Upload folder using huggingface_hub

.gitattributes

@@ -36,3 +36,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 0413-7B/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 link-0501-7B/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 0523-7B/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+0624-7B/tokenizer.json filter=lfs diff=lfs merge=lfs -text

0624-7B/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

0624-7B/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "/home/xulilin/Qwen-2.5-coder-instruct-7b",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "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,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 152064
+}

0624-7B/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": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.46.0"
+}

0624-7B/latest

@@ -0,0 +1 @@
+global_step400

0624-7B/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d00da5ae15d596be26abab13383dc215942993fcc7094974610bda031a9f24a
+size 4877660776

0624-7B/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f6ca0399ff4bb41d6e86f355805794f0865d7e7a4e8fac7591b6da3a976c08be
+size 4932751008

0624-7B/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1dba7682c4e17355e227713e0eaa60f72fab0cfba57efbe39632f6cee54b42c
+size 4330865200

0624-7B/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06279031a34337ec981246cb3c0df3150e9186e49a09c8ab0772e7ce9407f0c0
+size 1089994880

0624-7B/model.safetensors.index.json

@@ -0,0 +1,346 @@
+{
+  "metadata": {
+    "total_size": 15231233024
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00003-of-00004.safetensors"
+  }
+}

0624-7B/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cb795a5cea0baa625c50007a6c9da09c6bbb5c16b560424070384a479e7d8a6
+size 14512

0624-7B/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f19604377bd828eb366c68946ad997a4ff4d69beaeea93ee58915135768ec63
+size 14512

0624-7B/special_tokens_map.json

@@ -0,0 +1,32 @@
+{
+  "additional_special_tokens": [
+    {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false
+    },
+    {
+      "content": "<|im_start|>",
+      "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
+  }
+}

0624-7B/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

0624-7B/tokenizer_config.json

@@ -0,0 +1,198 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "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
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_end|>",
+    "<|im_start|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 2048,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "truncation": true,
+  "unk_token": null
+}

