Upload folder using huggingface_hub

added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "/apdcephfs/share_302520718/share_info/llm_models/Qwen2-7B-Instruct",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 128245,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 3584,
+  "initializer_range": 0.02,
+  "intermediate_size": 18944,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 28,
+  "model_type": "qwen2",
+  "num_attention_heads": 28,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 4,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.44.0",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151646
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.44.0"
+}

latest

@@ -0,0 +1 @@
+global_step264

model-00001-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:176dae3d16a3334a5d17b982e0dc82f37c192628c04d0767099771c74ef9c858
+size 4970694768

model-00002-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ac77f62f543270f3d374df282ed2b679a29b8168e7d0e0604bd837e1d744248
+size 4778622352

model-00003-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aed0ce1317aa6a283b21f911cc382bf6bf312ff6348bd66e7c95fd11800dab7f
+size 4932743960

model-00004-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ece7c7521bf018e41d9c413b16799065f3e31f1e81139702c16acaf99622386
+size 4932743992

model-00005-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:895c467810ad250571df4a1afb0a0809b91d50d15b8415c45a0e1b681148bfdf
+size 4998852296

model-00006-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f29fb0bca794ed95ddd5091eeb4a12c407720dd33a65a4f98992f32f895b0e32
+size 3662865184

model-00007-of-00007.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:660c4a515e46e1f77aa4a2107d27d6bc62bb167d8f1b787a3ae8df6052695ddf
+size 2173997184

model.safetensors.index.json

@@ -0,0 +1,346 @@
+{
+  "metadata": {
+    "total_size": 30450481152
+  },
+  "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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-00004-of-00007.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00007.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "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.bias": "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.bias": "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-00004-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.bias": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00004-of-00007.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00004-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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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-00005-of-00007.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00005-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-00005-of-00007.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "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.bias": "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.bias": "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-00005-of-00007.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00005-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.bias": "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.bias": "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.bias": "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.bias": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00005-of-00007.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00005-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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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-00006-of-00007.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "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-00006-of-00007.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "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.bias": "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-00006-of-00007.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00006-of-00007.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00006-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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "model-00006-of-00007.safetensors",
+    "model.layers.27.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.bias": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00002-of-00007.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00002-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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "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-00003-of-00007.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00007.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00002-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.bias": "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.bias": "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.bias": "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.bias": "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.bias": "model-00003-of-00007.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00007.safetensors",
+    "model.norm.weight": "model-00006-of-00007.safetensors"
+  }
+}

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0de366c30136073a3a0c31ebb2290502f8ef75fd54efda5df6f56df25b5a7f00
+size 15984

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9554f22a6d5ae3d3b393b6837eb160c7754c32cf937825b966efe64d23557e02
+size 15984

rng_state_10.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77f10e2c825a754a9c6ebe0682597681d28e0cbeecb75697cb3e66e5a74b8217
+size 15997

rng_state_11.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ea005818cdfea76a36d837575c12cf5dd4239d36edd73ead7d2da31bc72f231
+size 15997

rng_state_12.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0fa9fa6615cce499065a24a80775bee104f61fbb47adbceecf910f7f2a4f5dd
+size 15997

rng_state_13.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46d67f16a06bc40f5566d22f62fa5f3ac47f4d1549a83e82450b9b3fb3fa9be8
+size 15997

rng_state_14.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51983b4af2244563270a66ff5c09efb20f071e45cae88450e708597fcc607772
+size 15997

rng_state_15.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc0daa6984ccf29cbc9fed94ddd7d3dad9d2fb6c8d333d177e0c84a650a2052e
+size 15997

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4f78e693f6792779c6128da5fd5741738c1f7d8c48a5746e810301f3efc6f02
+size 15984

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c46645cc186d565845666fc3529f7f823aabb50d2a24f09b467322f3441ae074
+size 15984

rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e255eb35b3057165a202ef2df03bef9d5adb4dcf592f474ae36641677c5e5b5
+size 15984

rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdf4217d0873c8f7ea406f930eeee3c047a55ffd054662d4a520a5ec0f0e09f0
+size 15984

rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62c79c54f22bad76685ed183b0d934ca2d367b939106013b4f56f6b0888bb0da
+size 15984

rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ecb2dd1a04220fa26737941337203e9901682ee83a2e40e12a28ab6529f00e40
+size 15984

rng_state_8.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b42876183d46d0cf8fd02e8cbb96f226fae4814352be9d47fdc4d41368585c4f
+size 15984

rng_state_9.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36bca8713d5d11f600c73c13d8a7c601b48d2522d30fedaeba165ce1455c00f9
+size 15984

special_tokens_map.json

@@ -0,0 +1,34 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,61 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "128244": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": "<s>",
+  "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 10000,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "truncation_side": "left",
+  "unk_token": "<unk>"
+}

