "step": 88
+    },
+    {
+      "epoch": 0.43414634146341463,
+      "grad_norm": 2.7914488315582275,
+      "learning_rate": 4.977300877065347e-06,
+      "loss": 0.8466,
+      "step": 89
+    },
+    {
+      "epoch": 0.43902439024390244,
+      "grad_norm": 2.416043758392334,
+      "learning_rate": 4.976782861233053e-06,
+      "loss": 0.7132,
+      "step": 90
+    },
+    {
+      "epoch": 0.44390243902439025,
+      "grad_norm": 3.7616264820098877,
+      "learning_rate": 4.976259028657812e-06,
+      "loss": 0.7639,
+      "step": 91
+    },
+    {
+      "epoch": 0.44878048780487806,
+      "grad_norm": 2.6081621646881104,
+      "learning_rate": 4.975729380569845e-06,
+      "loss": 0.8055,
+      "step": 92
+    },
+    {
+      "epoch": 0.45365853658536587,
+      "grad_norm": 3.3343570232391357,
+      "learning_rate": 4.975193918213035e-06,
+      "loss": 0.6042,
+      "step": 93
+    },
+    {
+      "epoch": 0.4585365853658537,
+      "grad_norm": 2.517544746398926,
+      "learning_rate": 4.974652642844921e-06,
+      "loss": 0.7672,
+      "step": 94
+    },
+    {
+      "epoch": 0.4634146341463415,
+      "grad_norm": 4.173468589782715,
+      "learning_rate": 4.974105555736693e-06,
+      "loss": 1.0682,
+      "step": 95
+    },
+    {
+      "epoch": 0.4682926829268293,
+      "grad_norm": 2.8422317504882812,
+      "learning_rate": 4.973552658173186e-06,
+      "loss": 0.7841,
+      "step": 96
+    },
+    {
+      "epoch": 0.47317073170731705,
+      "grad_norm": 5.042182445526123,
+      "learning_rate": 4.972993951452887e-06,
+      "loss": 0.8851,
+      "step": 97
+    },
+    {
+      "epoch": 0.47804878048780486,
+      "grad_norm": 5.977590560913086,
+      "learning_rate": 4.9724294368879214e-06,
+      "loss": 0.9059,
+      "step": 98
+    },
+    {
+      "epoch": 0.48292682926829267,
+      "grad_norm": 4.227641582489014,
+      "learning_rate": 4.971859115804055e-06,
+      "loss": 1.0152,
+      "step": 99
+    },
+    {
+      "epoch": 0.4878048780487805,
+      "grad_norm": 3.180952548980713,
+      "learning_rate": 4.9712829895406935e-06,
+      "loss": 0.8092,
+      "step": 100
+    },
+    {
+      "epoch": 0.4926829268292683,
+      "grad_norm": 11.220394134521484,
+      "learning_rate": 4.970701059450872e-06,
+      "loss": 0.8239,
+      "step": 101
+    },
+    {
+      "epoch": 0.4975609756097561,
+      "grad_norm": 2.346975088119507,
+      "learning_rate": 4.970113326901258e-06,
+      "loss": 0.9283,
+      "step": 102
+    },
+    {
+      "epoch": 0.5024390243902439,
+      "grad_norm": 2.9470982551574707,
+      "learning_rate": 4.9695197932721455e-06,
+      "loss": 0.9429,
+      "step": 103
+    },
+    {
+      "epoch": 0.5073170731707317,
+      "grad_norm": 3.6048219203948975,
+      "learning_rate": 4.968920459957453e-06,
+      "loss": 0.9231,
+      "step": 104
+    },
+    {
+      "epoch": 0.5121951219512195,
+      "grad_norm": 2.8181886672973633,
+      "learning_rate": 4.968315328364719e-06,
+      "loss": 1.0005,
+      "step": 105
+    },
+    {
+      "epoch": 0.5170731707317073,
+      "grad_norm": 3.114147424697876,
+      "learning_rate": 4.9677043999151e-06,
+      "loss": 1.1326,
+      "step": 106
+    },
+    {
+      "epoch": 0.5219512195121951,
+      "grad_norm": 2.965885639190674,
+      "learning_rate": 4.967087676043366e-06,
+      "loss": 0.541,
+      "step": 107
+    },
+    {
+      "epoch": 0.526829268292683,
+      "grad_norm": 3.098677635192871,
+      "learning_rate": 4.966465158197897e-06,