0624-7B/trainer_state.json

@@ -0,0 +1,2833 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.6,
+  "eval_steps": 500,
+  "global_step": 400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.004,
+      "grad_norm": 4.334173679351807,
+      "learning_rate": 0.0,
+      "loss": 0.6307,
+      "step": 1
+    },
+    {
+      "epoch": 0.008,
+      "grad_norm": 2.2055771350860596,
+      "learning_rate": 7.525749891599529e-06,
+      "loss": 0.2344,
+      "step": 2
+    },
+    {
+      "epoch": 0.012,
+      "grad_norm": 2.218201160430908,
+      "learning_rate": 1.192803136799156e-05,
+      "loss": 0.2053,
+      "step": 3
+    },
+    {
+      "epoch": 0.016,
+      "grad_norm": 1.4988631010055542,
+      "learning_rate": 1.5051499783199057e-05,
+      "loss": 0.2045,
+      "step": 4
+    },
+    {
+      "epoch": 0.02,
+      "grad_norm": 1.0644451379776,
+      "learning_rate": 1.7474250108400467e-05,
+      "loss": 0.0942,
+      "step": 5
+    },
+    {
+      "epoch": 0.024,
+      "grad_norm": 0.6318012475967407,
+      "learning_rate": 1.945378125959109e-05,
+      "loss": 0.0534,
+      "step": 6
+    },
+    {
+      "epoch": 0.028,
+      "grad_norm": 0.4481346011161804,
+      "learning_rate": 2.1127451000356418e-05,
+      "loss": 0.0403,
+      "step": 7
+    },
+    {
+      "epoch": 0.032,
+      "grad_norm": 0.5680133104324341,
+      "learning_rate": 2.2577249674798584e-05,
+      "loss": 0.0308,
+      "step": 8
+    },
+    {
+      "epoch": 0.036,
+      "grad_norm": 0.43483495712280273,
+      "learning_rate": 2.385606273598312e-05,
+      "loss": 0.0265,
+      "step": 9
+    },
+    {
+      "epoch": 0.04,
+      "grad_norm": 0.4071185886859894,
+      "learning_rate": 2.4999999999999998e-05,
+      "loss": 0.0208,
+      "step": 10
+    },
+    {
+      "epoch": 0.044,
+      "grad_norm": 0.5254327654838562,
+      "learning_rate": 2.6034817128955623e-05,
+      "loss": 0.0342,
+      "step": 11
+    },
+    {
+      "epoch": 0.048,
+      "grad_norm": 0.4345680773258209,
+      "learning_rate": 2.6979531151190617e-05,
+      "loss": 0.02,
+      "step": 12
+    },
+    {
+      "epoch": 0.052,
+      "grad_norm": 0.5350413918495178,
+      "learning_rate": 2.7848583807670913e-05,
+      "loss": 0.0235,
+      "step": 13
+    },
+    {
+      "epoch": 0.056,
+      "grad_norm": 0.42923253774642944,
+      "learning_rate": 2.8653200891955945e-05,
+      "loss": 0.0187,
+      "step": 14
+    },
+    {
+      "epoch": 0.06,
+      "grad_norm": 0.45483624935150146,
+      "learning_rate": 2.940228147639203e-05,
+      "loss": 0.0243,
+      "step": 15
+    },
+    {
+      "epoch": 0.064,
+      "grad_norm": 0.5695978403091431,
+      "learning_rate": 3.0102999566398115e-05,
+      "loss": 0.0176,
+      "step": 16
+    },
+    {
+      "epoch": 0.068,
+      "grad_norm": 0.369960755109787,
+      "learning_rate": 3.076122303445685e-05,
+      "loss": 0.022,
+      "step": 17
+    },
+    {
+      "epoch": 0.072,
+      "grad_norm": 0.27982962131500244,
+      "learning_rate": 3.1381812627582646e-05,
+      "loss": 0.0135,
+      "step": 18
+    },
+    {
+      "epoch": 0.076,
+      "grad_norm": 0.4172382056713104,
+      "learning_rate": 3.1968840023820715e-05,
+      "loss": 0.0168,
+      "step": 19
+    },
+    {
+      "epoch": 0.08,
+      "grad_norm": 0.3223382532596588,
+      "learning_rate": 3.2525749891599525e-05,
+      "loss": 0.0098,
+      "step": 20
+    },
+    {
+      "epoch": 0.084,
+      "grad_norm": 0.3248063027858734,
+      "learning_rate": 3.305548236834798e-05,
+      "loss": 0.0089,
+      "step": 21
+    },
+    {
+      "epoch": 0.088,
+      "grad_norm": 0.2587601840496063,
+      "learning_rate": 3.3560567020555153e-05,
+      "loss": 0.0127,
+      "step": 22
+    },
+    {
+      "epoch": 0.092,
+      "grad_norm": 0.8799129128456116,
+      "learning_rate": 3.404319590043982e-05,
+      "loss": 0.0109,
+      "step": 23
+    },
+    {
+      "epoch": 0.096,
+      "grad_norm": 0.36437511444091797,
+      "learning_rate": 3.450528104279015e-05,
+      "loss": 0.0109,
+      "step": 24
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.27609172463417053,
+      "learning_rate": 3.4948500216800935e-05,
+      "loss": 0.006,
+      "step": 25
+    },
+    {
+      "epoch": 0.104,
+      "grad_norm": 0.2486644685268402,
+      "learning_rate": 3.537433369927044e-05,
+      "loss": 0.0072,
+      "step": 26
+    },
+    {
+      "epoch": 0.108,
+      "grad_norm": 0.4478244185447693,
+      "learning_rate": 3.578409410397468e-05,
+      "loss": 0.0096,
+      "step": 27
+    },
+    {
+      "epoch": 0.112,
+      "grad_norm": 0.23096731305122375,
+      "learning_rate": 3.6178950783555475e-05,
+      "loss": 0.0063,
+      "step": 28
+    },
+    {
+      "epoch": 0.116,
+      "grad_norm": 0.43742746114730835,
+      "learning_rate": 3.65599499474739e-05,
+      "loss": 0.0089,
+      "step": 29
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.23472078144550323,
+      "learning_rate": 3.6928031367991554e-05,
+      "loss": 0.0045,
+      "step": 30
+    },
+    {
+      "epoch": 0.124,
+      "grad_norm": 0.4394713044166565,
+      "learning_rate": 3.728404234585681e-05,
+      "loss": 0.009,
+      "step": 31
+    },
+    {
+      "epoch": 0.128,
+      "grad_norm": 0.3383356034755707,
+      "learning_rate": 3.762874945799765e-05,
+      "loss": 0.009,
+      "step": 32
+    },
+    {
+      "epoch": 0.132,
+      "grad_norm": 0.2296154648065567,
+      "learning_rate": 3.796284849694718e-05,
+      "loss": 0.0112,
+      "step": 33
+    },
+    {
+      "epoch": 0.136,
+      "grad_norm": 0.30197566747665405,
+      "learning_rate": 3.8286972926056376e-05,
+      "loss": 0.0084,
+      "step": 34
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.16092810034751892,
+      "learning_rate": 3.8601701108756885e-05,
+      "loss": 0.0032,
+      "step": 35
+    },
+    {
+      "epoch": 0.144,
+      "grad_norm": 0.3336141109466553,
+      "learning_rate": 3.890756251918218e-05,
+      "loss": 0.0038,
+      "step": 36
+    },
+    {
+      "epoch": 0.148,
+      "grad_norm": 0.15752413868904114,
+      "learning_rate": 3.920504310167487e-05,
+      "loss": 0.0033,
+      "step": 37
+    },
+    {
+      "epoch": 0.152,
+      "grad_norm": 0.2086404263973236,
+      "learning_rate": 3.949458991542025e-05,
+      "loss": 0.0028,
+      "step": 38
+    },
+    {
+      "epoch": 0.156,
+      "grad_norm": 0.2414998561143875,
+      "learning_rate": 3.977661517566247e-05,
+      "loss": 0.0067,
+      "step": 39
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.15021997690200806,
+      "learning_rate": 4.005149978319905e-05,
+      "loss": 0.0079,
+      "step": 40
+    },
+    {
+      "epoch": 0.164,
+      "grad_norm": 0.18097592890262604,
+      "learning_rate": 4.031959641799338e-05,
+      "loss": 0.0022,
+      "step": 41
+    },
+    {
+      "epoch": 0.168,
+      "grad_norm": 0.18316392600536346,
+      "learning_rate": 4.058123225994751e-05,
+      "loss": 0.0028,
+      "step": 42
+    },
+    {
+      "epoch": 0.172,
+      "grad_norm": 0.24015618860721588,
+      "learning_rate": 4.0836711389489654e-05,
+      "loss": 0.0037,
+      "step": 43
+    },
+    {
+      "epoch": 0.176,
+      "grad_norm": 0.15191113948822021,
+      "learning_rate": 4.108631691215468e-05,
+      "loss": 0.0028,
+      "step": 44
+    },
+    {
+      "epoch": 0.18,
+      "grad_norm": 0.18598900735378265,
+      "learning_rate": 4.133031284438358e-05,
+      "loss": 0.0017,
+      "step": 45
+    },
+    {
+      "epoch": 0.184,
+      "grad_norm": 0.2727285325527191,
+      "learning_rate": 4.156894579203935e-05,
+      "loss": 0.0061,
+      "step": 46
+    },
+    {
+      "epoch": 0.188,
+      "grad_norm": 0.19456899166107178,
+      "learning_rate": 4.180244644839293e-05,
+      "loss": 0.0028,
+      "step": 47
+    },
+    {
+      "epoch": 0.192,
+      "grad_norm": 0.20612072944641113,
+      "learning_rate": 4.203103093438968e-05,
+      "loss": 0.0028,
+      "step": 48
+    },
+    {
+      "epoch": 0.196,
+      "grad_norm": 0.498530775308609,
+      "learning_rate": 4.2254902000712836e-05,
+      "loss": 0.0073,
+      "step": 49
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.13147921860218048,
+      "learning_rate": 4.247425010840046e-05,
+      "loss": 0.005,
+      "step": 50
+    },
+    {
+      "epoch": 0.204,
+      "grad_norm": 0.8498079180717468,
+      "learning_rate": 4.2689254402448405e-05,
+      "loss": 0.0045,
+      "step": 51
+    },
+    {
+      "epoch": 0.208,
+      "grad_norm": 0.5296034216880798,
+      "learning_rate": 4.290008359086998e-05,
+      "loss": 0.0282,
+      "step": 52
+    },
+    {
+      "epoch": 0.212,
+      "grad_norm": 0.2918979823589325,
+      "learning_rate": 4.310689674001973e-05,
+      "loss": 0.009,
+      "step": 53
+    },
+    {
+      "epoch": 0.216,
+      "grad_norm": 0.555036723613739,
+      "learning_rate": 4.330984399557421e-05,
+      "loss": 0.0065,
+      "step": 54
+    },
+    {
+      "epoch": 0.22,
+      "grad_norm": 0.12301270663738251,
+      "learning_rate": 4.350906723735609e-05,
+      "loss": 0.0064,
+      "step": 55
+    },
+    {
+      "epoch": 0.224,
+      "grad_norm": 0.08401217311620712,