trainer_state.json

@@ -0,0 +1,2433 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 2000.0,
+  "global_step": 264,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.011363636363636364,
+      "grad_norm": 5.703509592126629,
+      "kl": 0.0,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 0.5688,
+      "step": 1,
+      "step_loss": 0.5533416867256165
+    },
+    {
+      "epoch": 0.022727272727272728,
+      "grad_norm": 2.414278384928731,
+      "kl": 0.1508263200521469,
+      "learning_rate": 2.438044511330269e-06,
+      "loss": 0.4403,
+      "step": 2,
+      "step_loss": 0.5107974410057068
+    },
+    {
+      "epoch": 0.03409090909090909,
+      "grad_norm": 2.5451897972697393,
+      "kl": 0.19495707750320435,
+      "learning_rate": 3.5717278751869343e-06,
+      "loss": 0.4802,
+      "step": 3,
+      "step_loss": 0.5942587852478027
+    },
+    {
+      "epoch": 0.045454545454545456,
+      "grad_norm": 1.362435064605772,
+      "kl": 0.14722466468811035,
+      "learning_rate": 4.376089022660538e-06,
+      "loss": 0.4463,
+      "step": 4,
+      "step_loss": 0.460592657327652
+    },
+    {
+      "epoch": 0.056818181818181816,
+      "grad_norm": 1.1743592442810231,
+      "kl": 0.12074092030525208,
+      "learning_rate": 5e-06,
+      "loss": 0.4001,
+      "step": 5,
+      "step_loss": 0.39433369040489197
+    },
+    {
+      "epoch": 0.06818181818181818,
+      "grad_norm": 1.2999424793282481,
+      "kl": 0.1568446159362793,
+      "learning_rate": 4.999941322464311e-06,
+      "loss": 0.4067,
+      "step": 6,
+      "step_loss": 0.356352835893631
+    },
+    {
+      "epoch": 0.07954545454545454,
+      "grad_norm": 1.160541252191944,
+      "kl": 0.23409639298915863,
+      "learning_rate": 4.999765292917736e-06,
+      "loss": 0.3998,
+      "step": 7,
+      "step_loss": 0.4068807065486908
+    },
+    {
+      "epoch": 0.09090909090909091,
+      "grad_norm": 1.227986620612679,
+      "kl": 0.24431678652763367,
+      "learning_rate": 4.999471920541589e-06,
+      "loss": 0.368,
+      "step": 8,
+      "step_loss": 0.41308918595314026
+    },
+    {
+      "epoch": 0.10227272727272728,
+      "grad_norm": 1.118558082955084,
+      "kl": 0.17894110083580017,
+      "learning_rate": 4.999061220637533e-06,
+      "loss": 0.3714,
+      "step": 9,
+      "step_loss": 0.2875981032848358
+    },
+    {
+      "epoch": 0.11363636363636363,
+      "grad_norm": 1.044241767662719,
+      "kl": 0.2728317379951477,
+      "learning_rate": 4.998533214626773e-06,
+      "loss": 0.3746,
+      "step": 10,
+      "step_loss": 0.381584107875824
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 1.012833206073384,
+      "kl": 0.24114784598350525,
+      "learning_rate": 4.9978879300489484e-06,
+      "loss": 0.3425,
+      "step": 11,
+      "step_loss": 0.3758848309516907
+    },
+    {
+      "epoch": 0.13636363636363635,
+      "grad_norm": 0.9561570826746042,
+      "kl": 0.24886220693588257,
+      "learning_rate": 4.9971254005606865e-06,
+      "loss": 0.3832,
+      "step": 12,
+      "step_loss": 0.40528154373168945
+    },
+    {
+      "epoch": 0.14772727272727273,
+      "grad_norm": 0.884172522884705,
+      "kl": 0.23464588820934296,
+      "learning_rate": 4.996245665933857e-06,
+      "loss": 0.3678,
+      "step": 13,
+      "step_loss": 0.3059731721878052
+    },
+    {
+      "epoch": 0.1590909090909091,
+      "grad_norm": 1.0353201682385085,
+      "kl": 0.1602363884449005,
+      "learning_rate": 4.9952487720534905e-06,
+      "loss": 0.3777,
+      "step": 14,
+      "step_loss": 0.3509451448917389
+    },
+    {
+      "epoch": 0.17045454545454544,
+      "grad_norm": 0.8535378320105195,
+      "kl": 0.2152172029018402,
+      "learning_rate": 4.9941347709153875e-06,
+      "loss": 0.3539,
+      "step": 15,
+      "step_loss": 0.36119481921195984
+    },
+    {
+      "epoch": 0.18181818181818182,
+      "grad_norm": 1.0118408129491079,
+      "kl": 0.2640606164932251,
+      "learning_rate": 4.992903720623408e-06,
+      "loss": 0.3701,
+      "step": 16,
+      "step_loss": 0.43707239627838135
+    },
+    {
+      "epoch": 0.19318181818181818,
+      "grad_norm": 0.9252651401652947,
+      "kl": 0.24001294374465942,
+      "learning_rate": 4.991555685386438e-06,
+      "loss": 0.3464,
+      "step": 17,
+      "step_loss": 0.39861610531806946
+    },
+    {
+      "epoch": 0.20454545454545456,
+      "grad_norm": 0.9546545294579731,
+      "kl": 0.2848950922489166,
+      "learning_rate": 4.990090735515043e-06,
+      "loss": 0.3729,
+      "step": 18,
+      "step_loss": 0.5053370594978333
+    },
+    {
+      "epoch": 0.2159090909090909,
+      "grad_norm": 0.8545398633463176,
+      "kl": 0.161998450756073,
+      "learning_rate": 4.988508947417799e-06,
+      "loss": 0.359,
+      "step": 19,
+      "step_loss": 0.3288613259792328
+    },
+    {
+      "epoch": 0.22727272727272727,
+      "grad_norm": 0.8025839263192421,
+      "kl": 0.2056322991847992,
+      "learning_rate": 4.98681040359731e-06,
+      "loss": 0.3439,
+      "step": 20,
+      "step_loss": 0.35805100202560425
+    },
+    {
+      "epoch": 0.23863636363636365,
+      "grad_norm": 0.8744693236774967,
+      "kl": 0.2929481267929077,
+      "learning_rate": 4.984995192645897e-06,
+      "loss": 0.3792,
+      "step": 21,
+      "step_loss": 0.45748788118362427
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.8729052332734021,
+      "kl": 0.2217060774564743,
+      "learning_rate": 4.983063409240992e-06,
+      "loss": 0.3505,
+      "step": 22,
+      "step_loss": 0.35472381114959717
+    },
+    {
+      "epoch": 0.26136363636363635,
+      "grad_norm": 0.8762945408187117,
+      "kl": 0.2572169601917267,
+      "learning_rate": 4.981015154140181e-06,
+      "loss": 0.3625,
+      "step": 23,
+      "step_loss": 0.4360867738723755
+    },
+    {
+      "epoch": 0.2727272727272727,
+      "grad_norm": 0.8055639586247896,
+      "kl": 0.3179939091205597,
+      "learning_rate": 4.978850534175967e-06,
+      "loss": 0.3703,
+      "step": 24,
+      "step_loss": 0.4685637056827545
+    },
+    {
+      "epoch": 0.2840909090909091,
+      "grad_norm": 0.7776274043752576,
+      "kl": 0.21350209414958954,
+      "learning_rate": 4.976569662250185e-06,
+      "loss": 0.3606,
+      "step": 25,
+      "step_loss": 0.3343837857246399
+    },
+    {
+      "epoch": 0.29545454545454547,
+      "grad_norm": 0.8563133494756662,
+      "kl": 0.23110716044902802,
+      "learning_rate": 4.974172657328117e-06,
+      "loss": 0.3468,
+      "step": 26,
+      "step_loss": 0.27567654848098755
+    },
+    {
+      "epoch": 0.3068181818181818,
+      "grad_norm": 0.8318009751857389,
+      "kl": 0.23791126906871796,
+      "learning_rate": 4.97165964443229e-06,
+      "loss": 0.3766,
+      "step": 27,
+      "step_loss": 0.43361616134643555
+    },
+    {
+      "epoch": 0.3181818181818182,
+      "grad_norm": 0.8256316165817771,
+      "kl": 0.2149941325187683,
+      "learning_rate": 4.96903075463595e-06,
+      "loss": 0.3507,
+      "step": 28,
+      "step_loss": 0.41853615641593933
+    },
+    {
+      "epoch": 0.32954545454545453,
+      "grad_norm": 0.897517985957999,
+      "kl": 0.24494698643684387,
+      "learning_rate": 4.966286125056234e-06,
+      "loss": 0.3647,
+      "step": 29,
+      "step_loss": 0.3512318730354309
+    },
+    {
+      "epoch": 0.3409090909090909,
+      "grad_norm": 0.9026937755400121,
+      "kl": 0.184300035238266,
+      "learning_rate": 4.963425898847006e-06,
+      "loss": 0.3518,
+      "step": 30,
+      "step_loss": 0.3595576882362366
+    },
+    {
+      "epoch": 0.3522727272727273,
+      "grad_norm": 0.85148287759613,
+      "kl": 0.24502253532409668,
+      "learning_rate": 4.960450225191402e-06,
+      "loss": 0.3627,
+      "step": 31,
+      "step_loss": 0.35294580459594727
+    },
+    {
+      "epoch": 0.36363636363636365,
+      "grad_norm": 0.7962414335930865,
+      "kl": 0.2057955265045166,
+      "learning_rate": 4.957359259294038e-06,
+      "loss": 0.3458,
+      "step": 32,
+      "step_loss": 0.34822237491607666
+    },
+    {
+      "epoch": 0.375,
+      "grad_norm": 0.800931144269417,
+      "kl": 0.21041449904441833,
+      "learning_rate": 4.954153162372928e-06,
+      "loss": 0.377,
+      "step": 33,
+      "step_loss": 0.40625178813934326
+    },
+    {
+      "epoch": 0.38636363636363635,
+      "grad_norm": 0.8558589097843969,
+      "kl": 0.2527558505535126,
+      "learning_rate": 4.950832101651063e-06,
+      "loss": 0.3845,
+      "step": 34,
+      "step_loss": 0.38101768493652344
+    },
+    {
+      "epoch": 0.3977272727272727,
+      "grad_norm": 0.8387953491003776,
+      "kl": 0.2718813717365265,
+      "learning_rate": 4.947396250347695e-06,
+      "loss": 0.3527,
+      "step": 35,
+      "step_loss": 0.3798375129699707
+    },
+    {
+      "epoch": 0.4090909090909091,
+      "grad_norm": 0.9507472286861052,
+      "kl": 0.1885739266872406,
+      "learning_rate": 4.943845787669303e-06,
+      "loss": 0.3589,
+      "step": 36,
+      "step_loss": 0.3146296441555023
+    },
+    {
+      "epoch": 0.42045454545454547,
+      "grad_norm": 0.8385876510679799,
+      "kl": 0.16516371071338654,
+      "learning_rate": 4.9401808988002425e-06,
+      "loss": 0.3578,
+      "step": 37,
+      "step_loss": 0.2985036373138428
+    },
+    {
+      "epoch": 0.4318181818181818,
+      "grad_norm": 0.8166686735971924,
+      "kl": 0.24522481858730316,
+      "learning_rate": 4.936401774893088e-06,
+      "loss": 0.3593,
+      "step": 38,
+      "step_loss": 0.4174098074436188
+    },
+    {
+      "epoch": 0.4431818181818182,
+      "grad_norm": 0.9075664298384456,
+      "kl": 0.23953908681869507,
+      "learning_rate": 4.932508613058665e-06,
