upload

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

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
+}

config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "moussaKam/fr-qwen-3B-base",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 36,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.1",
+  "use_cache": false,
+  "use_mrope": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.46.1"
+}

latest

@@ -0,0 +1 @@
+global_step3962

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9bbcb05eaaeeba15b23df2a066c482e8f653ed65e2219b1331160012a64ba8c
+size 4957560304

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f7b0c7dbcc57dbf41f14791ca57edcfad3dfbc935d3a12f1ec74150cb76c693
+size 1836696752

model.safetensors.index.json

@@ -0,0 +1,442 @@
+{
+  "metadata": {
+    "total_size": 6794207232
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.bias": "model-00002-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "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
+  }
+}

tokenizer.json

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

tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "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_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "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 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 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": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

trainer_state.json

@@ -0,0 +1,2805 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9998107374929026,
+  "eval_steps": 500,
+  "global_step": 3962,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025235000946312535,
+      "grad_norm": 4.200274467468262,
+      "learning_rate": 1.2594458438287156e-06,
+      "loss": 0.9928,
+      "step": 10
+    },
+    {
+      "epoch": 0.005047000189262507,
+      "grad_norm": 5.791783332824707,
+      "learning_rate": 2.518891687657431e-06,
+      "loss": 0.9146,
+      "step": 20
+    },
+    {
+      "epoch": 0.007570500283893761,
+      "grad_norm": 1.0928738117218018,
+      "learning_rate": 3.7783375314861467e-06,
+      "loss": 0.8451,
+      "step": 30
+    },
+    {
+      "epoch": 0.010094000378525014,
+      "grad_norm": 0.792613685131073,
+      "learning_rate": 5.037783375314862e-06,
+      "loss": 0.8395,
+      "step": 40
+    },
+    {
+      "epoch": 0.012617500473156268,
+      "grad_norm": 0.8177407383918762,
+      "learning_rate": 6.297229219143577e-06,
+      "loss": 0.7893,
+      "step": 50
+    },
+    {
+      "epoch": 0.015141000567787522,
+      "grad_norm": 0.7625726461410522,
+      "learning_rate": 7.556675062972293e-06,
+      "loss": 0.8163,
+      "step": 60
+    },
+    {
+      "epoch": 0.017664500662418776,
+      "grad_norm": 0.8031908869743347,
+      "learning_rate": 8.816120906801008e-06,
+      "loss": 0.8137,
+      "step": 70
+    },
+    {
+      "epoch": 0.020188000757050028,
+      "grad_norm": 0.8055213093757629,
+      "learning_rate": 1.0075566750629725e-05,
+      "loss": 0.8012,
+      "step": 80
+    },
+    {
+      "epoch": 0.022711500851681284,
+      "grad_norm": 0.8089568614959717,
+      "learning_rate": 1.133501259445844e-05,
+      "loss": 0.7879,
+      "step": 90
+    },
+    {
+      "epoch": 0.025235000946312536,
+      "grad_norm": 0.798632025718689,
+      "learning_rate": 1.2594458438287154e-05,
+      "loss": 0.7832,
+      "step": 100
+    },
+    {
+      "epoch": 0.027758501040943788,