+      "learning_rate": 4.370470067515501e-05,
+      "loss": 0.002,
+      "step": 56
+    },
+    {
+      "epoch": 0.228,
+      "grad_norm": 0.4035862982273102,
+      "learning_rate": 4.3896871391812285e-05,
+      "loss": 0.0067,
+      "step": 57
+    },
+    {
+      "epoch": 0.232,
+      "grad_norm": 0.1083892211318016,
+      "learning_rate": 4.408569983907343e-05,
+      "loss": 0.0069,
+      "step": 58
+    },
+    {
+      "epoch": 0.236,
+      "grad_norm": 0.10449299961328506,
+      "learning_rate": 4.42713002910536e-05,
+      "loss": 0.0016,
+      "step": 59
+    },
+    {
+      "epoch": 0.24,
+      "grad_norm": 0.20857375860214233,
+      "learning_rate": 4.445378125959108e-05,
+      "loss": 0.0018,
+      "step": 60
+    },
+    {
+      "epoch": 0.244,
+      "grad_norm": 0.13858433067798615,
+      "learning_rate": 4.463324587526917e-05,
+      "loss": 0.0036,
+      "step": 61
+    },
+    {
+      "epoch": 0.248,
+      "grad_norm": 0.20314235985279083,
+      "learning_rate": 4.4809792237456346e-05,
+      "loss": 0.0028,
+      "step": 62
+    },
+    {
+      "epoch": 0.252,
+      "grad_norm": 0.15137360990047455,
+      "learning_rate": 4.498351373633954e-05,
+      "loss": 0.0026,
+      "step": 63
+    },
+    {
+      "epoch": 0.256,
+      "grad_norm": 0.08835887908935547,
+      "learning_rate": 4.515449934959717e-05,
+      "loss": 0.0012,
+      "step": 64
+    },
+    {
+      "epoch": 0.26,
+      "grad_norm": 0.07209821045398712,
+      "learning_rate": 4.532283391607138e-05,
+      "loss": 0.001,
+      "step": 65
+    },
+    {
+      "epoch": 0.264,
+      "grad_norm": 0.14767873287200928,
+      "learning_rate": 4.548859838854671e-05,
+      "loss": 0.0015,
+      "step": 66
+    },
+    {
+      "epoch": 0.268,
+      "grad_norm": 0.11616674065589905,
+      "learning_rate": 4.565187006752065e-05,
+      "loss": 0.001,
+      "step": 67
+    },
+    {
+      "epoch": 0.272,
+      "grad_norm": 0.08439616113901138,
+      "learning_rate": 4.581272281765591e-05,
+      "loss": 0.0046,
+      "step": 68
+    },
+    {
+      "epoch": 0.276,
+      "grad_norm": 0.07998619973659515,
+      "learning_rate": 4.597122726843138e-05,
+      "loss": 0.0045,
+      "step": 69
+    },
+    {
+      "epoch": 0.28,
+      "grad_norm": 0.05273636057972908,
+      "learning_rate": 4.612745100035642e-05,
+      "loss": 0.0005,
+      "step": 70
+    },
+    {
+      "epoch": 0.284,
+      "grad_norm": 0.07583728432655334,
+      "learning_rate": 4.628145871797688e-05,
+      "loss": 0.0018,
+      "step": 71
+    },
+    {
+      "epoch": 0.288,
+      "grad_norm": 0.12443387508392334,
+      "learning_rate": 4.643331241078171e-05,
+      "loss": 0.0007,
+      "step": 72
+    },
+    {
+      "epoch": 0.292,
+      "grad_norm": 0.09343032538890839,
+      "learning_rate": 4.658307150301139e-05,
+      "loss": 0.0006,
+      "step": 73
+    },
+    {
+      "epoch": 0.296,
+      "grad_norm": 0.08595340698957443,
+      "learning_rate": 4.67307929932744e-05,
+      "loss": 0.0037,
+      "step": 74
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 0.053027234971523285,
+      "learning_rate": 4.687653158479249e-05,
+      "loss": 0.0003,
+      "step": 75
+    },
+    {
+      "epoch": 0.304,
+      "grad_norm": 0.05994114652276039,
+      "learning_rate": 4.702033980701978e-05,
+      "loss": 0.0007,
+      "step": 76
+    },
+    {
+      "epoch": 0.308,
+      "grad_norm": 0.08113588392734528,
+      "learning_rate": 4.716226812931204e-05,
+      "loss": 0.0038,
+      "step": 77
+    },
+    {
+      "epoch": 0.312,
+      "grad_norm": 0.048519495874643326,
+      "learning_rate": 4.7302365067262006e-05,
+      "loss": 0.0022,
+      "step": 78
+    },
+    {
+      "epoch": 0.316,
+      "grad_norm": 0.06878262013196945,
+      "learning_rate": 4.744067728226103e-05,
+      "loss": 0.0026,
+      "step": 79
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 0.0174928717315197,
+      "learning_rate": 4.757724967479858e-05,
+      "loss": 0.0001,
+      "step": 80
+    },
+    {
+      "epoch": 0.324,
+      "grad_norm": 0.055549003183841705,
+      "learning_rate": 4.771212547196624e-05,
+      "loss": 0.0005,
+      "step": 81
+    },
+    {
+      "epoch": 0.328,
+      "grad_norm": 0.06923408806324005,
+      "learning_rate": 4.7845346309592914e-05,
+      "loss": 0.0002,
+      "step": 82
+    },
+    {
+      "epoch": 0.332,
+      "grad_norm": 0.05054256692528725,
+      "learning_rate": 4.7976952309401844e-05,
+      "loss": 0.0042,
+      "step": 83
+    },
+    {
+      "epoch": 0.336,
+      "grad_norm": 0.042055755853652954,
+      "learning_rate": 4.810698215154703e-05,
+      "loss": 0.0022,
+      "step": 84
+    },
+    {
+      "epoch": 0.34,
+      "grad_norm": 0.1804121732711792,
+      "learning_rate": 4.823547314285732e-05,
+      "loss": 0.002,
+      "step": 85
+    },
+    {
+      "epoch": 0.344,
+      "grad_norm": 0.024929817765951157,
+      "learning_rate": 4.836246128108918e-05,
+      "loss": 0.0001,
+      "step": 86
+    },
+    {
+      "epoch": 0.348,
+      "grad_norm": 0.708000123500824,
+      "learning_rate": 4.8487981315465456e-05,
+      "loss": 0.0026,
+      "step": 87
+    },
+    {
+      "epoch": 0.352,
+      "grad_norm": 0.6377427577972412,
+      "learning_rate": 4.8612066803754214e-05,
+      "loss": 0.0123,
+      "step": 88
+    },
+    {
+      "epoch": 0.356,
+      "grad_norm": 0.25735020637512207,
+      "learning_rate": 4.873475016612281e-05,
+      "loss": 0.0055,
+      "step": 89
+    },
+    {
+      "epoch": 0.36,
+      "grad_norm": 1.950273871421814,
+      "learning_rate": 4.885606273598312e-05,
+      "loss": 0.0408,
+      "step": 90
+    },
+    {
+      "epoch": 0.364,
+      "grad_norm": 0.34996047616004944,
+      "learning_rate": 4.897603480802733e-05,
+      "loss": 0.0034,
+      "step": 91
+    },
+    {
+      "epoch": 0.368,
+      "grad_norm": 0.9180243611335754,
+      "learning_rate": 4.909469568363888e-05,
+      "loss": 0.02,
+      "step": 92
+    },
+    {
+      "epoch": 0.372,
+      "grad_norm": 0.8092993497848511,
+      "learning_rate": 4.9212073713848375e-05,
+      "loss": 0.0075,
+      "step": 93
+    },
+    {
+      "epoch": 0.376,
+      "grad_norm": 0.6170804500579834,
+      "learning_rate": 4.932819633999246e-05,
+      "loss": 0.0113,
+      "step": 94
+    },
+    {
+      "epoch": 0.38,
+      "grad_norm": 4.7827534675598145,
+      "learning_rate": 4.9443090132221186e-05,
+      "loss": 0.03,
+      "step": 95
+    },
+    {
+      "epoch": 0.384,
+      "grad_norm": 3.489516258239746,
+      "learning_rate": 4.9556780825989205e-05,
+      "loss": 0.0382,
+      "step": 96
+    },
+    {
+      "epoch": 0.388,
+      "grad_norm": 2.35620379447937,
+      "learning_rate": 4.9669293356656114e-05,
+      "loss": 0.0299,
+      "step": 97
+    },
+    {
+      "epoch": 0.392,
+      "grad_norm": 0.8634485602378845,
+      "learning_rate": 4.978065189231237e-05,
+      "loss": 0.0167,
+      "step": 98
+    },
+    {
+      "epoch": 0.396,
+      "grad_norm": 0.2725628614425659,
+      "learning_rate": 4.989087986493874e-05,
+      "loss": 0.0081,
+      "step": 99
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.17562685906887054,
+      "learning_rate": 4.9999999999999996e-05,
+      "loss": 0.0043,
+      "step": 100
+    },
+    {
+      "epoch": 0.404,
+      "grad_norm": 0.8643080592155457,
+      "learning_rate": 5e-05,
+      "loss": 0.0125,
+      "step": 101
+    },
+    {
+      "epoch": 0.408,
+      "grad_norm": 0.11818674951791763,
+      "learning_rate": 4.9875000000000006e-05,
+      "loss": 0.0034,
+      "step": 102
+    },
+    {
+      "epoch": 0.412,
+      "grad_norm": 0.20233666896820068,
+      "learning_rate": 4.975e-05,
+      "loss": 0.0029,
+      "step": 103
+    },
+    {
+      "epoch": 0.416,
+      "grad_norm": 0.23807714879512787,
+      "learning_rate": 4.962500000000001e-05,
+      "loss": 0.0026,
+      "step": 104
+    },
+    {
+      "epoch": 0.42,
+      "grad_norm": 0.1580943763256073,
+      "learning_rate": 4.9500000000000004e-05,
+      "loss": 0.0015,
+      "step": 105
+    },
+    {
+      "epoch": 0.424,
+      "grad_norm": 0.19758278131484985,
+      "learning_rate": 4.937500000000001e-05,
+      "loss": 0.0025,
+      "step": 106
+    },
+    {
+      "epoch": 0.428,
+      "grad_norm": 0.2816386818885803,
+      "learning_rate": 4.9250000000000004e-05,
+      "loss": 0.0028,
+      "step": 107
+    },
+    {
+      "epoch": 0.432,
+      "grad_norm": 0.3274244964122772,
+      "learning_rate": 4.9125e-05,
+      "loss": 0.0027,
+      "step": 108
+    },
+    {
+      "epoch": 0.436,
+      "grad_norm": 0.2742844820022583,
+      "learning_rate": 4.9e-05,
+      "loss": 0.0038,
+      "step": 109
+    },
+    {
+      "epoch": 0.44,
+      "grad_norm": 0.2021024376153946,
+      "learning_rate": 4.8875e-05,
+      "loss": 0.007,
+      "step": 110
+    },
+    {
+      "epoch": 0.444,
+      "grad_norm": 0.1884881854057312,
+      "learning_rate": 4.875e-05,
+      "loss": 0.0013,
+      "step": 111
+    },
+    {
+      "epoch": 0.448,
+      "grad_norm": 0.5865951776504517,
+      "learning_rate": 4.8625e-05,
+      "loss": 0.0053,
+      "step": 112