+      "loss": 0.3711,
+      "step": 39,
+      "step_loss": 0.3471056818962097
+    },
+    {
+      "epoch": 0.45454545454545453,
+      "grad_norm": 0.8181070744566579,
+      "kl": 0.30609607696533203,
+      "learning_rate": 4.928501616355768e-06,
+      "loss": 0.3599,
+      "step": 40,
+      "step_loss": 0.33987393975257874
+    },
+    {
+      "epoch": 0.4659090909090909,
+      "grad_norm": 0.7674769366399429,
+      "kl": 0.3086016774177551,
+      "learning_rate": 4.924380993780566e-06,
+      "loss": 0.3434,
+      "step": 41,
+      "step_loss": 0.3478284776210785
+    },
+    {
+      "epoch": 0.4772727272727273,
+      "grad_norm": 0.796828960509896,
+      "kl": 0.2363354116678238,
+      "learning_rate": 4.920146960255707e-06,
+      "loss": 0.3458,
+      "step": 42,
+      "step_loss": 0.3613693118095398
+    },
+    {
+      "epoch": 0.48863636363636365,
+      "grad_norm": 0.8201944175381407,
+      "kl": 0.21424362063407898,
+      "learning_rate": 4.915799736619105e-06,
+      "loss": 0.3572,
+      "step": 43,
+      "step_loss": 0.33215075731277466
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.846105466640005,
+      "kl": 0.17349004745483398,
+      "learning_rate": 4.911339549612422e-06,
+      "loss": 0.3782,
+      "step": 44,
+      "step_loss": 0.3519684672355652
+    },
+    {
+      "epoch": 0.5113636363636364,
+      "grad_norm": 0.8219513643708384,
+      "kl": 0.26255759596824646,
+      "learning_rate": 4.906766631869243e-06,
+      "loss": 0.362,
+      "step": 45,
+      "step_loss": 0.34211111068725586
+    },
+    {
+      "epoch": 0.5227272727272727,
+      "grad_norm": 0.8293883565647935,
+      "kl": 0.3788855969905853,
+      "learning_rate": 4.90208122190294e-06,
+      "loss": 0.3497,
+      "step": 46,
+      "step_loss": 0.40293213725090027
+    },
+    {
+      "epoch": 0.5340909090909091,
+      "grad_norm": 0.8369516655788055,
+      "kl": 0.27136877179145813,
+      "learning_rate": 4.897283564094233e-06,
+      "loss": 0.3761,
+      "step": 47,
+      "step_loss": 0.3964707851409912
+    },
+    {
+      "epoch": 0.5454545454545454,
+      "grad_norm": 0.8300152564082867,
+      "kl": 0.25666913390159607,
+      "learning_rate": 4.892373908678445e-06,
+      "loss": 0.3539,
+      "step": 48,
+      "step_loss": 0.3369652330875397
+    },
+    {
+      "epoch": 0.5568181818181818,
+      "grad_norm": 0.9104721437976564,
+      "kl": 0.274257630109787,
+      "learning_rate": 4.887352511732447e-06,
+      "loss": 0.389,
+      "step": 49,
+      "step_loss": 0.42897558212280273
+    },
+    {
+      "epoch": 0.5681818181818182,
+      "grad_norm": 0.8195403324929418,
+      "kl": 0.32372036576271057,
+      "learning_rate": 4.882219635161306e-06,
+      "loss": 0.3518,
+      "step": 50,
+      "step_loss": 0.3887690007686615
+    },
+    {
+      "epoch": 0.5795454545454546,
+      "grad_norm": 0.9168553194353436,
+      "kl": 0.28414681553840637,
+      "learning_rate": 4.876975546684619e-06,
+      "loss": 0.3581,
+      "step": 51,
+      "step_loss": 0.40320277214050293
+    },
+    {
+      "epoch": 0.5909090909090909,
+      "grad_norm": 0.8617294656704618,
+      "kl": 0.2004740983247757,
+      "learning_rate": 4.8716205198225525e-06,
+      "loss": 0.3805,
+      "step": 52,
+      "step_loss": 0.38499268889427185
+    },
+    {
+      "epoch": 0.6022727272727273,
+      "grad_norm": 0.7697913025754929,
+      "kl": 0.2655254900455475,
+      "learning_rate": 4.866154833881579e-06,
+      "loss": 0.3375,
+      "step": 53,
+      "step_loss": 0.3799101710319519
+    },
+    {
+      "epoch": 0.6136363636363636,
+      "grad_norm": 0.860433412224209,
+      "kl": 0.1964799165725708,
+      "learning_rate": 4.8605787739399055e-06,
+      "loss": 0.3614,
+      "step": 54,
+      "step_loss": 0.3518129587173462
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 0.8208795676466634,
+      "kl": 0.2083766907453537,
+      "learning_rate": 4.8548926308326025e-06,
+      "loss": 0.3497,
+      "step": 55,
+      "step_loss": 0.36417239904403687
+    },
+    {
+      "epoch": 0.6363636363636364,
+      "grad_norm": 0.8494420730908946,
+      "kl": 0.24671347439289093,
+      "learning_rate": 4.84909670113644e-06,
+      "loss": 0.3585,
+      "step": 56,
+      "step_loss": 0.28457289934158325
+    },
+    {
+      "epoch": 0.6477272727272727,
+      "grad_norm": 0.8001936082214277,
+      "kl": 0.20547156035900116,
+      "learning_rate": 4.843191287154415e-06,
+      "loss": 0.3387,
+      "step": 57,
+      "step_loss": 0.31630027294158936
+    },
+    {
+      "epoch": 0.6590909090909091,
+      "grad_norm": 0.8349937030988813,
+      "kl": 0.24216747283935547,
+      "learning_rate": 4.837176696899984e-06,
+      "loss": 0.3562,
+      "step": 58,
+      "step_loss": 0.31798893213272095
+    },
+    {
+      "epoch": 0.6704545454545454,
+      "grad_norm": 0.7997137630513312,
+      "kl": 0.22188925743103027,
+      "learning_rate": 4.8310532440810005e-06,
+      "loss": 0.3563,
+      "step": 59,
+      "step_loss": 0.30022716522216797
+    },
+    {
+      "epoch": 0.6818181818181818,
+      "grad_norm": 0.7732049611711472,
+      "kl": 0.2248784899711609,
+      "learning_rate": 4.82482124808335e-06,
+      "loss": 0.3191,
+      "step": 60,
+      "step_loss": 0.36109238862991333
+    },
+    {
+      "epoch": 0.6931818181818182,
+      "grad_norm": 0.8067392653775698,
+      "kl": 0.24406936764717102,
+      "learning_rate": 4.8184810339542925e-06,
+      "loss": 0.3501,
+      "step": 61,
+      "step_loss": 0.3913516402244568
+    },
+    {
+      "epoch": 0.7045454545454546,
+      "grad_norm": 0.7792045901924106,
+      "kl": 0.19702410697937012,
+      "learning_rate": 4.812032932385509e-06,
+      "loss": 0.3238,
+      "step": 62,
+      "step_loss": 0.2769385576248169
+    },
+    {
+      "epoch": 0.7159090909090909,
+      "grad_norm": 0.8595897752967626,
+      "kl": 0.2642907202243805,
+      "learning_rate": 4.805477279695852e-06,
+      "loss": 0.3538,
+      "step": 63,
+      "step_loss": 0.4088588356971741
+    },
+    {
+      "epoch": 0.7272727272727273,
+      "grad_norm": 0.8639662428592647,
+      "kl": 0.22237904369831085,
+      "learning_rate": 4.798814417813807e-06,
+      "loss": 0.3576,
+      "step": 64,
+      "step_loss": 0.3004940152168274
+    },
+    {
+      "epoch": 0.7386363636363636,
+      "grad_norm": 0.7875668542731125,
+      "kl": 0.22941315174102783,
+      "learning_rate": 4.7920446942596535e-06,
+      "loss": 0.3314,
+      "step": 65,
+      "step_loss": 0.2649787664413452
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.825774798553009,
+      "kl": 0.27141043543815613,
+      "learning_rate": 4.7851684621273435e-06,
+      "loss": 0.3694,
+      "step": 66,
+      "step_loss": 0.4899372458457947
+    },
+    {
+      "epoch": 0.7613636363636364,
+      "grad_norm": 0.7704537376313643,
+      "kl": 0.2449103444814682,
+      "learning_rate": 4.7781860800660836e-06,
+      "loss": 0.365,
+      "step": 67,
+      "step_loss": 0.41683661937713623
+    },
+    {
+      "epoch": 0.7727272727272727,
+      "grad_norm": 0.7904020967751078,
+      "kl": 0.30710798501968384,
+      "learning_rate": 4.771097912261626e-06,
+      "loss": 0.3577,
+      "step": 68,
+      "step_loss": 0.39199817180633545
+    },
+    {
+      "epoch": 0.7840909090909091,
+      "grad_norm": 0.7703764480757949,
+      "kl": 0.21822941303253174,
+      "learning_rate": 4.763904328417276e-06,
+      "loss": 0.3568,
+      "step": 69,
+      "step_loss": 0.3460182249546051
+    },
+    {
+      "epoch": 0.7954545454545454,
+      "grad_norm": 0.764160468250735,
+      "kl": 0.24635545909404755,
+      "learning_rate": 4.756605703734611e-06,
+      "loss": 0.3519,
+      "step": 70,
+      "step_loss": 0.4064783453941345
+    },
+    {
+      "epoch": 0.8068181818181818,
+      "grad_norm": 0.7481024101498955,
+      "kl": 0.14451748132705688,
+      "learning_rate": 4.7492024188939055e-06,
+      "loss": 0.3391,
+      "step": 71,
+      "step_loss": 0.25107917189598083
+    },
+    {
+      "epoch": 0.8181818181818182,
+      "grad_norm": 0.7802950172161581,
+      "kl": 0.1930258721113205,
+      "learning_rate": 4.741694860034281e-06,
+      "loss": 0.3564,
+      "step": 72,
+      "step_loss": 0.34083372354507446
+    },
+    {
+      "epoch": 0.8295454545454546,
+      "grad_norm": 0.7769810199285,
+      "kl": 0.23945628106594086,
+      "learning_rate": 4.734083418733563e-06,
+      "loss": 0.3725,
+      "step": 73,
+      "step_loss": 0.2961326241493225
+    },
+    {
+      "epoch": 0.8409090909090909,
+      "grad_norm": 0.825406272151327,
+      "kl": 0.23796045780181885,
+      "learning_rate": 4.726368491987854e-06,
+      "loss": 0.3376,
+      "step": 74,
+      "step_loss": 0.35123029351234436
+    },
+    {
+      "epoch": 0.8522727272727273,
+      "grad_norm": 0.7306727160298511,
+      "kl": 0.13507644832134247,
+      "learning_rate": 4.718550482190837e-06,
+      "loss": 0.3445,
+      "step": 75,
+      "step_loss": 0.2955048680305481
+    },
+    {
+      "epoch": 0.8636363636363636,
+      "grad_norm": 0.7579927918209679,
+      "kl": 0.24724549055099487,
+      "learning_rate": 4.7106297971127755e-06,
+      "loss": 0.349,
+      "step": 76,
+      "step_loss": 0.3554742932319641
+    },
+    {
+      "epoch": 0.875,
+      "grad_norm": 0.805419144828077,
+      "kl": 0.20437797904014587,
+      "learning_rate": 4.7026068498792535e-06,
+      "loss": 0.3439,
+      "step": 77,
+      "step_loss": 0.30100610852241516
+    },
+    {
+      "epoch": 0.8863636363636364,
+      "grad_norm": 0.8550394104357324,
+      "kl": 0.16827110946178436,
+      "learning_rate": 4.694482058949624e-06,
+      "loss": 0.3512,
+      "step": 78,