+      "loss": 0.9473,
+      "step": 108
+    },
+    {
+      "epoch": 0.5317073170731708,
+      "grad_norm": 2.8640191555023193,
+      "learning_rate": 4.965836847840681e-06,
+      "loss": 0.6678,
+      "step": 109
+    },
+    {
+      "epoch": 0.5365853658536586,
+      "grad_norm": 3.0950934886932373,
+      "learning_rate": 4.96520274644731e-06,
+      "loss": 0.9251,
+      "step": 110
+    },
+    {
+      "epoch": 0.5414634146341464,
+      "grad_norm": 2.99444317817688,
+      "learning_rate": 4.964562855506976e-06,
+      "loss": 0.7807,
+      "step": 111
+    },
+    {
+      "epoch": 0.5463414634146342,
+      "grad_norm": 2.348639726638794,
+      "learning_rate": 4.963917176522466e-06,
+      "loss": 0.6395,
+      "step": 112
+    },
+    {
+      "epoch": 0.551219512195122,
+      "grad_norm": 3.5988354682922363,
+      "learning_rate": 4.963265711010164e-06,
+      "loss": 1.0658,
+      "step": 113
+    },
+    {
+      "epoch": 0.5560975609756098,
+      "grad_norm": 3.3423564434051514,
+      "learning_rate": 4.9626084605000395e-06,
+      "loss": 0.8974,
+      "step": 114
+    },
+    {
+      "epoch": 0.5609756097560976,
+      "grad_norm": 2.8353331089019775,
+      "learning_rate": 4.961945426535652e-06,
+      "loss": 0.6144,
+      "step": 115
+    },
+    {
+      "epoch": 0.5658536585365853,
+      "grad_norm": 2.752387046813965,
+      "learning_rate": 4.961276610674141e-06,
+      "loss": 0.9083,
+      "step": 116
+    },
+    {
+      "epoch": 0.5707317073170731,
+      "grad_norm": 2.2654404640197754,
+      "learning_rate": 4.960602014486225e-06,
+      "loss": 1.0101,
+      "step": 117
+    },
+    {
+      "epoch": 0.5756097560975609,
+      "grad_norm": 3.344377040863037,
+      "learning_rate": 4.959921639556199e-06,
+      "loss": 0.8391,
+      "step": 118
+    },
+    {
+      "epoch": 0.5804878048780487,
+      "grad_norm": 3.1620500087738037,
+      "learning_rate": 4.959235487481928e-06,
+      "loss": 1.0431,
+      "step": 119
+    },
+    {
+      "epoch": 0.5853658536585366,
+      "grad_norm": 2.857048273086548,
+      "learning_rate": 4.958543559874846e-06,
+      "loss": 0.5864,
+      "step": 120
+    },
+    {
+      "epoch": 0.5902439024390244,
+      "grad_norm": 3.1736063957214355,
+      "learning_rate": 4.9578458583599495e-06,
+      "loss": 0.7868,
+      "step": 121
+    },
+    {
+      "epoch": 0.5951219512195122,
+      "grad_norm": 3.5520827770233154,
+      "learning_rate": 4.957142384575795e-06,
+      "loss": 0.7901,
+      "step": 122
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 3.265103578567505,
+      "learning_rate": 4.956433140174498e-06,
+      "loss": 0.9067,
+      "step": 123
+    },
+    {
+      "epoch": 0.6048780487804878,
+      "grad_norm": 3.1181187629699707,
+      "learning_rate": 4.9557181268217225e-06,
+      "loss": 0.8971,
+      "step": 124
+    },
+    {
+      "epoch": 0.6097560975609756,
+      "grad_norm": 2.4123694896698,
+      "learning_rate": 4.954997346196683e-06,
+      "loss": 1.2123,
+      "step": 125
+    },
+    {
+      "epoch": 0.6146341463414634,
+      "grad_norm": 2.9646875858306885,
+      "learning_rate": 4.954270799992138e-06,
+      "loss": 0.7696,
+      "step": 126
+    },
+    {
+      "epoch": 0.6195121951219512,
+      "grad_norm": 2.7457995414733887,
+      "learning_rate": 4.953538489914387e-06,
+      "loss": 0.7919,
+      "step": 127
+    },
+    {
+      "epoch": 0.624390243902439,
+      "grad_norm": 5.096850395202637,