+    },
+    {
+      "epoch": 0.452,
+      "grad_norm": 0.23957838118076324,
+      "learning_rate": 4.85e-05,
+      "loss": 0.005,
+      "step": 113
+    },
+    {
+      "epoch": 0.456,
+      "grad_norm": 0.3090040981769562,
+      "learning_rate": 4.8375000000000004e-05,
+      "loss": 0.0082,
+      "step": 114
+    },
+    {
+      "epoch": 0.46,
+      "grad_norm": 0.37422287464141846,
+      "learning_rate": 4.825e-05,
+      "loss": 0.0149,
+      "step": 115
+    },
+    {
+      "epoch": 0.464,
+      "grad_norm": 0.1816074252128601,
+      "learning_rate": 4.8125000000000004e-05,
+      "loss": 0.0053,
+      "step": 116
+    },
+    {
+      "epoch": 0.468,
+      "grad_norm": 0.38292816281318665,
+      "learning_rate": 4.8e-05,
+      "loss": 0.0057,
+      "step": 117
+    },
+    {
+      "epoch": 0.472,
+      "grad_norm": 0.10594677180051804,
+      "learning_rate": 4.7875000000000005e-05,
+      "loss": 0.0021,
+      "step": 118
+    },
+    {
+      "epoch": 0.476,
+      "grad_norm": 0.13979272544384003,
+      "learning_rate": 4.775e-05,
+      "loss": 0.0071,
+      "step": 119
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 0.08800072968006134,
+      "learning_rate": 4.7625000000000006e-05,
+      "loss": 0.0039,
+      "step": 120
+    },
+    {
+      "epoch": 0.484,
+      "grad_norm": 0.190990149974823,
+      "learning_rate": 4.75e-05,
+      "loss": 0.0037,
+      "step": 121
+    },
+    {
+      "epoch": 0.488,
+      "grad_norm": 0.13872408866882324,
+      "learning_rate": 4.7375e-05,
+      "loss": 0.0063,
+      "step": 122
+    },
+    {
+      "epoch": 0.492,
+      "grad_norm": 0.15142665803432465,
+      "learning_rate": 4.7249999999999997e-05,
+      "loss": 0.0009,
+      "step": 123
+    },
+    {
+      "epoch": 0.496,
+      "grad_norm": 0.15705378353595734,
+      "learning_rate": 4.7125e-05,
+      "loss": 0.002,
+      "step": 124
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.09296617656946182,
+      "learning_rate": 4.7e-05,
+      "loss": 0.0027,
+      "step": 125
+    },
+    {
+      "epoch": 0.504,
+      "grad_norm": 0.20458370447158813,
+      "learning_rate": 4.6875e-05,
+      "loss": 0.002,
+      "step": 126
+    },
+    {
+      "epoch": 0.508,
+      "grad_norm": 0.13503938913345337,
+      "learning_rate": 4.6750000000000005e-05,
+      "loss": 0.0018,
+      "step": 127
+    },
+    {
+      "epoch": 0.512,
+      "grad_norm": 0.10841026157140732,
+      "learning_rate": 4.6625e-05,
+      "loss": 0.0013,
+      "step": 128
+    },
+    {
+      "epoch": 0.516,
+      "grad_norm": 0.07126464694738388,
+      "learning_rate": 4.6500000000000005e-05,
+      "loss": 0.0026,
+      "step": 129
+    },
+    {
+      "epoch": 0.52,
+      "grad_norm": 0.1297496259212494,
+      "learning_rate": 4.6375e-05,
+      "loss": 0.0014,
+      "step": 130
+    },
+    {
+      "epoch": 0.524,
+      "grad_norm": 0.11263352632522583,
+      "learning_rate": 4.6250000000000006e-05,
+      "loss": 0.0028,
+      "step": 131
+    },
+    {
+      "epoch": 0.528,
+      "grad_norm": 0.10150951147079468,
+      "learning_rate": 4.6125e-05,
+      "loss": 0.0015,
+      "step": 132
+    },
+    {
+      "epoch": 0.532,
+      "grad_norm": 0.18945740163326263,
+      "learning_rate": 4.600000000000001e-05,
+      "loss": 0.0021,
+      "step": 133
+    },
+    {
+      "epoch": 0.536,
+      "grad_norm": 0.17706206440925598,
+      "learning_rate": 4.5875000000000004e-05,
+      "loss": 0.0019,
+      "step": 134
+    },
+    {
+      "epoch": 0.54,
+      "grad_norm": 0.05546269938349724,
+      "learning_rate": 4.575e-05,
+      "loss": 0.0005,
+      "step": 135
+    },
+    {
+      "epoch": 0.544,
+      "grad_norm": 0.12179642915725708,
+      "learning_rate": 4.5625e-05,
+      "loss": 0.0012,
+      "step": 136
+    },
+    {
+      "epoch": 0.548,
+      "grad_norm": 0.05075937882065773,
+      "learning_rate": 4.55e-05,
+      "loss": 0.0003,
+      "step": 137
+    },
+    {
+      "epoch": 0.552,
+      "grad_norm": 0.07799898087978363,
+      "learning_rate": 4.5375e-05,
+      "loss": 0.0016,
+      "step": 138
+    },
+    {
+      "epoch": 0.556,
+      "grad_norm": 0.13191546499729156,
+      "learning_rate": 4.525e-05,
+      "loss": 0.0033,
+      "step": 139
+    },
+    {
+      "epoch": 0.56,
+      "grad_norm": 0.14412976801395416,
+      "learning_rate": 4.5125e-05,
+      "loss": 0.003,
+      "step": 140
+    },
+    {
+      "epoch": 0.564,
+      "grad_norm": 0.09449886530637741,
+      "learning_rate": 4.5e-05,
+      "loss": 0.0013,
+      "step": 141
+    },
+    {
+      "epoch": 0.568,
+      "grad_norm": 0.03718528896570206,
+      "learning_rate": 4.4875e-05,
+      "loss": 0.0003,
+      "step": 142
+    },
+    {
+      "epoch": 0.572,
+      "grad_norm": 0.05031194910407066,
+      "learning_rate": 4.4750000000000004e-05,
+      "loss": 0.0004,
+      "step": 143
+    },
+    {
+      "epoch": 0.576,
+      "grad_norm": 0.24351955950260162,
+      "learning_rate": 4.4625e-05,
+      "loss": 0.0022,
+      "step": 144
+    },
+    {
+      "epoch": 0.58,
+      "grad_norm": 0.10770543664693832,
+      "learning_rate": 4.4500000000000004e-05,
+      "loss": 0.003,
+      "step": 145
+    },
+    {
+      "epoch": 0.584,
+      "grad_norm": 0.0956270769238472,
+      "learning_rate": 4.4375e-05,
+      "loss": 0.0075,
+      "step": 146
+    },
+    {
+      "epoch": 0.588,
+      "grad_norm": 0.28642475605010986,
+      "learning_rate": 4.4250000000000005e-05,
+      "loss": 0.0019,
+      "step": 147
+    },
+    {
+      "epoch": 0.592,
+      "grad_norm": 0.1973092406988144,
+      "learning_rate": 4.4125e-05,
+      "loss": 0.0006,
+      "step": 148
+    },
+    {
+      "epoch": 0.596,
+      "grad_norm": 0.48930782079696655,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.0018,
+      "step": 149
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.08301931619644165,
+      "learning_rate": 4.3875e-05,
+      "loss": 0.0031,
+      "step": 150
+    },
+    {
+      "epoch": 0.604,
+      "grad_norm": 0.09609490633010864,
+      "learning_rate": 4.375e-05,
+      "loss": 0.0007,
+      "step": 151
+    },
+    {
+      "epoch": 0.608,
+      "grad_norm": 0.07038693130016327,
+      "learning_rate": 4.3625e-05,
+      "loss": 0.0006,
+      "step": 152
+    },
+    {
+      "epoch": 0.612,
+      "grad_norm": 0.034853026270866394,
+      "learning_rate": 4.35e-05,
+      "loss": 0.0003,
+      "step": 153
+    },
+    {
+      "epoch": 0.616,
+      "grad_norm": 0.1049143597483635,
+      "learning_rate": 4.3375000000000004e-05,
+      "loss": 0.0032,
+      "step": 154
+    },
+    {
+      "epoch": 0.62,
+      "grad_norm": 0.1149531900882721,
+      "learning_rate": 4.325e-05,
+      "loss": 0.002,
+      "step": 155
+    },
+    {
+      "epoch": 0.624,
+      "grad_norm": 0.13460274040699005,
+      "learning_rate": 4.3125000000000005e-05,
+      "loss": 0.0012,
+      "step": 156
+    },
+    {
+      "epoch": 0.628,
+      "grad_norm": 0.16842494904994965,
+      "learning_rate": 4.3e-05,
+      "loss": 0.0019,
+      "step": 157
+    },
+    {
+      "epoch": 0.632,
+      "grad_norm": 0.09362483024597168,
+      "learning_rate": 4.2875000000000005e-05,
+      "loss": 0.0038,
+      "step": 158
+    },
+    {
+      "epoch": 0.636,
+      "grad_norm": 0.23620232939720154,
+      "learning_rate": 4.275e-05,
+      "loss": 0.0021,
+      "step": 159
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 0.04187873750925064,
+      "learning_rate": 4.2625000000000006e-05,
+      "loss": 0.0003,
+      "step": 160
+    },
+    {
+      "epoch": 0.644,
+      "grad_norm": 0.15332873165607452,
+      "learning_rate": 4.25e-05,
+      "loss": 0.0034,
+      "step": 161
+    },
+    {
+      "epoch": 0.648,
+      "grad_norm": 0.052837781608104706,
+      "learning_rate": 4.237500000000001e-05,
+      "loss": 0.0047,
+      "step": 162
+    },
+    {
+      "epoch": 0.652,
+      "grad_norm": 0.08028219640254974,
+      "learning_rate": 4.2250000000000004e-05,
+      "loss": 0.0016,
+      "step": 163
+    },
+    {
+      "epoch": 0.656,
+      "grad_norm": 0.11588768661022186,
+      "learning_rate": 4.2125e-05,
+      "loss": 0.0034,
+      "step": 164
+    },
+    {
+      "epoch": 0.66,
+      "grad_norm": 0.055087096989154816,
+      "learning_rate": 4.2e-05,
+      "loss": 0.0009,
+      "step": 165
+    },
+    {
+      "epoch": 0.664,
+      "grad_norm": 0.241183340549469,
+      "learning_rate": 4.1875e-05,
+      "loss": 0.0024,
+      "step": 166
+    },
+    {
+      "epoch": 0.668,
+      "grad_norm": 0.15006832778453827,
+      "learning_rate": 4.175e-05,
+      "loss": 0.0018,
+      "step": 167
+    },
+    {
+      "epoch": 0.672,
+      "grad_norm": 0.46611738204956055,
+      "learning_rate": 4.1625e-05,
+      "loss": 0.0037,
+      "step": 168
+    },
+    {
+      "epoch": 0.676,
+      "grad_norm": 0.19828003644943237,
+      "learning_rate": 4.15e-05,
+      "loss": 0.0018,
+      "step": 169
+    },
+    {
+      "epoch": 0.68,
+      "grad_norm": 0.12308524549007416,
+      "learning_rate": 4.1375e-05,
+      "loss": 0.001,
+      "step": 170
+    },
+    {
+      "epoch": 0.684,
+      "grad_norm": 0.09313371032476425,