+      "grad_norm": 0.8331993818283081,
+      "learning_rate": 1.385390428211587e-05,
+      "loss": 0.8476,
+      "step": 110
+    },
+    {
+      "epoch": 0.030282001135575044,
+      "grad_norm": 0.6234100461006165,
+      "learning_rate": 1.5113350125944587e-05,
+      "loss": 0.8122,
+      "step": 120
+    },
+    {
+      "epoch": 0.0328055012302063,
+      "grad_norm": 0.841749906539917,
+      "learning_rate": 1.63727959697733e-05,
+      "loss": 0.7766,
+      "step": 130
+    },
+    {
+      "epoch": 0.03532900132483755,
+      "grad_norm": 0.8041658997535706,
+      "learning_rate": 1.7632241813602016e-05,
+      "loss": 0.7657,
+      "step": 140
+    },
+    {
+      "epoch": 0.037852501419468804,
+      "grad_norm": 0.8009690046310425,
+      "learning_rate": 1.8891687657430733e-05,
+      "loss": 0.8194,
+      "step": 150
+    },
+    {
+      "epoch": 0.040376001514100056,
+      "grad_norm": 0.776654064655304,
+      "learning_rate": 2.015113350125945e-05,
+      "loss": 0.8009,
+      "step": 160
+    },
+    {
+      "epoch": 0.04289950160873131,
+      "grad_norm": 0.7446156740188599,
+      "learning_rate": 2.1410579345088162e-05,
+      "loss": 0.7832,
+      "step": 170
+    },
+    {
+      "epoch": 0.04542300170336257,
+      "grad_norm": 0.7450286149978638,
+      "learning_rate": 2.267002518891688e-05,
+      "loss": 0.8252,
+      "step": 180
+    },
+    {
+      "epoch": 0.04794650179799382,
+      "grad_norm": 0.788154125213623,
+      "learning_rate": 2.392947103274559e-05,
+      "loss": 0.8131,
+      "step": 190
+    },
+    {
+      "epoch": 0.05047000189262507,
+      "grad_norm": 0.8005194067955017,
+      "learning_rate": 2.5188916876574308e-05,
+      "loss": 0.816,
+      "step": 200
+    },
+    {
+      "epoch": 0.052993501987256324,
+      "grad_norm": 0.849112868309021,
+      "learning_rate": 2.6448362720403024e-05,
+      "loss": 0.8028,
+      "step": 210
+    },
+    {
+      "epoch": 0.055517002081887576,
+      "grad_norm": 0.7202879786491394,
+      "learning_rate": 2.770780856423174e-05,
+      "loss": 0.8145,
+      "step": 220
+    },
+    {
+      "epoch": 0.058040502176518835,
+      "grad_norm": 0.8000121116638184,
+      "learning_rate": 2.8967254408060457e-05,
+      "loss": 0.7571,
+      "step": 230
+    },
+    {
+      "epoch": 0.06056400227115009,
+      "grad_norm": 0.7495560050010681,
+      "learning_rate": 3.0226700251889174e-05,
+      "loss": 0.768,
+      "step": 240
+    },
+    {
+      "epoch": 0.06308750236578134,
+      "grad_norm": 0.8274487853050232,
+      "learning_rate": 3.148614609571788e-05,
+      "loss": 0.8411,
+      "step": 250
+    },
+    {
+      "epoch": 0.0656110024604126,
+      "grad_norm": 0.9089523553848267,
+      "learning_rate": 3.27455919395466e-05,
+      "loss": 0.8114,
+      "step": 260
+    },
+    {
+      "epoch": 0.06813450255504384,
+      "grad_norm": 0.8551489114761353,
+      "learning_rate": 3.4005037783375316e-05,
+      "loss": 0.8122,
+      "step": 270
+    },
+    {
+      "epoch": 0.0706580026496751,
+      "grad_norm": 0.8696727156639099,
+      "learning_rate": 3.526448362720403e-05,
+      "loss": 0.7858,
+      "step": 280
+    },
+    {
+      "epoch": 0.07318150274430635,
+      "grad_norm": 0.8770012259483337,
+      "learning_rate": 3.652392947103275e-05,
+      "loss": 0.7971,
+      "step": 290
+    },
+    {
+      "epoch": 0.07570500283893761,
+      "grad_norm": 0.8717426657676697,