+      "learning_rate": 4.9528004176832654e-06,
+      "loss": 0.6494,
+      "step": 128
+    },
+    {
+      "epoch": 0.6292682926829268,
+      "grad_norm": 3.124955177307129,
+      "learning_rate": 4.952056585032142e-06,
+      "loss": 1.0546,
+      "step": 129
+    },
+    {
+      "epoch": 0.6341463414634146,
+      "grad_norm": 2.4860167503356934,
+      "learning_rate": 4.951306993707913e-06,
+      "loss": 0.7907,
+      "step": 130
+    },
+    {
+      "epoch": 0.6390243902439025,
+      "grad_norm": 2.3380239009857178,
+      "learning_rate": 4.950551645470998e-06,
+      "loss": 0.7433,
+      "step": 131
+    },
+    {
+      "epoch": 0.6439024390243903,
+      "grad_norm": 2.8945236206054688,
+      "learning_rate": 4.9497905420953406e-06,
+      "loss": 0.7682,
+      "step": 132
+    },
+    {
+      "epoch": 0.6487804878048781,
+      "grad_norm": 3.429776430130005,
+      "learning_rate": 4.949023685368395e-06,
+      "loss": 0.8411,
+      "step": 133
+    },
+    {
+      "epoch": 0.6536585365853659,
+      "grad_norm": 2.8853516578674316,
+      "learning_rate": 4.948251077091131e-06,
+      "loss": 1.0792,
+      "step": 134
+    },
+    {
+      "epoch": 0.6585365853658537,
+      "grad_norm": 2.145598888397217,
+      "learning_rate": 4.947472719078025e-06,
+      "loss": 0.8033,
+      "step": 135
+    },
+    {
+      "epoch": 0.6634146341463415,
+      "grad_norm": 2.5064377784729004,
+      "learning_rate": 4.9466886131570565e-06,
+      "loss": 0.939,
+      "step": 136
+    },
+    {
+      "epoch": 0.6682926829268293,
+      "grad_norm": 2.5700225830078125,
+      "learning_rate": 4.945898761169704e-06,
+      "loss": 1.0418,
+      "step": 137
+    },
+    {
+      "epoch": 0.6731707317073171,
+      "grad_norm": 2.3390917778015137,
+      "learning_rate": 4.945103164970941e-06,
+      "loss": 0.6158,
+      "step": 138
+    },
+    {
+      "epoch": 0.6780487804878049,
+      "grad_norm": 2.1538751125335693,
+      "learning_rate": 4.9443018264292304e-06,
+      "loss": 0.6995,
+      "step": 139
+    },
+    {
+      "epoch": 0.6829268292682927,
+      "grad_norm": 5.255710601806641,
+      "learning_rate": 4.9434947474265225e-06,
+      "loss": 1.0382,
+      "step": 140
+    },
+    {
+      "epoch": 0.6878048780487804,
+      "grad_norm": 2.5547356605529785,
+      "learning_rate": 4.942681929858249e-06,
+      "loss": 1.037,
+      "step": 141
+    },
+    {
+      "epoch": 0.6926829268292682,
+      "grad_norm": 2.613280773162842,
+      "learning_rate": 4.941863375633315e-06,
+      "loss": 0.9071,
+      "step": 142
+    },
+    {
+      "epoch": 0.697560975609756,
+      "grad_norm": 2.9957327842712402,
+      "learning_rate": 4.9410390866741056e-06,
+      "loss": 0.7908,
+      "step": 143
+    },
+    {
+      "epoch": 0.7024390243902439,
+      "grad_norm": 2.410107374191284,
+      "learning_rate": 4.9402090649164655e-06,
+      "loss": 0.7739,
+      "step": 144
+    },
+    {
+      "epoch": 0.7073170731707317,
+      "grad_norm": 2.352013349533081,
+      "learning_rate": 4.9393733123097085e-06,
+      "loss": 0.939,
+      "step": 145
+    },
+    {
+      "epoch": 0.7121951219512195,
+      "grad_norm": 2.5164194107055664,
+      "learning_rate": 4.9385318308166065e-06,
+      "loss": 0.8729,
+      "step": 146
+    },
+    {
+      "epoch": 0.7170731707317073,
+      "grad_norm": 4.213881015777588,
+      "learning_rate": 4.937684622413385e-06,
+      "loss": 0.6124,
+      "step": 147
+    },
+    {