+      "learning_rate": 4.125e-05,
+      "loss": 0.0009,
+      "step": 171
+    },
+    {
+      "epoch": 0.688,
+      "grad_norm": 0.4643174111843109,
+      "learning_rate": 4.1125000000000004e-05,
+      "loss": 0.0036,
+      "step": 172
+    },
+    {
+      "epoch": 0.692,
+      "grad_norm": 0.12040840089321136,
+      "learning_rate": 4.1e-05,
+      "loss": 0.0024,
+      "step": 173
+    },
+    {
+      "epoch": 0.696,
+      "grad_norm": 0.22784455120563507,
+      "learning_rate": 4.0875000000000004e-05,
+      "loss": 0.0027,
+      "step": 174
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.1869855672121048,
+      "learning_rate": 4.075e-05,
+      "loss": 0.0043,
+      "step": 175
+    },
+    {
+      "epoch": 0.704,
+      "grad_norm": 0.11409322917461395,
+      "learning_rate": 4.0625000000000005e-05,
+      "loss": 0.0026,
+      "step": 176
+    },
+    {
+      "epoch": 0.708,
+      "grad_norm": 0.09818801283836365,
+      "learning_rate": 4.05e-05,
+      "loss": 0.0019,
+      "step": 177
+    },
+    {
+      "epoch": 0.712,
+      "grad_norm": 0.0799233466386795,
+      "learning_rate": 4.0375e-05,
+      "loss": 0.0008,
+      "step": 178
+    },
+    {
+      "epoch": 0.716,
+      "grad_norm": 0.22596658766269684,
+      "learning_rate": 4.025e-05,
+      "loss": 0.0047,
+      "step": 179
+    },
+    {
+      "epoch": 0.72,
+      "grad_norm": 0.07836493104696274,
+      "learning_rate": 4.0125e-05,
+      "loss": 0.0029,
+      "step": 180
+    },
+    {
+      "epoch": 0.724,
+      "grad_norm": 0.1314069926738739,
+      "learning_rate": 4e-05,
+      "loss": 0.0013,
+      "step": 181
+    },
+    {
+      "epoch": 0.728,
+      "grad_norm": 0.12211080640554428,
+      "learning_rate": 3.9875e-05,
+      "loss": 0.004,
+      "step": 182
+    },
+    {
+      "epoch": 0.732,
+      "grad_norm": 0.5613779425621033,
+      "learning_rate": 3.9750000000000004e-05,
+      "loss": 0.0065,
+      "step": 183
+    },
+    {
+      "epoch": 0.736,
+      "grad_norm": 0.03666184842586517,
+      "learning_rate": 3.9625e-05,
+      "loss": 0.0003,
+      "step": 184
+    },
+    {
+      "epoch": 0.74,
+      "grad_norm": 0.06119844689965248,
+      "learning_rate": 3.9500000000000005e-05,
+      "loss": 0.0009,
+      "step": 185
+    },
+    {
+      "epoch": 0.744,
+      "grad_norm": 0.08475872874259949,
+      "learning_rate": 3.9375e-05,
+      "loss": 0.0009,
+      "step": 186
+    },
+    {
+      "epoch": 0.748,
+      "grad_norm": 0.16239683330059052,
+      "learning_rate": 3.9250000000000005e-05,
+      "loss": 0.0012,
+      "step": 187
+    },
+    {
+      "epoch": 0.752,
+      "grad_norm": 0.07799831032752991,
+      "learning_rate": 3.9125e-05,
+      "loss": 0.0055,
+      "step": 188
+    },
+    {
+      "epoch": 0.756,
+      "grad_norm": 0.04904457554221153,
+      "learning_rate": 3.9000000000000006e-05,
+      "loss": 0.0024,
+      "step": 189
+    },
+    {
+      "epoch": 0.76,
+      "grad_norm": 0.04471833258867264,
+      "learning_rate": 3.8875e-05,
+      "loss": 0.0002,
+      "step": 190
+    },
+    {
+      "epoch": 0.764,
+      "grad_norm": 0.24567581713199615,
+      "learning_rate": 3.875e-05,
+      "loss": 0.0026,
+      "step": 191
+    },
+    {
+      "epoch": 0.768,
+      "grad_norm": 0.043222181499004364,
+      "learning_rate": 3.8625e-05,
+      "loss": 0.0002,
+      "step": 192
+    },
+    {
+      "epoch": 0.772,
+      "grad_norm": 0.043726395815610886,
+      "learning_rate": 3.85e-05,
+      "loss": 0.0001,
+      "step": 193
+    },
+    {
+      "epoch": 0.776,
+      "grad_norm": 0.11048438400030136,
+      "learning_rate": 3.8375e-05,
+      "loss": 0.0007,
+      "step": 194
+    },
+    {
+      "epoch": 0.78,
+      "grad_norm": 0.2425050139427185,
+      "learning_rate": 3.825e-05,
+      "loss": 0.0055,
+      "step": 195
+    },
+    {
+      "epoch": 0.784,
+      "grad_norm": 0.09772332012653351,
+      "learning_rate": 3.8125e-05,
+      "loss": 0.0004,
+      "step": 196
+    },
+    {
+      "epoch": 0.788,
+      "grad_norm": 0.024435682222247124,
+      "learning_rate": 3.8e-05,
+      "loss": 0.0005,
+      "step": 197
+    },
+    {
+      "epoch": 0.792,
+      "grad_norm": 0.04484609141945839,
+      "learning_rate": 3.7875e-05,
+      "loss": 0.0003,
+      "step": 198
+    },
+    {
+      "epoch": 0.796,
+      "grad_norm": 0.035425156354904175,
+      "learning_rate": 3.775e-05,
+      "loss": 0.0003,
+      "step": 199
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.04564299434423447,
+      "learning_rate": 3.7625e-05,
+      "loss": 0.0019,
+      "step": 200
+    },
+    {
+      "epoch": 0.804,
+      "grad_norm": 0.04999084025621414,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 0.0012,
+      "step": 201
+    },
+    {
+      "epoch": 0.808,
+      "grad_norm": 0.050590988248586655,
+      "learning_rate": 3.737500000000001e-05,
+      "loss": 0.0037,
+      "step": 202
+    },
+    {
+      "epoch": 0.812,
+      "grad_norm": 0.03431128337979317,
+      "learning_rate": 3.7250000000000004e-05,
+      "loss": 0.0025,
+      "step": 203
+    },
+    {
+      "epoch": 0.816,
+      "grad_norm": 0.06956495344638824,
+      "learning_rate": 3.7125e-05,
+      "loss": 0.0032,
+      "step": 204
+    },
+    {
+      "epoch": 0.82,
+      "grad_norm": 0.027486376464366913,
+      "learning_rate": 3.7e-05,
+      "loss": 0.0005,
+      "step": 205
+    },
+    {
+      "epoch": 0.824,
+      "grad_norm": 0.027509450912475586,
+      "learning_rate": 3.6875e-05,
+      "loss": 0.0003,
+      "step": 206
+    },
+    {
+      "epoch": 0.828,
+      "grad_norm": 0.04087129607796669,
+      "learning_rate": 3.675e-05,
+      "loss": 0.0003,
+      "step": 207
+    },
+    {
+      "epoch": 0.832,
+      "grad_norm": 0.03978107497096062,
+      "learning_rate": 3.6625e-05,
+      "loss": 0.0023,
+      "step": 208
+    },
+    {
+      "epoch": 0.836,
+      "grad_norm": 0.04381946101784706,
+      "learning_rate": 3.65e-05,
+      "loss": 0.0032,
+      "step": 209
+    },
+    {
+      "epoch": 0.84,
+      "grad_norm": 0.018398966640233994,
+      "learning_rate": 3.6375e-05,
+      "loss": 0.0002,
+      "step": 210
+    },
+    {
+      "epoch": 0.844,
+      "grad_norm": 0.27152058482170105,
+      "learning_rate": 3.625e-05,
+      "loss": 0.001,
+      "step": 211
+    },
+    {
+      "epoch": 0.848,
+      "grad_norm": 0.02850174717605114,
+      "learning_rate": 3.6125000000000004e-05,
+      "loss": 0.0007,
+      "step": 212
+    },
+    {
+      "epoch": 0.852,
+      "grad_norm": 0.06494108587503433,
+      "learning_rate": 3.6e-05,
+      "loss": 0.0014,
+      "step": 213
+    },
+    {
+      "epoch": 0.856,
+      "grad_norm": 0.042822666466236115,
+      "learning_rate": 3.5875000000000005e-05,
+      "loss": 0.0008,
+      "step": 214
+    },
+    {
+      "epoch": 0.86,
+      "grad_norm": 0.04977138340473175,
+      "learning_rate": 3.575e-05,
+      "loss": 0.0005,
+      "step": 215
+    },
+    {
+      "epoch": 0.864,
+      "grad_norm": 0.06221337243914604,
+      "learning_rate": 3.5625000000000005e-05,
+      "loss": 0.0016,
+      "step": 216
+    },
+    {
+      "epoch": 0.868,
+      "grad_norm": 0.011892068199813366,
+      "learning_rate": 3.55e-05,
+      "loss": 0.0001,
+      "step": 217
+    },
+    {
+      "epoch": 0.872,
+      "grad_norm": 0.0454794205725193,
+      "learning_rate": 3.5375e-05,
+      "loss": 0.0008,
+      "step": 218
+    },
+    {
+      "epoch": 0.876,
+      "grad_norm": 0.017043855041265488,
+      "learning_rate": 3.525e-05,
+      "loss": 0.0003,
+      "step": 219
+    },
+    {
+      "epoch": 0.88,
+      "grad_norm": 0.017138540744781494,
+      "learning_rate": 3.5125e-05,
+      "loss": 0.0002,
+      "step": 220
+    },
+    {
+      "epoch": 0.884,
+      "grad_norm": 0.10911066085100174,
+      "learning_rate": 3.5e-05,
+      "loss": 0.003,
+      "step": 221
+    },
+    {
+      "epoch": 0.888,
+      "grad_norm": 0.07072826474905014,
+      "learning_rate": 3.4875e-05,
+      "loss": 0.0028,
+      "step": 222
+    },
+    {
+      "epoch": 0.892,
+      "grad_norm": 0.06483879685401917,
+      "learning_rate": 3.475e-05,
+      "loss": 0.0011,
+      "step": 223
+    },
+    {
+      "epoch": 0.896,
+      "grad_norm": 0.032061319798231125,
+      "learning_rate": 3.4625e-05,
+      "loss": 0.0003,
+      "step": 224
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.10455196350812912,
+      "learning_rate": 3.45e-05,
+      "loss": 0.0006,
+      "step": 225
+    },
+    {
+      "epoch": 0.904,
+      "grad_norm": 0.03009301982820034,
+      "learning_rate": 3.4375e-05,
+      "loss": 0.0009,
+      "step": 226
+    },
+    {
+      "epoch": 0.908,
+      "grad_norm": 0.21497099101543427,
+      "learning_rate": 3.4250000000000006e-05,
+      "loss": 0.0022,
+      "step": 227
+    },
+    {
+      "epoch": 0.912,
+      "grad_norm": 0.05242502689361572,
+      "learning_rate": 3.4125e-05,
+      "loss": 0.0032,
+      "step": 228
+    },
+    {
+      "epoch": 0.916,
+      "grad_norm": 0.03429080918431282,
+      "learning_rate": 3.4000000000000007e-05,
+      "loss": 0.0015,
+      "step": 229
+    },
+    {
+      "epoch": 0.92,