+      "step_loss": 0.2656687796115875
+    },
+    {
+      "epoch": 0.8977272727272727,
+      "grad_norm": 0.7699285730450894,
+      "kl": 0.17964152991771698,
+      "learning_rate": 4.686255848095186e-06,
+      "loss": 0.3407,
+      "step": 79,
+      "step_loss": 0.23917566239833832
+    },
+    {
+      "epoch": 0.9090909090909091,
+      "grad_norm": 0.7978865990480669,
+      "kl": 0.27529603242874146,
+      "learning_rate": 4.677928646377076e-06,
+      "loss": 0.3433,
+      "step": 80,
+      "step_loss": 0.33308878540992737
+    },
+    {
+      "epoch": 0.9204545454545454,
+      "grad_norm": 0.7840380701095863,
+      "kl": 0.22533753514289856,
+      "learning_rate": 4.669500888123897e-06,
+      "loss": 0.344,
+      "step": 81,
+      "step_loss": 0.33834126591682434
+    },
+    {
+      "epoch": 0.9318181818181818,
+      "grad_norm": 0.856812662574151,
+      "kl": 0.2573813498020172,
+      "learning_rate": 4.660973012909057e-06,
+      "loss": 0.359,
+      "step": 82,
+      "step_loss": 0.3473721146583557
+    },
+    {
+      "epoch": 0.9431818181818182,
+      "grad_norm": 0.7265171734197157,
+      "kl": 0.2186812162399292,
+      "learning_rate": 4.652345465527847e-06,
+      "loss": 0.3341,
+      "step": 83,
+      "step_loss": 0.3598743975162506
+    },
+    {
+      "epoch": 0.9545454545454546,
+      "grad_norm": 0.8049253093828008,
+      "kl": 0.2602262794971466,
+      "learning_rate": 4.64361869597424e-06,
+      "loss": 0.3464,
+      "step": 84,
+      "step_loss": 0.3444975018501282
+    },
+    {
+      "epoch": 0.9659090909090909,
+      "grad_norm": 0.7680402534237456,
+      "kl": 0.2810295522212982,
+      "learning_rate": 4.634793159417421e-06,
+      "loss": 0.3373,
+      "step": 85,
+      "step_loss": 0.26042619347572327
+    },
+    {
+      "epoch": 0.9772727272727273,
+      "grad_norm": 0.8011651956250969,
+      "kl": 0.24380852282047272,
+      "learning_rate": 4.625869316178043e-06,
+      "loss": 0.3676,
+      "step": 86,
+      "step_loss": 0.4119197428226471
+    },
+    {
+      "epoch": 0.9886363636363636,
+      "grad_norm": 0.7751572880029601,
+      "kl": 0.25345954298973083,
+      "learning_rate": 4.6168476317042224e-06,
+      "loss": 0.3449,
+      "step": 87,
+      "step_loss": 0.38151469826698303
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.7947674725517713,
+      "kl": 0.2837441861629486,
+      "learning_rate": 4.60772857654726e-06,
+      "loss": 0.3646,
+      "step": 88,
+      "step_loss": 0.4021923542022705
+    },
+    {
+      "epoch": 1.0,
+      "eval_test_transformed.json_loss": 0.6147641539573669,
+      "eval_test_transformed.json_runtime": 57.2166,
+      "eval_test_transformed.json_samples_per_second": 8.739,
+      "eval_test_transformed.json_steps_per_second": 0.559,
+      "step": 88
+    },
+    {
+      "epoch": 1.0113636363636365,
+      "grad_norm": 0.8300174107506423,
+      "kl": 0.08372548967599869,
+      "learning_rate": 4.598512626337096e-06,
+      "loss": 0.2738,
+      "step": 89,
+      "step_loss": 0.5384058952331543
+    },
+    {
+      "epoch": 1.0227272727272727,
+      "grad_norm": 0.77573239267709,
+      "kl": 0.2220146656036377,
+      "learning_rate": 4.589200261757507e-06,
+      "loss": 0.2835,
+      "step": 90,
+      "step_loss": 0.23898854851722717
+    },
+    {
+      "epoch": 1.0340909090909092,
+      "grad_norm": 0.724932328716244,
+      "kl": 0.3047032654285431,
+      "learning_rate": 4.5797919685210306e-06,
+      "loss": 0.2681,
+      "step": 91,
+      "step_loss": 0.30961817502975464
+    },
+    {
+      "epoch": 1.0454545454545454,
+      "grad_norm": 0.6895314893413088,
+      "kl": 0.24007435142993927,
+      "learning_rate": 4.570288237343632e-06,
+      "loss": 0.2602,
+      "step": 92,
+      "step_loss": 0.29084306955337524
+    },
+    {
+      "epoch": 1.0568181818181819,
+      "grad_norm": 0.8345390101480717,
+      "kl": 0.2587500810623169,
+      "learning_rate": 4.560689563919113e-06,
+      "loss": 0.2838,
+      "step": 93,
+      "step_loss": 0.2709003686904907
+    },
+    {
+      "epoch": 1.0681818181818181,
+      "grad_norm": 0.8688624933533857,
+      "kl": 0.26922354102134705,
+      "learning_rate": 4.550996448893253e-06,
+      "loss": 0.2866,
+      "step": 94,
+      "step_loss": 0.3056609034538269
+    },
+    {
+      "epoch": 1.0795454545454546,
+      "grad_norm": 0.7821685731421445,
+      "kl": 0.25485578179359436,
+      "learning_rate": 4.5412093978376986e-06,
+      "loss": 0.2694,
+      "step": 95,
+      "step_loss": 0.22295084595680237
+    },
+    {
+      "epoch": 1.0909090909090908,
+      "grad_norm": 0.8329103508983827,
+      "kl": 0.3171152174472809,
+      "learning_rate": 4.531328921223596e-06,
+      "loss": 0.2676,
+      "step": 96,
+      "step_loss": 0.24555529654026031
+    },
+    {
+      "epoch": 1.1022727272727273,
+      "grad_norm": 0.8278389202666937,
+      "kl": 0.268743634223938,
+      "learning_rate": 4.521355534394959e-06,
+      "loss": 0.2806,
+      "step": 97,
+      "step_loss": 0.2850462794303894
+    },
+    {
+      "epoch": 1.1136363636363635,
+      "grad_norm": 0.7515633501773713,
+      "kl": 0.21348519623279572,
+      "learning_rate": 4.511289757541801e-06,
+      "loss": 0.2714,
+      "step": 98,
+      "step_loss": 0.22777102887630463
+    },
+    {
+      "epoch": 1.125,
+      "grad_norm": 0.7901926539707923,
+      "kl": 0.2479882836341858,
+      "learning_rate": 4.501132115672992e-06,
+      "loss": 0.2621,
+      "step": 99,
+      "step_loss": 0.27556145191192627
+    },
+    {
+      "epoch": 1.1363636363636362,
+      "grad_norm": 0.7721311538003944,
+      "kl": 0.2527960538864136,
+      "learning_rate": 4.490883138588882e-06,
+      "loss": 0.2673,
+      "step": 100,
+      "step_loss": 0.2762888967990875
+    },
+    {
+      "epoch": 1.1477272727272727,
+      "grad_norm": 0.7810424529002389,
+      "kl": 0.26764342188835144,
+      "learning_rate": 4.4805433608536655e-06,
+      "loss": 0.269,
+      "step": 101,
+      "step_loss": 0.2725462317466736
+    },
+    {
+      "epoch": 1.1590909090909092,
+      "grad_norm": 0.7955281338155472,
+      "kl": 0.3150593042373657,
+      "learning_rate": 4.470113321767499e-06,
+      "loss": 0.2761,
+      "step": 102,
+      "step_loss": 0.311062216758728
+    },
+    {
+      "epoch": 1.1704545454545454,
+      "grad_norm": 0.7565768284407913,
+      "kl": 0.2376776784658432,
+      "learning_rate": 4.459593565338376e-06,
+      "loss": 0.2851,
+      "step": 103,
+      "step_loss": 0.2452271282672882
+    },
+    {
+      "epoch": 1.1818181818181819,
+      "grad_norm": 0.8304952134574923,
+      "kl": 0.2546381950378418,
+      "learning_rate": 4.448984640253747e-06,
+      "loss": 0.2818,
+      "step": 104,
+      "step_loss": 0.3039626479148865
+    },
+    {
+      "epoch": 1.1931818181818181,
+      "grad_norm": 0.8207694461248964,
+      "kl": 0.41062766313552856,
+      "learning_rate": 4.438287099851905e-06,
+      "loss": 0.2698,
+      "step": 105,
+      "step_loss": 0.3451313078403473
+    },
+    {
+      "epoch": 1.2045454545454546,
+      "grad_norm": 0.763176799019057,
+      "kl": 0.24396805465221405,
+      "learning_rate": 4.427501502093126e-06,
+      "loss": 0.2666,
+      "step": 106,
+      "step_loss": 0.24862422049045563
+    },
+    {
+      "epoch": 1.2159090909090908,
+      "grad_norm": 0.7653625194049708,
+      "kl": 0.3014935255050659,
+      "learning_rate": 4.416628409530563e-06,
+      "loss": 0.2615,
+      "step": 107,
+      "step_loss": 0.23431876301765442
+    },
+    {
+      "epoch": 1.2272727272727273,
+      "grad_norm": 0.7932503486822488,
+      "kl": 0.23210333287715912,
+      "learning_rate": 4.405668389280906e-06,
+      "loss": 0.2615,
+      "step": 108,
+      "step_loss": 0.239517942070961
+    },
+    {
+      "epoch": 1.2386363636363638,
+      "grad_norm": 0.8303257471879192,
+      "kl": 0.25863131880760193,
+      "learning_rate": 4.394622012994803e-06,
+      "loss": 0.2733,
+      "step": 109,
+      "step_loss": 0.2567521035671234
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.7822296104419395,
+      "kl": 0.18686552345752716,
+      "learning_rate": 4.383489856827045e-06,
+      "loss": 0.2772,
+      "step": 110,
+      "step_loss": 0.23640470206737518
+    },
+    {
+      "epoch": 1.2613636363636362,
+      "grad_norm": 0.7749159198189497,
+      "kl": 0.26931869983673096,
+      "learning_rate": 4.372272501406511e-06,
+      "loss": 0.2754,
+      "step": 111,
+      "step_loss": 0.29603561758995056
+    },
+    {
+      "epoch": 1.2727272727272727,
+      "grad_norm": 0.8008832913025072,
+      "kl": 0.2696314752101898,
+      "learning_rate": 4.360970531805891e-06,
+      "loss": 0.2646,
+      "step": 112,
+      "step_loss": 0.2858111262321472
+    },
+    {
+      "epoch": 1.2840909090909092,
+      "grad_norm": 0.8001913735849326,
+      "kl": 0.2806592881679535,
+      "learning_rate": 4.34958453751116e-06,
+      "loss": 0.2662,
+      "step": 113,
+      "step_loss": 0.2791585922241211
+    },
+    {
+      "epoch": 1.2954545454545454,
+      "grad_norm": 0.8024500136687336,
+      "kl": 0.24742648005485535,
+      "learning_rate": 4.338115112390839e-06,
+      "loss": 0.2777,
+      "step": 114,
+      "step_loss": 0.27738770842552185
+    },
+    {
+      "epoch": 1.3068181818181819,
+      "grad_norm": 0.7743853146387183,
+      "kl": 0.24617239832878113,
+      "learning_rate": 4.32656285466502e-06,
+      "loss": 0.272,
+      "step": 115,
+      "step_loss": 0.2725653350353241
+    },
+    {
+      "epoch": 1.3181818181818181,
+      "grad_norm": 0.8260411729526528,
+      "kl": 0.24938064813613892,
+      "learning_rate": 4.314928366874162e-06,
+      "loss": 0.2709,