+      "epoch": 0.7219512195121951,
+      "grad_norm": 2.7950191497802734,
+      "learning_rate": 4.9368316890897185e-06,
+      "loss": 0.975,
+      "step": 148
+    },
+    {
+      "epoch": 0.7268292682926829,
+      "grad_norm": 2.8618874549865723,
+      "learning_rate": 4.9359730328487264e-06,
+      "loss": 0.5832,
+      "step": 149
+    },
+    {
+      "epoch": 0.7317073170731707,
+      "grad_norm": 2.6943812370300293,
+      "learning_rate": 4.935108655706972e-06,
+      "loss": 0.8124,
+      "step": 150
+    },
+    {
+      "epoch": 0.7365853658536585,
+      "grad_norm": 3.2164082527160645,
+      "learning_rate": 4.934238559694448e-06,
+      "loss": 1.1446,
+      "step": 151
+    },
+    {
+      "epoch": 0.7414634146341463,
+      "grad_norm": 3.05002498626709,
+      "learning_rate": 4.9333627468545845e-06,
+      "loss": 0.7884,
+      "step": 152
+    },
+    {
+      "epoch": 0.7463414634146341,
+      "grad_norm": 2.863351583480835,
+      "learning_rate": 4.932481219244231e-06,
+      "loss": 0.7918,
+      "step": 153
+    },
+    {
+      "epoch": 0.751219512195122,
+      "grad_norm": 2.4947102069854736,
+      "learning_rate": 4.931593978933666e-06,
+      "loss": 0.775,
+      "step": 154
+    },
+    {
+      "epoch": 0.7560975609756098,
+      "grad_norm": 2.918886184692383,
+      "learning_rate": 4.930701028006577e-06,
+      "loss": 0.993,
+      "step": 155
+    },
+    {
+      "epoch": 0.7609756097560976,
+      "grad_norm": 2.835956573486328,
+      "learning_rate": 4.929802368560066e-06,
+      "loss": 0.7911,
+      "step": 156
+    },
+    {
+      "epoch": 0.7658536585365854,
+      "grad_norm": 3.3073575496673584,
+      "learning_rate": 4.928898002704642e-06,
+      "loss": 0.9346,
+      "step": 157
+    },
+    {
+      "epoch": 0.7707317073170732,
+      "grad_norm": 3.086146354675293,
+      "learning_rate": 4.927987932564215e-06,
+      "loss": 0.817,
+      "step": 158
+    },
+    {
+      "epoch": 0.775609756097561,
+      "grad_norm": 2.5419743061065674,
+      "learning_rate": 4.927072160276092e-06,
+      "loss": 0.7918,
+      "step": 159
+    },
+    {
+      "epoch": 0.7804878048780488,
+      "grad_norm": 3.984297275543213,
+      "learning_rate": 4.926150687990969e-06,
+      "loss": 0.7153,
+      "step": 160
+    },
+    {
+      "epoch": 0.7853658536585366,
+      "grad_norm": 2.4703335762023926,
+      "learning_rate": 4.925223517872934e-06,
+      "loss": 0.8982,
+      "step": 161
+    },
+    {
+      "epoch": 0.7902439024390244,
+      "grad_norm": 2.81785249710083,
+      "learning_rate": 4.9242906520994484e-06,
+      "loss": 0.9839,
+      "step": 162
+    },
+    {
+      "epoch": 0.7951219512195122,
+      "grad_norm": 2.3304924964904785,
+      "learning_rate": 4.923352092861358e-06,
+      "loss": 0.8406,
+      "step": 163
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 2.339498519897461,
+      "learning_rate": 4.922407842362875e-06,
+      "loss": 0.6602,
+      "step": 164
+    },
+    {
+      "epoch": 0.8048780487804879,
+      "grad_norm": 3.488255262374878,
+      "learning_rate": 4.921457902821578e-06,
+      "loss": 0.9779,
+      "step": 165
+    },
+    {
+      "epoch": 0.8097560975609757,
+      "grad_norm": 2.8528945446014404,
+      "learning_rate": 4.920502276468408e-06,
+      "loss": 0.8821,
+      "step": 166
+    },
+    {
+      "epoch": 0.8146341463414634,
+      "grad_norm": 3.4649784564971924,
+      "learning_rate": 4.9195409655476605e-06,
+      "loss": 0.7539,