+      "grad_norm": 0.3183232843875885,
+      "learning_rate": 3.3875000000000003e-05,
+      "loss": 0.0033,
+      "step": 230
+    },
+    {
+      "epoch": 0.924,
+      "grad_norm": 0.033974889665842056,
+      "learning_rate": 3.375000000000001e-05,
+      "loss": 0.0004,
+      "step": 231
+    },
+    {
+      "epoch": 0.928,
+      "grad_norm": 0.08924167603254318,
+      "learning_rate": 3.3625000000000004e-05,
+      "loss": 0.0014,
+      "step": 232
+    },
+    {
+      "epoch": 0.932,
+      "grad_norm": 0.093363918364048,
+      "learning_rate": 3.35e-05,
+      "loss": 0.0006,
+      "step": 233
+    },
+    {
+      "epoch": 0.936,
+      "grad_norm": 0.105079285800457,
+      "learning_rate": 3.3375e-05,
+      "loss": 0.0028,
+      "step": 234
+    },
+    {
+      "epoch": 0.94,
+      "grad_norm": 0.057525135576725006,
+      "learning_rate": 3.325e-05,
+      "loss": 0.0004,
+      "step": 235
+    },
+    {
+      "epoch": 0.944,
+      "grad_norm": 0.07350476831197739,
+      "learning_rate": 3.3125e-05,
+      "loss": 0.0024,
+      "step": 236
+    },
+    {
+      "epoch": 0.948,
+      "grad_norm": 0.020194334909319878,
+      "learning_rate": 3.3e-05,
+      "loss": 0.0002,
+      "step": 237
+    },
+    {
+      "epoch": 0.952,
+      "grad_norm": 0.010576704517006874,
+      "learning_rate": 3.2875e-05,
+      "loss": 0.0001,
+      "step": 238
+    },
+    {
+      "epoch": 0.956,
+      "grad_norm": 0.007914057932794094,
+      "learning_rate": 3.275e-05,
+      "loss": 0.0001,
+      "step": 239
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 0.06846226006746292,
+      "learning_rate": 3.2625e-05,
+      "loss": 0.0022,
+      "step": 240
+    },
+    {
+      "epoch": 0.964,
+      "grad_norm": 0.002107464475557208,
+      "learning_rate": 3.2500000000000004e-05,
+      "loss": 0.0,
+      "step": 241
+    },
+    {
+      "epoch": 0.968,
+      "grad_norm": 0.005104186944663525,
+      "learning_rate": 3.2375e-05,
+      "loss": 0.0001,
+      "step": 242
+    },
+    {
+      "epoch": 0.972,
+      "grad_norm": 0.027005469426512718,
+      "learning_rate": 3.2250000000000005e-05,
+      "loss": 0.0012,
+      "step": 243
+    },
+    {
+      "epoch": 0.976,
+      "grad_norm": 0.01935047283768654,
+      "learning_rate": 3.2125e-05,
+      "loss": 0.0001,
+      "step": 244
+    },
+    {
+      "epoch": 0.98,
+      "grad_norm": 0.012384717352688313,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 0.0001,
+      "step": 245
+    },
+    {
+      "epoch": 0.984,
+      "grad_norm": 0.006367299240082502,
+      "learning_rate": 3.1875e-05,
+      "loss": 0.0001,
+      "step": 246
+    },
+    {
+      "epoch": 0.988,
+      "grad_norm": 0.003837389638647437,
+      "learning_rate": 3.175e-05,
+      "loss": 0.0,
+      "step": 247
+    },
+    {
+      "epoch": 0.992,
+      "grad_norm": 0.17976249754428864,
+      "learning_rate": 3.1624999999999996e-05,
+      "loss": 0.0003,
+      "step": 248
+    },
+    {
+      "epoch": 0.996,
+      "grad_norm": 0.0030153498519212008,
+      "learning_rate": 3.15e-05,
+      "loss": 0.0,
+      "step": 249
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.011528008617460728,
+      "learning_rate": 3.1375e-05,
+      "loss": 0.0,
+      "step": 250
+    },
+    {
+      "epoch": 1.004,
+      "grad_norm": 0.0014031054452061653,
+      "learning_rate": 3.125e-05,
+      "loss": 0.0,
+      "step": 251
+    },
+    {
+      "epoch": 1.008,
+      "grad_norm": 0.048355042934417725,
+      "learning_rate": 3.1125000000000004e-05,
+      "loss": 0.0039,
+      "step": 252
+    },
+    {
+      "epoch": 1.012,
+      "grad_norm": 0.3448329269886017,
+      "learning_rate": 3.1e-05,
+      "loss": 0.001,
+      "step": 253
+    },
+    {
+      "epoch": 1.016,
+      "grad_norm": 0.04559427127242088,
+      "learning_rate": 3.0875000000000005e-05,
+      "loss": 0.0001,
+      "step": 254
+    },
+    {
+      "epoch": 1.02,
+      "grad_norm": 0.059698574244976044,
+      "learning_rate": 3.075e-05,
+      "loss": 0.0024,
+      "step": 255
+    },
+    {
+      "epoch": 1.024,
+      "grad_norm": 0.08983492106199265,
+      "learning_rate": 3.0625000000000006e-05,
+      "loss": 0.0002,
+      "step": 256
+    },
+    {
+      "epoch": 1.028,
+      "grad_norm": 0.025842411443591118,
+      "learning_rate": 3.05e-05,
+      "loss": 0.0001,
+      "step": 257
+    },
+    {
+      "epoch": 1.032,
+      "grad_norm": 0.009762114845216274,
+      "learning_rate": 3.0375000000000003e-05,
+      "loss": 0.0001,
+      "step": 258
+    },
+    {
+      "epoch": 1.036,
+      "grad_norm": 0.03436556085944176,
+      "learning_rate": 3.025e-05,
+      "loss": 0.0013,
+      "step": 259
+    },
+    {
+      "epoch": 1.04,
+      "grad_norm": 0.13744613528251648,
+      "learning_rate": 3.0125000000000004e-05,
+      "loss": 0.0002,
+      "step": 260
+    },
+    {
+      "epoch": 1.044,
+      "grad_norm": 0.08697144687175751,
+      "learning_rate": 3e-05,
+      "loss": 0.0009,
+      "step": 261
+    },
+    {
+      "epoch": 1.048,
+      "grad_norm": 0.005860824603587389,
+      "learning_rate": 2.9875000000000004e-05,
+      "loss": 0.0,
+      "step": 262
+    },
+    {
+      "epoch": 1.052,
+      "grad_norm": 0.031256936490535736,
+      "learning_rate": 2.975e-05,
+      "loss": 0.002,
+      "step": 263
+    },
+    {
+      "epoch": 1.056,
+      "grad_norm": 0.007167185191065073,
+      "learning_rate": 2.9625000000000002e-05,
+      "loss": 0.0,
+      "step": 264
+    },
+    {
+      "epoch": 1.06,
+      "grad_norm": 0.026170356199145317,
+      "learning_rate": 2.95e-05,
+      "loss": 0.0015,
+      "step": 265
+    },
+    {
+      "epoch": 1.064,
+      "grad_norm": 0.000729930296074599,
+      "learning_rate": 2.9375000000000003e-05,
+      "loss": 0.0,
+      "step": 266
+    },
+    {
+      "epoch": 1.068,
+      "grad_norm": 0.08990947157144547,
+      "learning_rate": 2.925e-05,
+      "loss": 0.0007,
+      "step": 267
+    },
+    {
+      "epoch": 1.072,
+      "grad_norm": 0.0014114304212853312,
+      "learning_rate": 2.9125000000000003e-05,
+      "loss": 0.0,
+      "step": 268
+    },
+    {
+      "epoch": 1.076,
+      "grad_norm": 0.0023166867904365063,
+      "learning_rate": 2.9e-05,
+      "loss": 0.0,
+      "step": 269
+    },
+    {
+      "epoch": 1.08,
+      "grad_norm": 0.027701951563358307,
+      "learning_rate": 2.8875e-05,
+      "loss": 0.0016,
+      "step": 270
+    },
+    {
+      "epoch": 1.084,
+      "grad_norm": 0.012332534417510033,
+      "learning_rate": 2.8749999999999997e-05,
+      "loss": 0.0001,
+      "step": 271
+    },
+    {
+      "epoch": 1.088,
+      "grad_norm": 0.1477350890636444,
+      "learning_rate": 2.8625e-05,
+      "loss": 0.0011,
+      "step": 272
+    },
+    {
+      "epoch": 1.092,
+      "grad_norm": 0.04740828648209572,
+      "learning_rate": 2.8499999999999998e-05,
+      "loss": 0.0028,
+      "step": 273
+    },
+    {
+      "epoch": 1.096,
+      "grad_norm": 0.06496315449476242,
+      "learning_rate": 2.8375000000000002e-05,
+      "loss": 0.0012,
+      "step": 274
+    },
+    {
+      "epoch": 1.1,
+      "grad_norm": 0.025945995002985,
+      "learning_rate": 2.825e-05,
+      "loss": 0.001,
+      "step": 275
+    },
+    {
+      "epoch": 1.104,
+      "grad_norm": 0.002442737575620413,
+      "learning_rate": 2.8125000000000003e-05,
+      "loss": 0.0,
+      "step": 276
+    },
+    {
+      "epoch": 1.108,
+      "grad_norm": 0.048240870237350464,
+      "learning_rate": 2.8000000000000003e-05,
+      "loss": 0.0001,
+      "step": 277
+    },
+    {
+      "epoch": 1.112,
+      "grad_norm": 0.025942618027329445,
+      "learning_rate": 2.7875e-05,
+      "loss": 0.0011,
+      "step": 278
+    },
+    {
+      "epoch": 1.116,
+      "grad_norm": 0.0020797245670109987,
+      "learning_rate": 2.7750000000000004e-05,
+      "loss": 0.0,
+      "step": 279
+    },
+    {
+      "epoch": 1.12,
+      "grad_norm": 0.0007060124189592898,
+      "learning_rate": 2.7625e-05,
+      "loss": 0.0,
+      "step": 280
+    },
+    {
+      "epoch": 1.124,
+      "grad_norm": 0.05631349980831146,
+      "learning_rate": 2.7500000000000004e-05,
+      "loss": 0.0055,
+      "step": 281
+    },
+    {
+      "epoch": 1.1280000000000001,
+      "grad_norm": 0.0010315525578334928,
+      "learning_rate": 2.7375e-05,
+      "loss": 0.0,
+      "step": 282
+    },
+    {
+      "epoch": 1.1320000000000001,
+      "grad_norm": 0.0005242518382146955,
+      "learning_rate": 2.725e-05,
+      "loss": 0.0,
+      "step": 283
+    },
+    {
+      "epoch": 1.1360000000000001,
+      "grad_norm": 0.0018985565984621644,
+      "learning_rate": 2.7125000000000002e-05,
+      "loss": 0.0,
+      "step": 284
+    },
+    {
+      "epoch": 1.1400000000000001,
+      "grad_norm": 0.016066987067461014,
+      "learning_rate": 2.7000000000000002e-05,
+      "loss": 0.0002,
+      "step": 285
+    },
+    {
+      "epoch": 1.144,
+      "grad_norm": 0.0014664505142718554,
+      "learning_rate": 2.6875e-05,
+      "loss": 0.0,
+      "step": 286
+    },
+    {
+      "epoch": 1.148,
+      "grad_norm": 0.037049151957035065,
+      "learning_rate": 2.6750000000000003e-05,
+      "loss": 0.0021,