+      "step": 116,
+      "step_loss": 0.23085035383701324
+    },
+    {
+      "epoch": 1.3295454545454546,
+      "grad_norm": 0.7593579608277142,
+      "kl": 0.2912304401397705,
+      "learning_rate": 4.30321225584766e-06,
+      "loss": 0.2595,
+      "step": 117,
+      "step_loss": 0.2982204258441925
+    },
+    {
+      "epoch": 1.3409090909090908,
+      "grad_norm": 0.7590166337347966,
+      "kl": 0.3144587576389313,
+      "learning_rate": 4.291415132672202e-06,
+      "loss": 0.2715,
+      "step": 118,
+      "step_loss": 0.2491433173418045
+    },
+    {
+      "epoch": 1.3522727272727273,
+      "grad_norm": 0.8096110997434255,
+      "kl": 0.3628700077533722,
+      "learning_rate": 4.279537612659893e-06,
+      "loss": 0.2747,
+      "step": 119,
+      "step_loss": 0.29215359687805176
+    },
+    {
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.7676948672131104,
+      "kl": 0.32500603795051575,
+      "learning_rate": 4.267580315316156e-06,
+      "loss": 0.2692,
+      "step": 120,
+      "step_loss": 0.2846797704696655
+    },
+    {
+      "epoch": 1.375,
+      "grad_norm": 0.7763972054682284,
+      "kl": 0.3070172071456909,
+      "learning_rate": 4.2555438643074315e-06,
+      "loss": 0.2748,
+      "step": 121,
+      "step_loss": 0.29722610116004944
+    },
+    {
+      "epoch": 1.3863636363636362,
+      "grad_norm": 0.765562794552249,
+      "kl": 0.24159571528434753,
+      "learning_rate": 4.243428887428635e-06,
+      "loss": 0.2648,
+      "step": 122,
+      "step_loss": 0.20421911776065826
+    },
+    {
+      "epoch": 1.3977272727272727,
+      "grad_norm": 0.7829115174919172,
+      "kl": 0.1845472753047943,
+      "learning_rate": 4.231236016570425e-06,
+      "loss": 0.2671,
+      "step": 123,
+      "step_loss": 0.24147802591323853
+    },
+    {
+      "epoch": 1.4090909090909092,
+      "grad_norm": 0.8644726847122038,
+      "kl": 0.24288871884346008,
+      "learning_rate": 4.218965887686236e-06,
+      "loss": 0.2914,
+      "step": 124,
+      "step_loss": 0.25647714734077454
+    },
+    {
+      "epoch": 1.4204545454545454,
+      "grad_norm": 0.8339544286421284,
+      "kl": 0.231180340051651,
+      "learning_rate": 4.206619140759113e-06,
+      "loss": 0.2704,
+      "step": 125,
+      "step_loss": 0.22522114217281342
+    },
+    {
+      "epoch": 1.4318181818181819,
+      "grad_norm": 0.8221761695605967,
+      "kl": 0.2961800992488861,
+      "learning_rate": 4.1941964197683294e-06,
+      "loss": 0.2769,
+      "step": 126,
+      "step_loss": 0.3132496774196625
+    },
+    {
+      "epoch": 1.4431818181818181,
+      "grad_norm": 0.8281811749101126,
+      "kl": 0.228164404630661,
+      "learning_rate": 4.181698372655802e-06,
+      "loss": 0.296,
+      "step": 127,
+      "step_loss": 0.2789953351020813
+    },
+    {
+      "epoch": 1.4545454545454546,
+      "grad_norm": 0.7805168144692464,
+      "kl": 0.29118138551712036,
+      "learning_rate": 4.16912565129229e-06,
+      "loss": 0.2691,
+      "step": 128,
+      "step_loss": 0.2938961982727051
+    },
+    {
+      "epoch": 1.4659090909090908,
+      "grad_norm": 0.764790917540533,
+      "kl": 0.21116551756858826,
+      "learning_rate": 4.156478911443399e-06,
+      "loss": 0.2728,
+      "step": 129,
+      "step_loss": 0.19998157024383545
+    },
+    {
+      "epoch": 1.4772727272727273,
+      "grad_norm": 0.8182095775205943,
+      "kl": 0.2947465479373932,
+      "learning_rate": 4.143758812735377e-06,
+      "loss": 0.276,
+      "step": 130,
+      "step_loss": 0.23230603337287903
+    },
+    {
+      "epoch": 1.4886363636363638,
+      "grad_norm": 0.7882946546339218,
+      "kl": 0.24360516667366028,
+      "learning_rate": 4.130966018620709e-06,
+      "loss": 0.2677,
+      "step": 131,
+      "step_loss": 0.2349737137556076
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.8656008871968565,
+      "kl": 0.35519665479660034,
+      "learning_rate": 4.118101196343515e-06,
+      "loss": 0.2729,
+      "step": 132,
+      "step_loss": 0.26551371812820435
+    },
+    {
+      "epoch": 1.5113636363636362,
+      "grad_norm": 0.8133559486757417,
+      "kl": 0.31251227855682373,
+      "learning_rate": 4.105165016904744e-06,
+      "loss": 0.2703,
+      "step": 133,
+      "step_loss": 0.30454349517822266
+    },
+    {
+      "epoch": 1.5227272727272727,
+      "grad_norm": 0.7341218877408442,
+      "kl": 0.3486112952232361,
+      "learning_rate": 4.092158155027177e-06,
+      "loss": 0.2778,
+      "step": 134,
+      "step_loss": 0.3166317045688629
+    },
+    {
+      "epoch": 1.5340909090909092,
+      "grad_norm": 0.8500600386901499,
+      "kl": 0.26721635460853577,
+      "learning_rate": 4.07908128912024e-06,
+      "loss": 0.2822,
+      "step": 135,
+      "step_loss": 0.31385499238967896
+    },
+    {
+      "epoch": 1.5454545454545454,
+      "grad_norm": 0.7849382316075052,
+      "kl": 0.25328248739242554,
+      "learning_rate": 4.065935101244614e-06,
+      "loss": 0.2798,
+      "step": 136,
+      "step_loss": 0.3054334223270416
+    },
+    {
+      "epoch": 1.5568181818181817,
+      "grad_norm": 0.8045542338117103,
+      "kl": 0.22720679640769958,
+      "learning_rate": 4.05272027707666e-06,
+      "loss": 0.2688,
+      "step": 137,
+      "step_loss": 0.2093893587589264
+    },
+    {
+      "epoch": 1.5681818181818183,
+      "grad_norm": 0.8059325354646184,
+      "kl": 0.33818334341049194,
+      "learning_rate": 4.039437505872661e-06,
+      "loss": 0.2755,
+      "step": 138,
+      "step_loss": 0.25241315364837646
+    },
+    {
+      "epoch": 1.5795454545454546,
+      "grad_norm": 0.7947964956021049,
+      "kl": 0.2659541666507721,
+      "learning_rate": 4.026087480432867e-06,
+      "loss": 0.2714,
+      "step": 139,
+      "step_loss": 0.22505882382392883
+    },
+    {
+      "epoch": 1.5909090909090908,
+      "grad_norm": 0.7798080488748865,
+      "kl": 0.3098128139972687,
+      "learning_rate": 4.012670897065361e-06,
+      "loss": 0.2645,
+      "step": 140,
+      "step_loss": 0.30545055866241455
+    },
+    {
+      "epoch": 1.6022727272727273,
+      "grad_norm": 0.775197771433234,
+      "kl": 0.22397831082344055,
+      "learning_rate": 3.999188455549743e-06,
+      "loss": 0.2666,
+      "step": 141,
+      "step_loss": 0.23903930187225342
+    },
+    {
+      "epoch": 1.6136363636363638,
+      "grad_norm": 0.8077084029693482,
+      "kl": 0.27500656247138977,
+      "learning_rate": 3.98564085910063e-06,
+      "loss": 0.2722,
+      "step": 142,
+      "step_loss": 0.35808488726615906
+    },
+    {
+      "epoch": 1.625,
+      "grad_norm": 0.8147386009423194,
+      "kl": 0.267314612865448,
+      "learning_rate": 3.972028814330974e-06,
+      "loss": 0.2723,
+      "step": 143,
+      "step_loss": 0.22990119457244873
+    },
+    {
+      "epoch": 1.6363636363636362,
+      "grad_norm": 0.8458877608603808,
+      "kl": 0.23753684759140015,
+      "learning_rate": 3.958353031215216e-06,
+      "loss": 0.2876,
+      "step": 144,
+      "step_loss": 0.2537928521633148
+    },
+    {
+      "epoch": 1.6477272727272727,
+      "grad_norm": 0.8197579462143825,
+      "kl": 0.23161743581295013,
+      "learning_rate": 3.944614223052246e-06,
+      "loss": 0.2801,
+      "step": 145,
+      "step_loss": 0.2637965679168701
+    },
+    {
+      "epoch": 1.6590909090909092,
+      "grad_norm": 0.8314592069504814,
+      "kl": 0.24659006297588348,
+      "learning_rate": 3.930813106428202e-06,
+      "loss": 0.2817,
+      "step": 146,
+      "step_loss": 0.2755964696407318
+    },
+    {
+      "epoch": 1.6704545454545454,
+      "grad_norm": 0.7638963920533027,
+      "kl": 0.20747093856334686,
+      "learning_rate": 3.916950401179096e-06,
+      "loss": 0.2635,
+      "step": 147,
+      "step_loss": 0.22498083114624023
+    },
+    {
+      "epoch": 1.6818181818181817,
+      "grad_norm": 0.8172422080755228,
+      "kl": 0.3187533915042877,
+      "learning_rate": 3.903026830353268e-06,
+      "loss": 0.273,
+      "step": 148,
+      "step_loss": 0.3263009190559387
+    },
+    {
+      "epoch": 1.6931818181818183,
+      "grad_norm": 0.8620536672122645,
+      "kl": 0.27937936782836914,
+      "learning_rate": 3.8890431201736715e-06,
+      "loss": 0.2776,
+      "step": 149,
+      "step_loss": 0.3091718554496765
+    },
+    {
+      "epoch": 1.7045454545454546,
+      "grad_norm": 0.8127977485887442,
+      "kl": 0.3105573356151581,
+      "learning_rate": 3.875e-06,
+      "loss": 0.2668,
+      "step": 150,
+      "step_loss": 0.3028494715690613
+    },
+    {
+      "epoch": 1.7159090909090908,
+      "grad_norm": 0.8201951442514309,
+      "kl": 0.3465476334095001,
+      "learning_rate": 3.86089820229064e-06,
+      "loss": 0.2791,
+      "step": 151,
+      "step_loss": 0.31831085681915283
+    },
+    {
+      "epoch": 1.7272727272727273,
+      "grad_norm": 0.835516351157446,
+      "kl": 0.21952536702156067,
+      "learning_rate": 3.846738462564469e-06,
+      "loss": 0.2721,
+      "step": 152,
+      "step_loss": 0.2234291136264801
+    },
+    {
+      "epoch": 1.7386363636363638,
+      "grad_norm": 0.8665975677625136,
+      "kl": 0.29508858919143677,
+      "learning_rate": 3.8325215193624945e-06,
+      "loss": 0.2802,
+      "step": 153,
+      "step_loss": 0.3162890076637268
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 0.8064955980472845,
+      "kl": 0.32985877990722656,
+      "learning_rate": 3.8182481142093315e-06,
+      "loss": 0.2566,
+      "step": 154,
+      "step_loss": 0.2641649842262268
+    },
+    {
+      "epoch": 1.7613636363636362,
+      "grad_norm": 0.826451499999824,
+      "kl": 0.2728135287761688,
+      "learning_rate": 3.8039189915745286e-06,
+      "loss": 0.292,
+      "step": 155,