+      "learning_rate": 3.7783375314861465e-05,
+      "loss": 0.7703,
+      "step": 300
+    },
+    {
+      "epoch": 0.07822850293356887,
+      "grad_norm": 0.8454267382621765,
+      "learning_rate": 3.904282115869018e-05,
+      "loss": 0.8146,
+      "step": 310
+    },
+    {
+      "epoch": 0.08075200302820011,
+      "grad_norm": 0.7605656981468201,
+      "learning_rate": 4.03022670025189e-05,
+      "loss": 0.7875,
+      "step": 320
+    },
+    {
+      "epoch": 0.08327550312283137,
+      "grad_norm": 0.7611861228942871,
+      "learning_rate": 4.1561712846347615e-05,
+      "loss": 0.8089,
+      "step": 330
+    },
+    {
+      "epoch": 0.08579900321746262,
+      "grad_norm": 0.7927576303482056,
+      "learning_rate": 4.2821158690176324e-05,
+      "loss": 0.7802,
+      "step": 340
+    },
+    {
+      "epoch": 0.08832250331209388,
+      "grad_norm": 0.8251197338104248,
+      "learning_rate": 4.408060453400504e-05,
+      "loss": 0.7648,
+      "step": 350
+    },
+    {
+      "epoch": 0.09084600340672513,
+      "grad_norm": 0.7899374961853027,
+      "learning_rate": 4.534005037783376e-05,
+      "loss": 0.8011,
+      "step": 360
+    },
+    {
+      "epoch": 0.09336950350135638,
+      "grad_norm": 0.8102702498435974,
+      "learning_rate": 4.659949622166247e-05,
+      "loss": 0.8063,
+      "step": 370
+    },
+    {
+      "epoch": 0.09589300359598764,
+      "grad_norm": 0.8451895713806152,
+      "learning_rate": 4.785894206549118e-05,
+      "loss": 0.8162,
+      "step": 380
+    },
+    {
+      "epoch": 0.09841650369061888,
+      "grad_norm": 0.664681077003479,
+      "learning_rate": 4.91183879093199e-05,
+      "loss": 0.8095,
+      "step": 390
+    },
+    {
+      "epoch": 0.10094000378525014,
+      "grad_norm": 1.5052850246429443,
+      "learning_rate": 4.999991263591223e-05,
+      "loss": 0.7586,
+      "step": 400
+    },
+    {
+      "epoch": 0.1034635038798814,
+      "grad_norm": 0.7986663579940796,
+      "learning_rate": 4.9998359513560176e-05,
+      "loss": 0.783,
+      "step": 410
+    },
+    {
+      "epoch": 0.10598700397451265,
+      "grad_norm": 0.7573216557502747,
+      "learning_rate": 4.999486510586282e-05,
+      "loss": 0.8245,
+      "step": 420
+    },
+    {
+      "epoch": 0.1085105040691439,
+      "grad_norm": 0.7997973561286926,
+      "learning_rate": 4.9989429684183686e-05,
+      "loss": 0.8228,
+      "step": 430
+    },
+    {
+      "epoch": 0.11103400416377515,
+      "grad_norm": 0.7936347723007202,
+      "learning_rate": 4.9982053670618626e-05,
+      "loss": 0.8098,
+      "step": 440
+    },
+    {
+      "epoch": 0.11355750425840641,
+      "grad_norm": 0.7175100445747375,
+      "learning_rate": 4.997273763796312e-05,
+      "loss": 0.7959,
+      "step": 450
+    },
+    {
+      "epoch": 0.11608100435303767,
+      "grad_norm": 0.8202365040779114,
+      "learning_rate": 4.996148230966775e-05,
+      "loss": 0.7682,
+      "step": 460
+    },
+    {
+      "epoch": 0.11860450444766892,
+      "grad_norm": 0.7965167164802551,
+      "learning_rate": 4.994828855978202e-05,
+      "loss": 0.8459,
+      "step": 470
+    },
+    {
+      "epoch": 0.12112800454230017,
+      "grad_norm": 0.7704299092292786,
+      "learning_rate": 4.99331574128865e-05,
+      "loss": 0.7659,
+      "step": 480
+    },
+    {
+      "epoch": 0.12365150463693142,
+      "grad_norm": 0.7578620910644531,
+      "learning_rate": 4.991609004401324e-05,
+      "loss": 0.8097,
+      "step": 490