+      "step": 287
+    },
+    {
+      "epoch": 1.152,
+      "grad_norm": 0.003005338367074728,
+      "learning_rate": 2.6625e-05,
+      "loss": 0.0,
+      "step": 288
+    },
+    {
+      "epoch": 1.156,
+      "grad_norm": 0.001254579983651638,
+      "learning_rate": 2.6500000000000004e-05,
+      "loss": 0.0,
+      "step": 289
+    },
+    {
+      "epoch": 1.16,
+      "grad_norm": 0.03759104385972023,
+      "learning_rate": 2.6375e-05,
+      "loss": 0.0011,
+      "step": 290
+    },
+    {
+      "epoch": 1.164,
+      "grad_norm": 0.1227162629365921,
+      "learning_rate": 2.625e-05,
+      "loss": 0.0012,
+      "step": 291
+    },
+    {
+      "epoch": 1.168,
+      "grad_norm": 0.03856627270579338,
+      "learning_rate": 2.6124999999999998e-05,
+      "loss": 0.0022,
+      "step": 292
+    },
+    {
+      "epoch": 1.172,
+      "grad_norm": 0.024704601615667343,
+      "learning_rate": 2.6000000000000002e-05,
+      "loss": 0.0002,
+      "step": 293
+    },
+    {
+      "epoch": 1.176,
+      "grad_norm": 0.0007420993642881513,
+      "learning_rate": 2.5875e-05,
+      "loss": 0.0,
+      "step": 294
+    },
+    {
+      "epoch": 1.18,
+      "grad_norm": 0.036463238298892975,
+      "learning_rate": 2.5750000000000002e-05,
+      "loss": 0.0022,
+      "step": 295
+    },
+    {
+      "epoch": 1.184,
+      "grad_norm": 0.15049096941947937,
+      "learning_rate": 2.5625e-05,
+      "loss": 0.0017,
+      "step": 296
+    },
+    {
+      "epoch": 1.188,
+      "grad_norm": 0.0008732306305319071,
+      "learning_rate": 2.5500000000000003e-05,
+      "loss": 0.0,
+      "step": 297
+    },
+    {
+      "epoch": 1.192,
+      "grad_norm": 0.0014195935800671577,
+      "learning_rate": 2.5375e-05,
+      "loss": 0.0,
+      "step": 298
+    },
+    {
+      "epoch": 1.196,
+      "grad_norm": 0.0003884572652168572,
+      "learning_rate": 2.525e-05,
+      "loss": 0.0,
+      "step": 299
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.0012515120906755328,
+      "learning_rate": 2.5124999999999997e-05,
+      "loss": 0.0,
+      "step": 300
+    },
+    {
+      "epoch": 1.204,
+      "grad_norm": 0.0020221874583512545,
+      "learning_rate": 2.5e-05,
+      "loss": 0.0,
+      "step": 301
+    },
+    {
+      "epoch": 1.208,
+      "grad_norm": 0.05133691802620888,
+      "learning_rate": 2.4875e-05,
+      "loss": 0.0019,
+      "step": 302
+    },
+    {
+      "epoch": 1.212,
+      "grad_norm": 0.0008544109296053648,
+      "learning_rate": 2.4750000000000002e-05,
+      "loss": 0.0,
+      "step": 303
+    },
+    {
+      "epoch": 1.216,
+      "grad_norm": 0.0014570467174053192,
+      "learning_rate": 2.4625000000000002e-05,
+      "loss": 0.0,
+      "step": 304
+    },
+    {
+      "epoch": 1.22,
+      "grad_norm": 0.0011205587070435286,
+      "learning_rate": 2.45e-05,
+      "loss": 0.0,
+      "step": 305
+    },
+    {
+      "epoch": 1.224,
+      "grad_norm": 0.02561991848051548,
+      "learning_rate": 2.4375e-05,
+      "loss": 0.0007,
+      "step": 306
+    },
+    {
+      "epoch": 1.228,
+      "grad_norm": 0.07412664592266083,
+      "learning_rate": 2.425e-05,
+      "loss": 0.0001,
+      "step": 307
+    },
+    {
+      "epoch": 1.232,
+      "grad_norm": 0.00037382490700110793,
+      "learning_rate": 2.4125e-05,
+      "loss": 0.0,
+      "step": 308
+    },
+    {
+      "epoch": 1.236,
+      "grad_norm": 0.00031290503102354705,
+      "learning_rate": 2.4e-05,
+      "loss": 0.0,
+      "step": 309
+    },
+    {
+      "epoch": 1.24,
+      "grad_norm": 0.004732325207442045,
+      "learning_rate": 2.3875e-05,
+      "loss": 0.0,
+      "step": 310
+    },
+    {
+      "epoch": 1.244,
+      "grad_norm": 0.020535649731755257,
+      "learning_rate": 2.375e-05,
+      "loss": 0.0001,
+      "step": 311
+    },
+    {
+      "epoch": 1.248,
+      "grad_norm": 0.03397507593035698,
+      "learning_rate": 2.3624999999999998e-05,
+      "loss": 0.0022,
+      "step": 312
+    },
+    {
+      "epoch": 1.252,
+      "grad_norm": 0.001010193140245974,
+      "learning_rate": 2.35e-05,
+      "loss": 0.0,
+      "step": 313
+    },
+    {
+      "epoch": 1.256,
+      "grad_norm": 0.0019539042841643095,
+      "learning_rate": 2.3375000000000002e-05,
+      "loss": 0.0,
+      "step": 314
+    },
+    {
+      "epoch": 1.26,
+      "grad_norm": 0.0031424874905496836,
+      "learning_rate": 2.3250000000000003e-05,
+      "loss": 0.0,
+      "step": 315
+    },
+    {
+      "epoch": 1.264,
+      "grad_norm": 0.02730289287865162,
+      "learning_rate": 2.3125000000000003e-05,
+      "loss": 0.0016,
+      "step": 316
+    },
+    {
+      "epoch": 1.268,
+      "grad_norm": 0.0027051756624132395,
+      "learning_rate": 2.3000000000000003e-05,
+      "loss": 0.0,
+      "step": 317
+    },
+    {
+      "epoch": 1.272,
+      "grad_norm": 0.03329765424132347,
+      "learning_rate": 2.2875e-05,
+      "loss": 0.0002,
+      "step": 318
+    },
+    {
+      "epoch": 1.276,
+      "grad_norm": 0.001193282543681562,
+      "learning_rate": 2.275e-05,
+      "loss": 0.0,
+      "step": 319
+    },
+    {
+      "epoch": 1.28,
+      "grad_norm": 0.031788360327482224,
+      "learning_rate": 2.2625e-05,
+      "loss": 0.0016,
+      "step": 320
+    },
+    {
+      "epoch": 1.284,
+      "grad_norm": 0.0014218081487342715,
+      "learning_rate": 2.25e-05,
+      "loss": 0.0,
+      "step": 321
+    },
+    {
+      "epoch": 1.288,
+      "grad_norm": 0.030428649857640266,
+      "learning_rate": 2.2375000000000002e-05,
+      "loss": 0.0002,
+      "step": 322
+    },
+    {
+      "epoch": 1.292,
+      "grad_norm": 0.006552345585078001,
+      "learning_rate": 2.2250000000000002e-05,
+      "loss": 0.0,
+      "step": 323
+    },
+    {
+      "epoch": 1.296,
+      "grad_norm": 0.015952585265040398,
+      "learning_rate": 2.2125000000000002e-05,
+      "loss": 0.0001,
+      "step": 324
+    },
+    {
+      "epoch": 1.3,
+      "grad_norm": 0.08881539851427078,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 0.0026,
+      "step": 325
+    },
+    {
+      "epoch": 1.304,
+      "grad_norm": 0.02848576381802559,
+      "learning_rate": 2.1875e-05,
+      "loss": 0.0015,
+      "step": 326
+    },
+    {
+      "epoch": 1.308,
+      "grad_norm": 0.025394096970558167,
+      "learning_rate": 2.175e-05,
+      "loss": 0.0001,
+      "step": 327
+    },
+    {
+      "epoch": 1.312,
+      "grad_norm": 0.10008195787668228,
+      "learning_rate": 2.1625e-05,
+      "loss": 0.0043,
+      "step": 328
+    },
+    {
+      "epoch": 1.316,
+      "grad_norm": 0.00036653419374488294,
+      "learning_rate": 2.15e-05,
+      "loss": 0.0,
+      "step": 329
+    },
+    {
+      "epoch": 1.32,
+      "grad_norm": 0.00029204797465354204,
+      "learning_rate": 2.1375e-05,
+      "loss": 0.0,
+      "step": 330
+    },
+    {
+      "epoch": 1.324,
+      "grad_norm": 0.08541154861450195,
+      "learning_rate": 2.125e-05,
+      "loss": 0.0006,
+      "step": 331
+    },
+    {
+      "epoch": 1.328,
+      "grad_norm": 0.0006791361374780536,
+      "learning_rate": 2.1125000000000002e-05,
+      "loss": 0.0,
+      "step": 332
+    },
+    {
+      "epoch": 1.332,
+      "grad_norm": 0.001022230600938201,
+      "learning_rate": 2.1e-05,
+      "loss": 0.0,
+      "step": 333
+    },
+    {
+      "epoch": 1.336,
+      "grad_norm": 0.0005869977758266032,
+      "learning_rate": 2.0875e-05,
+      "loss": 0.0,
+      "step": 334
+    },
+    {
+      "epoch": 1.34,
+      "grad_norm": 0.005593669600784779,
+      "learning_rate": 2.075e-05,
+      "loss": 0.0,
+      "step": 335
+    },
+    {
+      "epoch": 1.3439999999999999,
+      "grad_norm": 0.019772056490182877,
+      "learning_rate": 2.0625e-05,
+      "loss": 0.0002,
+      "step": 336
+    },
+    {
+      "epoch": 1.3479999999999999,
+      "grad_norm": 0.010209028609097004,
+      "learning_rate": 2.05e-05,
+      "loss": 0.0001,
+      "step": 337
+    },
+    {
+      "epoch": 1.3519999999999999,
+      "grad_norm": 0.0006436029216274619,
+      "learning_rate": 2.0375e-05,
+      "loss": 0.0,
+      "step": 338
+    },
+    {
+      "epoch": 1.3559999999999999,
+      "grad_norm": 0.0012518498115241528,
+      "learning_rate": 2.025e-05,
+      "loss": 0.0,
+      "step": 339
+    },
+    {
+      "epoch": 1.3599999999999999,
+      "grad_norm": 0.024785468354821205,
+      "learning_rate": 2.0125e-05,
+      "loss": 0.0014,
+      "step": 340
+    },
+    {
+      "epoch": 1.3639999999999999,
+      "grad_norm": 0.02820267528295517,
+      "learning_rate": 2e-05,
+      "loss": 0.0015,
+      "step": 341
+    },
+    {
+      "epoch": 1.3679999999999999,
+      "grad_norm": 0.018189677968621254,
+      "learning_rate": 1.9875000000000002e-05,
+      "loss": 0.0001,
+      "step": 342
+    },
+    {
+      "epoch": 1.3719999999999999,
+      "grad_norm": 0.0020627263002097607,
+      "learning_rate": 1.9750000000000002e-05,
+      "loss": 0.0,
+      "step": 343
+    },
+    {
+      "epoch": 1.376,
+      "grad_norm": 0.02280012145638466,
+      "learning_rate": 1.9625000000000003e-05,
+      "loss": 0.0001,
+      "step": 344
+    },
+    {
+      "epoch": 1.38,
+      "grad_norm": 0.03776618093252182,