+      "step_loss": 0.32532957196235657
+    },
+    {
+      "epoch": 1.7727272727272727,
+      "grad_norm": 0.8336078231467312,
+      "kl": 0.38967686891555786,
+      "learning_rate": 3.789534898833732e-06,
+      "loss": 0.278,
+      "step": 156,
+      "step_loss": 0.3321007192134857
+    },
+    {
+      "epoch": 1.7840909090909092,
+      "grad_norm": 0.7893976689434017,
+      "kl": 0.2202274650335312,
+      "learning_rate": 3.775096586229713e-06,
+      "loss": 0.2696,
+      "step": 157,
+      "step_loss": 0.2540920376777649
+    },
+    {
+      "epoch": 1.7954545454545454,
+      "grad_norm": 0.9319185366269114,
+      "kl": 0.4245775043964386,
+      "learning_rate": 3.760604806833228e-06,
+      "loss": 0.288,
+      "step": 158,
+      "step_loss": 0.35646578669548035
+    },
+    {
+      "epoch": 1.8068181818181817,
+      "grad_norm": 0.744000278965067,
+      "kl": 0.21765930950641632,
+      "learning_rate": 3.7460603165037475e-06,
+      "loss": 0.262,
+      "step": 159,
+      "step_loss": 0.2947049140930176
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "grad_norm": 0.7890825930010562,
+      "kl": 0.2992360591888428,
+      "learning_rate": 3.7314638738500275e-06,
+      "loss": 0.2797,
+      "step": 160,
+      "step_loss": 0.30612078309059143
+    },
+    {
+      "epoch": 1.8295454545454546,
+      "grad_norm": 0.806422206319446,
+      "kl": 0.2850240468978882,
+      "learning_rate": 3.7168162401905415e-06,
+      "loss": 0.2643,
+      "step": 161,
+      "step_loss": 0.25350436568260193
+    },
+    {
+      "epoch": 1.8409090909090908,
+      "grad_norm": 0.7772092463231446,
+      "kl": 0.294902503490448,
+      "learning_rate": 3.7021181795137777e-06,
+      "loss": 0.2752,
+      "step": 162,
+      "step_loss": 0.3195292353630066
+    },
+    {
+      "epoch": 1.8522727272727273,
+      "grad_norm": 0.8137138542667633,
+      "kl": 0.3464955985546112,
+      "learning_rate": 3.6873704584383834e-06,
+      "loss": 0.2879,
+      "step": 163,
+      "step_loss": 0.3319052755832672
+    },
+    {
+      "epoch": 1.8636363636363638,
+      "grad_norm": 0.8158519268417352,
+      "kl": 0.2651825547218323,
+      "learning_rate": 3.6725738461731865e-06,
+      "loss": 0.2698,
+      "step": 164,
+      "step_loss": 0.3068053722381592
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 0.7751669380913654,
+      "kl": 0.2937452793121338,
+      "learning_rate": 3.65772911447707e-06,
+      "loss": 0.2701,
+      "step": 165,
+      "step_loss": 0.2569812834262848
+    },
+    {
+      "epoch": 1.8863636363636362,
+      "grad_norm": 0.7494161240442716,
+      "kl": 0.27101945877075195,
+      "learning_rate": 3.6428370376187212e-06,
+      "loss": 0.2756,
+      "step": 166,
+      "step_loss": 0.3467923402786255
+    },
+    {
+      "epoch": 1.8977272727272727,
+      "grad_norm": 0.8176861818718908,
+      "kl": 0.27005308866500854,
+      "learning_rate": 3.627898392336248e-06,
+      "loss": 0.2652,
+      "step": 167,
+      "step_loss": 0.27606719732284546
+    },
+    {
+      "epoch": 1.9090909090909092,
+      "grad_norm": 0.7568912608938045,
+      "kl": 0.23509696125984192,
+      "learning_rate": 3.6129139577966665e-06,
+      "loss": 0.2764,
+      "step": 168,
+      "step_loss": 0.27201592922210693
+    },
+    {
+      "epoch": 1.9204545454545454,
+      "grad_norm": 0.8229709961416729,
+      "kl": 0.28099095821380615,
+      "learning_rate": 3.597884515555258e-06,
+      "loss": 0.2685,
+      "step": 169,
+      "step_loss": 0.29871946573257446
+    },
+    {
+      "epoch": 1.9318181818181817,
+      "grad_norm": 0.7594413163226815,
+      "kl": 0.2222573161125183,
+      "learning_rate": 3.5828108495148077e-06,
+      "loss": 0.2601,
+      "step": 170,
+      "step_loss": 0.262919157743454
+    },
+    {
+      "epoch": 1.9431818181818183,
+      "grad_norm": 0.7658440412811187,
+      "kl": 0.2369879186153412,
+      "learning_rate": 3.5676937458847177e-06,
+      "loss": 0.2898,
+      "step": 171,
+      "step_loss": 0.2267894744873047
+    },
+    {
+      "epoch": 1.9545454545454546,
+      "grad_norm": 0.7855144082392459,
+      "kl": 0.24906377494335175,
+      "learning_rate": 3.55253399314e-06,
+      "loss": 0.2927,
+      "step": 172,
+      "step_loss": 0.33061373233795166
+    },
+    {
+      "epoch": 1.9659090909090908,
+      "grad_norm": 0.793142140040205,
+      "kl": 0.24075445532798767,
+      "learning_rate": 3.5373323819801493e-06,
+      "loss": 0.2788,
+      "step": 173,
+      "step_loss": 0.3006718158721924
+    },
+    {
+      "epoch": 1.9772727272727273,
+      "grad_norm": 0.7402321203997657,
+      "kl": 0.2078879177570343,
+      "learning_rate": 3.5220897052879047e-06,
+      "loss": 0.283,
+      "step": 174,
+      "step_loss": 0.2104087769985199
+    },
+    {
+      "epoch": 1.9886363636363638,
+      "grad_norm": 0.7547346712459372,
+      "kl": 0.20877879858016968,
+      "learning_rate": 3.5068067580878947e-06,
+      "loss": 0.2724,
+      "step": 175,
+      "step_loss": 0.23320713639259338
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.7771753483379468,
+      "kl": 0.26442065834999084,
+      "learning_rate": 3.491484337505166e-06,
+      "loss": 0.268,
+      "step": 176,
+      "step_loss": 0.2616370916366577
+    },
+    {
+      "epoch": 2.0,
+      "eval_test_transformed.json_loss": 0.6573941707611084,
+      "eval_test_transformed.json_runtime": 57.072,
+      "eval_test_transformed.json_samples_per_second": 8.761,
+      "eval_test_transformed.json_steps_per_second": 0.561,
+      "step": 176
+    },
+    {
+      "epoch": 2.0113636363636362,
+      "grad_norm": 0.8267860023888801,
+      "kl": 0.10590674728155136,
+      "learning_rate": 3.476123242723614e-06,
+      "loss": 0.2097,
+      "step": 177,
+      "step_loss": 0.5581972599029541
+    },
+    {
+      "epoch": 2.022727272727273,
+      "grad_norm": 0.856141962071338,
+      "kl": 0.3062549829483032,
+      "learning_rate": 3.460724274944294e-06,
+      "loss": 0.2173,
+      "step": 178,
+      "step_loss": 0.23314087092876434
+    },
+    {
+      "epoch": 2.034090909090909,
+      "grad_norm": 0.7361155640769321,
+      "kl": 0.3152432143688202,
+      "learning_rate": 3.445288237343632e-06,
+      "loss": 0.1978,
+      "step": 179,
+      "step_loss": 0.20342102646827698
+    },
+    {
+      "epoch": 2.0454545454545454,
+      "grad_norm": 0.732358232912068,
+      "kl": 0.382126122713089,
+      "learning_rate": 3.4298159350315398e-06,
+      "loss": 0.2032,
+      "step": 180,
+      "step_loss": 0.24822968244552612
+    },
+    {
+      "epoch": 2.0568181818181817,
+      "grad_norm": 0.7942400733536357,
+      "kl": 0.3069852590560913,
+      "learning_rate": 3.4143081750094107e-06,
+      "loss": 0.1976,
+      "step": 181,
+      "step_loss": 0.21024365723133087
+    },
+    {
+      "epoch": 2.0681818181818183,
+      "grad_norm": 0.9225206215680616,
+      "kl": 0.4743329882621765,
+      "learning_rate": 3.3987657661280393e-06,
+      "loss": 0.2133,
+      "step": 182,
+      "step_loss": 0.22629733383655548
+    },
+    {
+      "epoch": 2.0795454545454546,
+      "grad_norm": 0.9790953481687389,
+      "kl": 0.31061556935310364,
+      "learning_rate": 3.383189519045428e-06,
+      "loss": 0.2031,
+      "step": 183,
+      "step_loss": 0.21077898144721985
+    },
+    {
+      "epoch": 2.090909090909091,
+      "grad_norm": 0.9744424313842567,
+      "kl": 0.380555659532547,
+      "learning_rate": 3.3675802461845036e-06,
+      "loss": 0.2004,
+      "step": 184,
+      "step_loss": 0.22242428362369537
+    },
+    {
+      "epoch": 2.102272727272727,
+      "grad_norm": 0.916179520314943,
+      "kl": 0.43431419134140015,
+      "learning_rate": 3.351938761690747e-06,
+      "loss": 0.1966,
+      "step": 185,
+      "step_loss": 0.2207513451576233
+    },
+    {
+      "epoch": 2.1136363636363638,
+      "grad_norm": 0.8004844301509703,
+      "kl": 0.28789108991622925,
+      "learning_rate": 3.3362658813897274e-06,
+      "loss": 0.1942,
+      "step": 186,
+      "step_loss": 0.26408788561820984
+    },
+    {
+      "epoch": 2.125,
+      "grad_norm": 0.8955852845271478,
+      "kl": 0.28126734495162964,
+      "learning_rate": 3.320562422744548e-06,
+      "loss": 0.2053,
+      "step": 187,
+      "step_loss": 0.2404702603816986
+    },
+    {
+      "epoch": 2.1363636363636362,
+      "grad_norm": 0.8109199492626206,
+      "kl": 0.31442955136299133,
+      "learning_rate": 3.3048292048132153e-06,
+      "loss": 0.2011,
+      "step": 188,
+      "step_loss": 0.21545666456222534
+    },
+    {
+      "epoch": 2.147727272727273,
+      "grad_norm": 0.8741546702827006,
+      "kl": 0.2675096094608307,
+      "learning_rate": 3.289067048205912e-06,
+      "loss": 0.2088,
+      "step": 189,
+      "step_loss": 0.1575331836938858
+    },
+    {
+      "epoch": 2.159090909090909,
+      "grad_norm": 0.8080813493615443,
+      "kl": 0.3244311213493347,
+      "learning_rate": 3.273276775042199e-06,
+      "loss": 0.2003,
+      "step": 190,
+      "step_loss": 0.187534362077713
+    },
+    {
+      "epoch": 2.1704545454545454,
+      "grad_norm": 0.8677916956527729,
+      "kl": 0.2513880729675293,
+      "learning_rate": 3.2574592089081375e-06,
+      "loss": 0.1989,
+      "step": 191,
+      "step_loss": 0.17420931160449982
+    },
+    {
+      "epoch": 2.1818181818181817,
+      "grad_norm": 0.798718321325761,
+      "kl": 0.291971892118454,
+      "learning_rate": 3.241615174813327e-06,
+      "loss": 0.207,
+      "step": 192,
+      "step_loss": 0.19951367378234863
+    },
+    {
+      "epoch": 2.1931818181818183,
+      "grad_norm": 0.8266125308725523,
+      "kl": 0.3032814562320709,
+      "learning_rate": 3.2257454991478825e-06,