+      "step": 167
+    },
+    {
+      "epoch": 0.8195121951219512,
+      "grad_norm": 2.3109042644500732,
+      "learning_rate": 4.918573972316982e-06,
+      "loss": 0.9807,
+      "step": 168
+    },
+    {
+      "epoch": 0.824390243902439,
+      "grad_norm": 2.678666353225708,
+      "learning_rate": 4.917601299047361e-06,
+      "loss": 0.8318,
+      "step": 169
+    },
+    {
+      "epoch": 0.8292682926829268,
+      "grad_norm": 2.730614185333252,
+      "learning_rate": 4.916622948023129e-06,
+      "loss": 0.7816,
+      "step": 170
+    },
+    {
+      "epoch": 0.8341463414634146,
+      "grad_norm": 2.9835665225982666,
+      "learning_rate": 4.915638921541952e-06,
+      "loss": 0.6633,
+      "step": 171
+    },
+    {
+      "epoch": 0.8390243902439024,
+      "grad_norm": 3.31217360496521,
+      "learning_rate": 4.914649221914822e-06,
+      "loss": 0.9296,
+      "step": 172
+    },
+    {
+      "epoch": 0.8439024390243902,
+      "grad_norm": 2.9021658897399902,
+      "learning_rate": 4.913653851466057e-06,
+      "loss": 0.6864,
+      "step": 173
+    },
+    {
+      "epoch": 0.848780487804878,
+      "grad_norm": 3.3672914505004883,
+      "learning_rate": 4.912652812533291e-06,
+      "loss": 0.8599,
+      "step": 174
+    },
+    {
+      "epoch": 0.8536585365853658,
+      "grad_norm": 2.4871644973754883,
+      "learning_rate": 4.911646107467472e-06,
+      "loss": 0.8949,
+      "step": 175
+    },
+    {
+      "epoch": 0.8585365853658536,
+      "grad_norm": 2.728022813796997,
+      "learning_rate": 4.9106337386328524e-06,
+      "loss": 0.9758,
+      "step": 176
+    },
+    {
+      "epoch": 0.8634146341463415,
+      "grad_norm": 2.704252243041992,
+      "learning_rate": 4.909615708406991e-06,
+      "loss": 0.8954,
+      "step": 177
+    },
+    {
+      "epoch": 0.8682926829268293,
+      "grad_norm": 2.4002223014831543,
+      "learning_rate": 4.908592019180738e-06,
+      "loss": 0.7157,
+      "step": 178
+    },
+    {
+      "epoch": 0.8731707317073171,
+      "grad_norm": 2.1927788257598877,
+      "learning_rate": 4.907562673358234e-06,
+      "loss": 0.6358,
+      "step": 179
+    },
+    {
+      "epoch": 0.8780487804878049,
+      "grad_norm": 2.458500623703003,
+      "learning_rate": 4.906527673356907e-06,
+      "loss": 0.6685,
+      "step": 180
+    },
+    {
+      "epoch": 0.8829268292682927,
+      "grad_norm": 2.5924787521362305,
+      "learning_rate": 4.905487021607462e-06,
+      "loss": 0.5686,
+      "step": 181
+    },
+    {
+      "epoch": 0.8878048780487805,
+      "grad_norm": 3.0923380851745605,
+      "learning_rate": 4.904440720553876e-06,
+      "loss": 0.8538,
+      "step": 182
+    },
+    {
+      "epoch": 0.8926829268292683,
+      "grad_norm": 2.8001527786254883,
+      "learning_rate": 4.903388772653396e-06,
+      "loss": 0.8292,
+      "step": 183
+    },
+    {
+      "epoch": 0.8975609756097561,
+      "grad_norm": 2.4344072341918945,
+      "learning_rate": 4.902331180376529e-06,
+      "loss": 0.7946,
+      "step": 184
+    },
+    {
+      "epoch": 0.9024390243902439,
+      "grad_norm": 2.6313226222991943,
+      "learning_rate": 4.901267946207038e-06,
+      "loss": 0.9269,
+      "step": 185
+    },
+    {
+      "epoch": 0.9073170731707317,
+      "grad_norm": 2.4776692390441895,
+      "learning_rate": 4.900199072641937e-06,
+      "loss": 0.7433,
+      "step": 186
+    },
+    {
+      "epoch": 0.9121951219512195,
+      "grad_norm": 2.339869260787964,