+      "learning_rate": 1.9500000000000003e-05,
+      "loss": 0.0025,
+      "step": 345
+    },
+    {
+      "epoch": 1.384,
+      "grad_norm": 0.001024254597723484,
+      "learning_rate": 1.9375e-05,
+      "loss": 0.0,
+      "step": 346
+    },
+    {
+      "epoch": 1.388,
+      "grad_norm": 0.006558741442859173,
+      "learning_rate": 1.925e-05,
+      "loss": 0.0001,
+      "step": 347
+    },
+    {
+      "epoch": 1.392,
+      "grad_norm": 0.0002272271813126281,
+      "learning_rate": 1.9125e-05,
+      "loss": 0.0,
+      "step": 348
+    },
+    {
+      "epoch": 1.396,
+      "grad_norm": 0.0006058457656763494,
+      "learning_rate": 1.9e-05,
+      "loss": 0.0,
+      "step": 349
+    },
+    {
+      "epoch": 1.4,
+      "grad_norm": 0.025721488520503044,
+      "learning_rate": 1.8875e-05,
+      "loss": 0.0013,
+      "step": 350
+    },
+    {
+      "epoch": 1.404,
+      "grad_norm": 0.006544401869177818,
+      "learning_rate": 1.8750000000000002e-05,
+      "loss": 0.0,
+      "step": 351
+    },
+    {
+      "epoch": 1.408,
+      "grad_norm": 0.02833346277475357,
+      "learning_rate": 1.8625000000000002e-05,
+      "loss": 0.002,
+      "step": 352
+    },
+    {
+      "epoch": 1.412,
+      "grad_norm": 0.003127289703115821,
+      "learning_rate": 1.85e-05,
+      "loss": 0.0,
+      "step": 353
+    },
+    {
+      "epoch": 1.416,
+      "grad_norm": 0.0018629053374752402,
+      "learning_rate": 1.8375e-05,
+      "loss": 0.0,
+      "step": 354
+    },
+    {
+      "epoch": 1.42,
+      "grad_norm": 0.016095077618956566,
+      "learning_rate": 1.825e-05,
+      "loss": 0.0005,
+      "step": 355
+    },
+    {
+      "epoch": 1.424,
+      "grad_norm": 0.002553897676989436,
+      "learning_rate": 1.8125e-05,
+      "loss": 0.0,
+      "step": 356
+    },
+    {
+      "epoch": 1.428,
+      "grad_norm": 0.0001830149703891948,
+      "learning_rate": 1.8e-05,
+      "loss": 0.0,
+      "step": 357
+    },
+    {
+      "epoch": 1.432,
+      "grad_norm": 0.00017017306527122855,
+      "learning_rate": 1.7875e-05,
+      "loss": 0.0,
+      "step": 358
+    },
+    {
+      "epoch": 1.436,
+      "grad_norm": 0.0004768996441271156,
+      "learning_rate": 1.775e-05,
+      "loss": 0.0,
+      "step": 359
+    },
+    {
+      "epoch": 1.44,
+      "grad_norm": 0.008216165006160736,
+      "learning_rate": 1.7625e-05,
+      "loss": 0.0,
+      "step": 360
+    },
+    {
+      "epoch": 1.444,
+      "grad_norm": 0.00048328310367651284,
+      "learning_rate": 1.75e-05,
+      "loss": 0.0,
+      "step": 361
+    },
+    {
+      "epoch": 1.448,
+      "grad_norm": 0.001905157696455717,
+      "learning_rate": 1.7375e-05,
+      "loss": 0.0,
+      "step": 362
+    },
+    {
+      "epoch": 1.452,
+      "grad_norm": 0.04279350861907005,
+      "learning_rate": 1.725e-05,
+      "loss": 0.0003,
+      "step": 363
+    },
+    {
+      "epoch": 1.456,
+      "grad_norm": 0.00044308413635008037,
+      "learning_rate": 1.7125000000000003e-05,
+      "loss": 0.0,
+      "step": 364
+    },
+    {
+      "epoch": 1.46,
+      "grad_norm": 0.01364831067621708,
+      "learning_rate": 1.7000000000000003e-05,
+      "loss": 0.0001,
+      "step": 365
+    },
+    {
+      "epoch": 1.464,
+      "grad_norm": 0.0178745798766613,
+      "learning_rate": 1.6875000000000004e-05,
+      "loss": 0.0002,
+      "step": 366
+    },
+    {
+      "epoch": 1.468,
+      "grad_norm": 0.014486960135400295,
+      "learning_rate": 1.675e-05,
+      "loss": 0.0003,
+      "step": 367
+    },
+    {
+      "epoch": 1.472,
+      "grad_norm": 0.0017555502709001303,
+      "learning_rate": 1.6625e-05,
+      "loss": 0.0,
+      "step": 368
+    },
+    {
+      "epoch": 1.476,
+      "grad_norm": 0.0017262659966945648,
+      "learning_rate": 1.65e-05,
+      "loss": 0.0,
+      "step": 369
+    },
+    {
+      "epoch": 1.48,
+      "grad_norm": 0.0001670898200245574,
+      "learning_rate": 1.6375e-05,
+      "loss": 0.0,
+      "step": 370
+    },
+    {
+      "epoch": 1.484,
+      "grad_norm": 0.04310227930545807,
+      "learning_rate": 1.6250000000000002e-05,
+      "loss": 0.0,
+      "step": 371
+    },
+    {
+      "epoch": 1.488,
+      "grad_norm": 0.022789426147937775,
+      "learning_rate": 1.6125000000000002e-05,
+      "loss": 0.0012,
+      "step": 372
+    },
+    {
+      "epoch": 1.492,
+      "grad_norm": 0.00021801373804919422,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 0.0,
+      "step": 373
+    },
+    {
+      "epoch": 1.496,
+      "grad_norm": 0.0049751815386116505,
+      "learning_rate": 1.5875e-05,
+      "loss": 0.0,
+      "step": 374
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.025986915454268456,
+      "learning_rate": 1.575e-05,
+      "loss": 0.0009,
+      "step": 375
+    },
+    {
+      "epoch": 1.504,
+      "grad_norm": 0.027431977912783623,
+      "learning_rate": 1.5625e-05,
+      "loss": 0.0011,
+      "step": 376
+    },
+    {
+      "epoch": 1.508,
+      "grad_norm": 0.000507499324157834,
+      "learning_rate": 1.55e-05,
+      "loss": 0.0,
+      "step": 377
+    },
+    {
+      "epoch": 1.512,
+      "grad_norm": 0.0003128699027001858,
+      "learning_rate": 1.5375e-05,
+      "loss": 0.0,
+      "step": 378
+    },
+    {
+      "epoch": 1.516,
+      "grad_norm": 0.00046913442201912403,
+      "learning_rate": 1.525e-05,
+      "loss": 0.0,
+      "step": 379
+    },
+    {
+      "epoch": 1.52,
+      "grad_norm": 0.029943570494651794,
+      "learning_rate": 1.5125e-05,
+      "loss": 0.0017,
+      "step": 380
+    },
+    {
+      "epoch": 1.524,
+      "grad_norm": 0.002284953836351633,
+      "learning_rate": 1.5e-05,
+      "loss": 0.0,
+      "step": 381
+    },
+    {
+      "epoch": 1.528,
+      "grad_norm": 0.03019576705992222,
+      "learning_rate": 1.4875e-05,
+      "loss": 0.0024,
+      "step": 382
+    },
+    {
+      "epoch": 1.532,
+      "grad_norm": 0.0012708123540505767,
+      "learning_rate": 1.475e-05,
+      "loss": 0.0,
+      "step": 383
+    },
+    {
+      "epoch": 1.536,
+      "grad_norm": 0.023377712815999985,
+      "learning_rate": 1.4625e-05,
+      "loss": 0.0013,
+      "step": 384
+    },
+    {
+      "epoch": 1.54,
+      "grad_norm": 0.00021962085156701505,
+      "learning_rate": 1.45e-05,
+      "loss": 0.0,
+      "step": 385
+    },
+    {
+      "epoch": 1.544,
+      "grad_norm": 0.00874408334493637,
+      "learning_rate": 1.4374999999999999e-05,
+      "loss": 0.0,
+      "step": 386
+    },
+    {
+      "epoch": 1.548,
+      "grad_norm": 0.002368140732869506,
+      "learning_rate": 1.4249999999999999e-05,
+      "loss": 0.0,
+      "step": 387
+    },
+    {
+      "epoch": 1.552,
+      "grad_norm": 0.02442522905766964,
+      "learning_rate": 1.4125e-05,
+      "loss": 0.0014,
+      "step": 388
+    },
+    {
+      "epoch": 1.556,
+      "grad_norm": 0.022362807765603065,
+      "learning_rate": 1.4000000000000001e-05,
+      "loss": 0.0011,
+      "step": 389
+    },
+    {
+      "epoch": 1.56,
+      "grad_norm": 0.00017425463011022657,
+      "learning_rate": 1.3875000000000002e-05,
+      "loss": 0.0,
+      "step": 390
+    },
+    {
+      "epoch": 1.564,
+      "grad_norm": 0.03626161813735962,
+      "learning_rate": 1.3750000000000002e-05,
+      "loss": 0.0028,
+      "step": 391
+    },
+    {
+      "epoch": 1.568,
+      "grad_norm": 0.04060878977179527,
+      "learning_rate": 1.3625e-05,
+      "loss": 0.0033,
+      "step": 392
+    },
+    {
+      "epoch": 1.572,
+      "grad_norm": 0.04934659227728844,
+      "learning_rate": 1.3500000000000001e-05,
+      "loss": 0.0036,
+      "step": 393
+    },
+    {
+      "epoch": 1.576,
+      "grad_norm": 0.023239271715283394,
+      "learning_rate": 1.3375000000000002e-05,
+      "loss": 0.001,
+      "step": 394
+    },
+    {
+      "epoch": 1.58,
+      "grad_norm": 0.025479884818196297,
+      "learning_rate": 1.3250000000000002e-05,
+      "loss": 0.0012,
+      "step": 395
+    },
+    {
+      "epoch": 1.584,
+      "grad_norm": 0.024552635848522186,
+      "learning_rate": 1.3125e-05,
+      "loss": 0.0,
+      "step": 396
+    },
+    {
+      "epoch": 1.588,
+      "grad_norm": 0.02806110680103302,
+      "learning_rate": 1.3000000000000001e-05,
+      "loss": 0.0013,
+      "step": 397
+    },
+    {
+      "epoch": 1.592,
+      "grad_norm": 0.0003899558796547353,
+      "learning_rate": 1.2875000000000001e-05,
+      "loss": 0.0,
+      "step": 398
+    },
+    {
+      "epoch": 1.596,
+      "grad_norm": 0.022901885211467743,
+      "learning_rate": 1.2750000000000002e-05,
+      "loss": 0.001,
+      "step": 399
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.00024660167400725186,
+      "learning_rate": 1.2625e-05,
+      "loss": 0.0,
+      "step": 400
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 500,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 23568689516544.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

0624-7B/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0115359d4f041e411883c09f27baa1de92bbfe16d482bcc8036c8e3273de6ef2
+size 7736

0624-7B/zero_to_fp32.py

@@ -0,0 +1,604 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_file,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)