+      "loss": 0.1893,
+      "step": 193,
+      "step_loss": 0.17084187269210815
+    },
+    {
+      "epoch": 2.2045454545454546,
+      "grad_norm": 0.7812137976841603,
+      "kl": 0.38334983587265015,
+      "learning_rate": 3.2098510096393214e-06,
+      "loss": 0.2027,
+      "step": 194,
+      "step_loss": 0.20809711515903473
+    },
+    {
+      "epoch": 2.215909090909091,
+      "grad_norm": 0.9554319759849772,
+      "kl": 0.3265298306941986,
+      "learning_rate": 3.1939325353094004e-06,
+      "loss": 0.2052,
+      "step": 195,
+      "step_loss": 0.19203175604343414
+    },
+    {
+      "epoch": 2.227272727272727,
+      "grad_norm": 0.8249089393067537,
+      "kl": 0.35938209295272827,
+      "learning_rate": 3.177990906430871e-06,
+      "loss": 0.1967,
+      "step": 196,
+      "step_loss": 0.2233177125453949
+    },
+    {
+      "epoch": 2.2386363636363638,
+      "grad_norm": 0.7933030950436119,
+      "kl": 0.32349923253059387,
+      "learning_rate": 3.1620269544841764e-06,
+      "loss": 0.1913,
+      "step": 197,
+      "step_loss": 0.1729939877986908
+    },
+    {
+      "epoch": 2.25,
+      "grad_norm": 0.8197164640788506,
+      "kl": 0.3547500669956207,
+      "learning_rate": 3.1460415121140804e-06,
+      "loss": 0.2006,
+      "step": 198,
+      "step_loss": 0.2586993873119354
+    },
+    {
+      "epoch": 2.2613636363636362,
+      "grad_norm": 0.8495735950170649,
+      "kl": 0.3509565591812134,
+      "learning_rate": 3.1300354130862404e-06,
+      "loss": 0.2045,
+      "step": 199,
+      "step_loss": 0.21880128979682922
+    },
+    {
+      "epoch": 2.2727272727272725,
+      "grad_norm": 0.7758237844371901,
+      "kl": 0.3566822409629822,
+      "learning_rate": 3.1140094922437213e-06,
+      "loss": 0.1906,
+      "step": 200,
+      "step_loss": 0.15825271606445312
+    },
+    {
+      "epoch": 2.284090909090909,
+      "grad_norm": 0.7748859598486534,
+      "kl": 0.23674319684505463,
+      "learning_rate": 3.097964585463449e-06,
+      "loss": 0.1904,
+      "step": 201,
+      "step_loss": 0.15398302674293518
+    },
+    {
+      "epoch": 2.2954545454545454,
+      "grad_norm": 0.8150881114674452,
+      "kl": 0.32412102818489075,
+      "learning_rate": 3.081901529612614e-06,
+      "loss": 0.2015,
+      "step": 202,
+      "step_loss": 0.1894824057817459
+    },
+    {
+      "epoch": 2.3068181818181817,
+      "grad_norm": 0.777056310216207,
+      "kl": 0.33722126483917236,
+      "learning_rate": 3.065821162505024e-06,
+      "loss": 0.2094,
+      "step": 203,
+      "step_loss": 0.24027691781520844
+    },
+    {
+      "epoch": 2.3181818181818183,
+      "grad_norm": 0.7521221020569652,
+      "kl": 0.30370503664016724,
+      "learning_rate": 3.0497243228574054e-06,
+      "loss": 0.2033,
+      "step": 204,
+      "step_loss": 0.2159876674413681
+    },
+    {
+      "epoch": 2.3295454545454546,
+      "grad_norm": 0.8056198200446986,
+      "kl": 0.27457597851753235,
+      "learning_rate": 3.0336118502456515e-06,
+      "loss": 0.2057,
+      "step": 205,
+      "step_loss": 0.1882285177707672
+    },
+    {
+      "epoch": 2.340909090909091,
+      "grad_norm": 0.7864857515183022,
+      "kl": 0.35345762968063354,
+      "learning_rate": 3.0174845850610395e-06,
+      "loss": 0.1896,
+      "step": 206,
+      "step_loss": 0.1616518199443817
+    },
+    {
+      "epoch": 2.3522727272727275,
+      "grad_norm": 0.7347377124860899,
+      "kl": 0.26013386249542236,
+      "learning_rate": 3.001343368466396e-06,
+      "loss": 0.1931,
+      "step": 207,
+      "step_loss": 0.15433846414089203
+    },
+    {
+      "epoch": 2.3636363636363638,
+      "grad_norm": 0.8262277498582172,
+      "kl": 0.41462039947509766,
+      "learning_rate": 2.9851890423522205e-06,
+      "loss": 0.1946,
+      "step": 208,
+      "step_loss": 0.17125703394412994
+    },
+    {
+      "epoch": 2.375,
+      "grad_norm": 0.8440169159473123,
+      "kl": 0.30189239978790283,
+      "learning_rate": 2.9690224492927777e-06,
+      "loss": 0.1972,
+      "step": 209,
+      "step_loss": 0.1643592119216919
+    },
+    {
+      "epoch": 2.3863636363636362,
+      "grad_norm": 0.7902010267144333,
+      "kl": 0.30659422278404236,
+      "learning_rate": 2.9528444325021475e-06,
+      "loss": 0.196,
+      "step": 210,
+      "step_loss": 0.16432619094848633
+    },
+    {
+      "epoch": 2.3977272727272725,
+      "grad_norm": 0.7897263557587403,
+      "kl": 0.35797473788261414,
+      "learning_rate": 2.9366558357902485e-06,
+      "loss": 0.2104,
+      "step": 211,
+      "step_loss": 0.18551866710186005
+    },
+    {
+      "epoch": 2.409090909090909,
+      "grad_norm": 0.8139541524586551,
+      "kl": 0.5444263815879822,
+      "learning_rate": 2.920457503518824e-06,
+      "loss": 0.2149,
+      "step": 212,
+      "step_loss": 0.2275647521018982
+    },
+    {
+      "epoch": 2.4204545454545454,
+      "grad_norm": 0.779793908267749,
+      "kl": 0.26403847336769104,
+      "learning_rate": 2.9042502805574022e-06,
+      "loss": 0.195,
+      "step": 213,
+      "step_loss": 0.16895370185375214
+    },
+    {
+      "epoch": 2.4318181818181817,
+      "grad_norm": 0.8129577814286397,
+      "kl": 0.3593398630619049,
+      "learning_rate": 2.8880350122392327e-06,
+      "loss": 0.1991,
+      "step": 214,
+      "step_loss": 0.19971898198127747
+    },
+    {
+      "epoch": 2.4431818181818183,
+      "grad_norm": 0.819141896928592,
+      "kl": 0.40604791045188904,
+      "learning_rate": 2.87181254431719e-06,
+      "loss": 0.1916,
+      "step": 215,
+      "step_loss": 0.19662940502166748
+    },
+    {
+      "epoch": 2.4545454545454546,
+      "grad_norm": 0.8326566611196083,
+      "kl": 0.30143654346466064,
+      "learning_rate": 2.8555837229196654e-06,
+      "loss": 0.2012,
+      "step": 216,
+      "step_loss": 0.19791531562805176
+    },
+    {
+      "epoch": 2.465909090909091,
+      "grad_norm": 0.7971460218117707,
+      "kl": 0.28109458088874817,
+      "learning_rate": 2.839349394506435e-06,
+      "loss": 0.2047,
+      "step": 217,
+      "step_loss": 0.1851027011871338
+    },
+    {
+      "epoch": 2.4772727272727275,
+      "grad_norm": 0.8160202720148206,
+      "kl": 0.3231932520866394,
+      "learning_rate": 2.8231104058245066e-06,
+      "loss": 0.2046,
+      "step": 218,
+      "step_loss": 0.1588630974292755
+    },
+    {
+      "epoch": 2.4886363636363638,
+      "grad_norm": 0.8870641900955384,
+      "kl": 0.39819225668907166,
+      "learning_rate": 2.8068676038639584e-06,
+      "loss": 0.2045,
+      "step": 219,
+      "step_loss": 0.21049059927463531
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.8580982781435952,
+      "kl": 0.21048355102539062,
+      "learning_rate": 2.790621835813761e-06,
+      "loss": 0.1954,
+      "step": 220,
+      "step_loss": 0.15752212703227997
+    },
+    {
+      "epoch": 2.5113636363636362,
+      "grad_norm": 0.8332029161126598,
+      "kl": 0.3022167980670929,
+      "learning_rate": 2.7743739490175902e-06,
+      "loss": 0.1993,
+      "step": 221,
+      "step_loss": 0.18184006214141846
+    },
+    {
+      "epoch": 2.5227272727272725,
+      "grad_norm": 0.7915764221822466,
+      "kl": 0.3532565236091614,
+      "learning_rate": 2.75812479092963e-06,
+      "loss": 0.2028,
+      "step": 222,
+      "step_loss": 0.2090645432472229
+    },
+    {
+      "epoch": 2.534090909090909,
+      "grad_norm": 0.816127163580774,
+      "kl": 0.3357287645339966,
+      "learning_rate": 2.7418752090703716e-06,
+      "loss": 0.1993,
+      "step": 223,
+      "step_loss": 0.22028760612010956
+    },
+    {
+      "epoch": 2.5454545454545454,
+      "grad_norm": 0.8169798227897933,
+      "kl": 0.3894638121128082,
+      "learning_rate": 2.72562605098241e-06,
+      "loss": 0.2095,
+      "step": 224,
+      "step_loss": 0.3154832422733307
+    },
+    {
+      "epoch": 2.5568181818181817,
+      "grad_norm": 0.8178116233042594,
+      "kl": 0.4051194190979004,
+      "learning_rate": 2.7093781641862387e-06,
+      "loss": 0.215,
+      "step": 225,
+      "step_loss": 0.21542856097221375
+    },
+    {
+      "epoch": 2.5681818181818183,
+      "grad_norm": 0.7927440832417105,
+      "kl": 0.48752185702323914,
+      "learning_rate": 2.6931323961360423e-06,
+      "loss": 0.1982,
+      "step": 226,
+      "step_loss": 0.2092168629169464
+    },
+    {
+      "epoch": 2.5795454545454546,
+      "grad_norm": 0.8036623600570535,
+      "kl": 0.25240644812583923,
+      "learning_rate": 2.6768895941754945e-06,
+      "loss": 0.1875,
+      "step": 227,
+      "step_loss": 0.1589215099811554
+    },
+    {
+      "epoch": 2.590909090909091,
+      "grad_norm": 0.8374058314659513,
+      "kl": 0.3620951473712921,
+      "learning_rate": 2.660650605493566e-06,
+      "loss": 0.1882,
+      "step": 228,
+      "step_loss": 0.19974978268146515
+    },
+    {
+      "epoch": 2.6022727272727275,
+      "grad_norm": 0.8335380700060886,
+      "kl": 0.41594362258911133,
+      "learning_rate": 2.644416277080335e-06,
+      "loss": 0.2165,
+      "step": 229,
+      "step_loss": 0.24422059953212738
+    },
+    {
+      "epoch": 2.6136363636363638,
+      "grad_norm": 0.8100345325173146,
+      "kl": 0.4320451319217682,
+      "learning_rate": 2.6281874556828106e-06,
+      "loss": 0.2252,
+      "step": 230,
+      "step_loss": 0.2085280418395996
+    },
+    {
+      "epoch": 2.625,
+      "grad_norm": 0.8530910666847069,
+      "kl": 0.3395635783672333,
+      "learning_rate": 2.6119649877607684e-06,
+      "loss": 0.2006,
+      "step": 231,
+      "step_loss": 0.18190529942512512
+    },
+    {
+      "epoch": 2.6363636363636362,
+      "grad_norm": 0.7566154700308811,
+      "kl": 0.3144618272781372,
+      "learning_rate": 2.5957497194425985e-06,