+      "learning_rate": 4.899124562191484e-06,
+      "loss": 0.6577,
+      "step": 187
+    },
+    {
+      "epoch": 0.9170731707317074,
+      "grad_norm": 3.076890468597412,
+      "learning_rate": 4.8980444173791735e-06,
+      "loss": 0.5989,
+      "step": 188
+    },
+    {
+      "epoch": 0.9219512195121952,
+      "grad_norm": 2.83957839012146,
+      "learning_rate": 4.896958640741735e-06,
+      "loss": 0.9364,
+      "step": 189
+    },
+    {
+      "epoch": 0.926829268292683,
+      "grad_norm": 2.770867347717285,
+      "learning_rate": 4.895867234829121e-06,
+      "loss": 1.0328,
+      "step": 190
+    },
+    {
+      "epoch": 0.9317073170731708,
+      "grad_norm": 2.7819619178771973,
+      "learning_rate": 4.894770202204509e-06,
+      "loss": 0.772,
+      "step": 191
+    },
+    {
+      "epoch": 0.9365853658536586,
+      "grad_norm": 3.925703763961792,
+      "learning_rate": 4.893667545444285e-06,
+      "loss": 0.8128,
+      "step": 192
+    },
+    {
+      "epoch": 0.9414634146341463,
+      "grad_norm": 3.034944534301758,
+      "learning_rate": 4.8925592671380495e-06,
+      "loss": 0.7418,
+      "step": 193
+    },
+    {
+      "epoch": 0.9463414634146341,
+      "grad_norm": 2.3350143432617188,
+      "learning_rate": 4.891445369888601e-06,
+      "loss": 0.5979,
+      "step": 194
+    },
+    {
+      "epoch": 0.9512195121951219,
+      "grad_norm": 2.6433160305023193,
+      "learning_rate": 4.890325856311936e-06,
+      "loss": 0.9664,
+      "step": 195
+    },
+    {
+      "epoch": 0.9560975609756097,
+      "grad_norm": 2.715142011642456,
+      "learning_rate": 4.889200729037241e-06,
+      "loss": 0.8482,
+      "step": 196
+    },
+    {
+      "epoch": 0.9609756097560975,
+      "grad_norm": 2.6157352924346924,
+      "learning_rate": 4.888069990706884e-06,
+      "loss": 0.7173,
+      "step": 197
+    },
+    {
+      "epoch": 0.9658536585365853,
+      "grad_norm": 3.7308952808380127,
+      "learning_rate": 4.886933643976414e-06,
+      "loss": 0.5433,
+      "step": 198
+    },
+    {
+      "epoch": 0.9707317073170731,
+      "grad_norm": 3.1134045124053955,
+      "learning_rate": 4.885791691514548e-06,
+      "loss": 0.5997,
+      "step": 199
+    },
+    {
+      "epoch": 0.975609756097561,
+      "grad_norm": 2.421365976333618,
+      "learning_rate": 4.884644136003172e-06,
+      "loss": 0.6477,
+      "step": 200
+    },
+    {
+      "epoch": 0.9804878048780488,
+      "grad_norm": 2.8676180839538574,
+      "learning_rate": 4.883490980137327e-06,
+      "loss": 1.3465,
+      "step": 201
+    },
+    {
+      "epoch": 0.9853658536585366,
+      "grad_norm": 2.236189603805542,
+      "learning_rate": 4.882332226625208e-06,
+      "loss": 0.7533,
+      "step": 202
+    },
+    {
+      "epoch": 0.9902439024390244,
+      "grad_norm": 2.2514970302581787,
+      "learning_rate": 4.881167878188158e-06,
+      "loss": 0.8555,
+      "step": 203
+    },
+    {
+      "epoch": 0.9951219512195122,
+      "grad_norm": 2.6856095790863037,
+      "learning_rate": 4.8799979375606565e-06,
+      "loss": 0.7634,
+      "step": 204
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 2.5563852787017822,
+      "learning_rate": 4.878822407490319e-06,
+      "loss": 0.66,
+      "step": 205
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 2050,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 206,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.897858756221338e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}