+      "loss": 0.1887,
+      "step": 232,
+      "step_loss": 0.18467919528484344
+    },
+    {
+      "epoch": 2.6477272727272725,
+      "grad_norm": 0.7310945410825066,
+      "kl": 0.3082126975059509,
+      "learning_rate": 2.579542496481178e-06,
+      "loss": 0.1922,
+      "step": 233,
+      "step_loss": 0.21097740530967712
+    },
+    {
+      "epoch": 2.659090909090909,
+      "grad_norm": 0.7837603862506226,
+      "kl": 0.30413469672203064,
+      "learning_rate": 2.5633441642097527e-06,
+      "loss": 0.204,
+      "step": 234,
+      "step_loss": 0.2184579074382782
+    },
+    {
+      "epoch": 2.6704545454545454,
+      "grad_norm": 0.7849444203230034,
+      "kl": 0.32521089911460876,
+      "learning_rate": 2.5471555674978536e-06,
+      "loss": 0.1902,
+      "step": 235,
+      "step_loss": 0.1784965693950653
+    },
+    {
+      "epoch": 2.6818181818181817,
+      "grad_norm": 0.8978400192531096,
+      "kl": 0.4359068274497986,
+      "learning_rate": 2.530977550707223e-06,
+      "loss": 0.1998,
+      "step": 236,
+      "step_loss": 0.24568131566047668
+    },
+    {
+      "epoch": 2.6931818181818183,
+      "grad_norm": 0.8296558697377163,
+      "kl": 0.34526827931404114,
+      "learning_rate": 2.5148109576477798e-06,
+      "loss": 0.2051,
+      "step": 237,
+      "step_loss": 0.1845228672027588
+    },
+    {
+      "epoch": 2.7045454545454546,
+      "grad_norm": 0.8456638318243491,
+      "kl": 0.4374793767929077,
+      "learning_rate": 2.4986566315336047e-06,
+      "loss": 0.2099,
+      "step": 238,
+      "step_loss": 0.2226126492023468
+    },
+    {
+      "epoch": 2.715909090909091,
+      "grad_norm": 0.7855160396523918,
+      "kl": 0.35535740852355957,
+      "learning_rate": 2.4825154149389608e-06,
+      "loss": 0.2064,
+      "step": 239,
+      "step_loss": 0.2260066717863083
+    },
+    {
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.7687481608795249,
+      "kl": 0.3926181495189667,
+      "learning_rate": 2.4663881497543497e-06,
+      "loss": 0.1978,
+      "step": 240,
+      "step_loss": 0.183305025100708
+    },
+    {
+      "epoch": 2.7386363636363638,
+      "grad_norm": 0.8743610433319449,
+      "kl": 0.3105931580066681,
+      "learning_rate": 2.450275677142595e-06,
+      "loss": 0.1982,
+      "step": 241,
+      "step_loss": 0.1785978525876999
+    },
+    {
+      "epoch": 2.75,
+      "grad_norm": 0.8103030286089026,
+      "kl": 0.28116321563720703,
+      "learning_rate": 2.4341788374949753e-06,
+      "loss": 0.2028,
+      "step": 242,
+      "step_loss": 0.16758720576763153
+    },
+    {
+      "epoch": 2.7613636363636362,
+      "grad_norm": 0.7780046576774838,
+      "kl": 0.31231576204299927,
+      "learning_rate": 2.418098470387387e-06,
+      "loss": 0.204,
+      "step": 243,
+      "step_loss": 0.21919600665569305
+    },
+    {
+      "epoch": 2.7727272727272725,
+      "grad_norm": 0.7900342339932793,
+      "kl": 0.36211156845092773,
+      "learning_rate": 2.4020354145365526e-06,
+      "loss": 0.1925,
+      "step": 244,
+      "step_loss": 0.22099542617797852
+    },
+    {
+      "epoch": 2.784090909090909,
+      "grad_norm": 0.7782246567974114,
+      "kl": 0.24455486238002777,
+      "learning_rate": 2.3859905077562802e-06,
+      "loss": 0.1871,
+      "step": 245,
+      "step_loss": 0.13276106119155884
+    },
+    {
+      "epoch": 2.7954545454545454,
+      "grad_norm": 0.7780058068595647,
+      "kl": 0.3040383756160736,
+      "learning_rate": 2.36996458691376e-06,
+      "loss": 0.203,
+      "step": 246,
+      "step_loss": 0.18567970395088196
+    },
+    {
+      "epoch": 2.8068181818181817,
+      "grad_norm": 0.797452119268861,
+      "kl": 0.3017336428165436,
+      "learning_rate": 2.35395848788592e-06,
+      "loss": 0.2045,
+      "step": 247,
+      "step_loss": 0.18115977942943573
+    },
+    {
+      "epoch": 2.8181818181818183,
+      "grad_norm": 0.8291981778954278,
+      "kl": 0.31822529435157776,
+      "learning_rate": 2.337973045515824e-06,
+      "loss": 0.2063,
+      "step": 248,
+      "step_loss": 0.18958526849746704
+    },
+    {
+      "epoch": 2.8295454545454546,
+      "grad_norm": 0.8192468169601297,
+      "kl": 0.36042526364326477,
+      "learning_rate": 2.3220090935691284e-06,
+      "loss": 0.2078,
+      "step": 249,
+      "step_loss": 0.198805570602417
+    },
+    {
+      "epoch": 2.840909090909091,
+      "grad_norm": 0.8386142691384693,
+      "kl": 0.3129850924015045,
+      "learning_rate": 2.3060674646906007e-06,
+      "loss": 0.1976,
+      "step": 250,
+      "step_loss": 0.15486395359039307
+    },
+    {
+      "epoch": 2.8522727272727275,
+      "grad_norm": 0.8542094258716569,
+      "kl": 0.328252911567688,
+      "learning_rate": 2.2901489903606793e-06,
+      "loss": 0.2094,
+      "step": 251,
+      "step_loss": 0.22234201431274414
+    },
+    {
+      "epoch": 2.8636363636363638,
+      "grad_norm": 0.8235027946049533,
+      "kl": 0.33431679010391235,
+      "learning_rate": 2.274254500852118e-06,
+      "loss": 0.2108,
+      "step": 252,
+      "step_loss": 0.23872962594032288
+    },
+    {
+      "epoch": 2.875,
+      "grad_norm": 0.7826979380094895,
+      "kl": 0.35612499713897705,
+      "learning_rate": 2.258384825186673e-06,
+      "loss": 0.2018,
+      "step": 253,
+      "step_loss": 0.22147932648658752
+    },
+    {
+      "epoch": 2.8863636363636362,
+      "grad_norm": 0.7727887293657175,
+      "kl": 0.4257417321205139,
+      "learning_rate": 2.2425407910918632e-06,
+      "loss": 0.1996,
+      "step": 254,
+      "step_loss": 0.25106555223464966
+    },
+    {
+      "epoch": 2.8977272727272725,
+      "grad_norm": 0.7880191514694174,
+      "kl": 0.3257736265659332,
+      "learning_rate": 2.2267232249578026e-06,
+      "loss": 0.1997,
+      "step": 255,
+      "step_loss": 0.17997223138809204
+    },
+    {
+      "epoch": 2.909090909090909,
+      "grad_norm": 0.8229368102750961,
+      "kl": 0.32289835810661316,
+      "learning_rate": 2.2109329517940897e-06,
+      "loss": 0.2049,
+      "step": 256,
+      "step_loss": 0.19980621337890625
+    },
+    {
+      "epoch": 2.9204545454545454,
+      "grad_norm": 0.8649527821304737,
+      "kl": 0.4314434826374054,
+      "learning_rate": 2.1951707951867862e-06,
+      "loss": 0.2071,
+      "step": 257,
+      "step_loss": 0.19861182570457458
+    },
+    {
+      "epoch": 2.9318181818181817,
+      "grad_norm": 0.7631295075779413,
+      "kl": 0.2881592810153961,
+      "learning_rate": 2.1794375772554525e-06,
+      "loss": 0.195,
+      "step": 258,
+      "step_loss": 0.17643983662128448
+    },
+    {
+      "epoch": 2.9431818181818183,
+      "grad_norm": 0.8176775074066636,
+      "kl": 0.2888239324092865,
+      "learning_rate": 2.163734118610274e-06,
+      "loss": 0.2033,
+      "step": 259,
+      "step_loss": 0.22650499641895294
+    },
+    {
+      "epoch": 2.9545454545454546,
+      "grad_norm": 0.7857239601852497,
+      "kl": 0.39291509985923767,
+      "learning_rate": 2.148061238309253e-06,
+      "loss": 0.1934,
+      "step": 260,
+      "step_loss": 0.19493892788887024
+    },
+    {
+      "epoch": 2.965909090909091,
+      "grad_norm": 0.773095874284012,
+      "kl": 0.37562623620033264,
+      "learning_rate": 2.132419753815497e-06,
+      "loss": 0.1945,
+      "step": 261,
+      "step_loss": 0.2021428793668747
+    },
+    {
+      "epoch": 2.9772727272727275,
+      "grad_norm": 0.8313683748212116,
+      "kl": 0.3755533695220947,
+      "learning_rate": 2.116810480954573e-06,
+      "loss": 0.2116,
+      "step": 262,
+      "step_loss": 0.2229767143726349
+    },
+    {
+      "epoch": 2.9886363636363638,
+      "grad_norm": 0.7820076892211333,
+      "kl": 0.32565972208976746,
+      "learning_rate": 2.101234233871961e-06,
+      "loss": 0.2024,
+      "step": 263,
+      "step_loss": 0.2286100536584854
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7846662975927288,
+      "kl": 0.4572798013687134,
+      "learning_rate": 2.0856918249905904e-06,
+      "loss": 0.1966,
+      "step": 264,
+      "step_loss": 0.2366623431444168
+    },
+    {
+      "epoch": 3.0,
+      "eval_test_transformed.json_loss": 0.7202348113059998,
+      "eval_test_transformed.json_runtime": 57.1103,
+      "eval_test_transformed.json_samples_per_second": 8.755,
+      "eval_test_transformed.json_steps_per_second": 0.56,
+      "step": 264
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 440,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 50.0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 105951870517248.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:70070ea7b8e7f33977358fe366ff1be0ffa42366cc9f04bff068425d226b4779
+size 8888

zero_to_fp32.py

@@ -0,0 +1,587 @@
+#!/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):
+    """
+    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)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+
+
+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 _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()
+            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):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _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):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _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):
+    """
+    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``
+
+    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)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None):
+    """
+    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``
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+    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("-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)