+    },
+    {
+      "epoch": 0.12617500473156268,
+      "grad_norm": 0.7264330983161926,
+      "learning_rate": 4.989708777855453e-05,
+      "loss": 0.8092,
+      "step": 500
+    },
+    {
+      "epoch": 0.12869850482619394,
+      "grad_norm": 0.8330686092376709,
+      "learning_rate": 4.9876152092159994e-05,
+      "loss": 0.8352,
+      "step": 510
+    },
+    {
+      "epoch": 0.1312220049208252,
+      "grad_norm": 0.6997131705284119,
+      "learning_rate": 4.985328461062195e-05,
+      "loss": 0.8109,
+      "step": 520
+    },
+    {
+      "epoch": 0.13374550501545643,
+      "grad_norm": 0.6822782754898071,
+      "learning_rate": 4.98284871097492e-05,
+      "loss": 0.8048,
+      "step": 530
+    },
+    {
+      "epoch": 0.1362690051100877,
+      "grad_norm": 0.741365373134613,
+      "learning_rate": 4.98017615152291e-05,
+      "loss": 0.8053,
+      "step": 540
+    },
+    {
+      "epoch": 0.13879250520471895,
+      "grad_norm": 0.7147175669670105,
+      "learning_rate": 4.977310990247807e-05,
+      "loss": 0.8005,
+      "step": 550
+    },
+    {
+      "epoch": 0.1413160052993502,
+      "grad_norm": 0.8730887770652771,
+      "learning_rate": 4.974253449648031e-05,
+      "loss": 0.8088,
+      "step": 560
+    },
+    {
+      "epoch": 0.14383950539398146,
+      "grad_norm": 0.7284607887268066,
+      "learning_rate": 4.971003767161516e-05,
+      "loss": 0.8023,
+      "step": 570
+    },
+    {
+      "epoch": 0.1463630054886127,
+      "grad_norm": 0.737747311592102,
+      "learning_rate": 4.9675621951472584e-05,
+      "loss": 0.808,
+      "step": 580
+    },
+    {
+      "epoch": 0.14888650558324396,
+      "grad_norm": 0.6779095530509949,
+      "learning_rate": 4.9639290008657304e-05,
+      "loss": 0.8142,
+      "step": 590
+    },
+    {
+      "epoch": 0.15141000567787521,
+      "grad_norm": 0.7556443214416504,
+      "learning_rate": 4.960104466458118e-05,
+      "loss": 0.8131,
+      "step": 600
+    },
+    {
+      "epoch": 0.15393350577250647,
+      "grad_norm": 0.7624189257621765,
+      "learning_rate": 4.956088888924414e-05,
+      "loss": 0.7894,
+      "step": 610
+    },
+    {
+      "epoch": 0.15645700586713773,
+      "grad_norm": 0.7633741497993469,
+      "learning_rate": 4.951882580100353e-05,
+      "loss": 0.8126,
+      "step": 620
+    },
+    {
+      "epoch": 0.15898050596176896,
+      "grad_norm": 0.6966682076454163,
+      "learning_rate": 4.947485866633199e-05,
+      "loss": 0.8111,
+      "step": 630
+    },
+    {
+      "epoch": 0.16150400605640022,
+      "grad_norm": 0.7578861117362976,
+      "learning_rate": 4.94289908995637e-05,
+      "loss": 0.7498,
+      "step": 640
+    },
+    {
+      "epoch": 0.16402750615103148,
+      "grad_norm": 0.7788159847259521,
+      "learning_rate": 4.938122606262936e-05,
+      "loss": 0.8281,
+      "step": 650
+    },
+    {
+      "epoch": 0.16655100624566274,
+      "grad_norm": 0.7012534737586975,
+      "learning_rate": 4.9331567864779457e-05,
+      "loss": 0.8298,
+      "step": 660
+    },
+    {
+      "epoch": 0.169074506340294,
+      "grad_norm": 0.6714362502098083,
+      "learning_rate": 4.928002016229634e-05,
+      "loss": 0.8126,
+      "step": 670
+    },
+    {
+      "epoch": 0.17159800643492523,
+      "grad_norm": 0.6076031923294067,
+      "learning_rate": 4.9226586958194647e-05,
+      "loss": 0.8289,
+      "step": 680
+    },
+    {
+      "epoch": 0.1741215065295565,
+      "grad_norm": 0.6331019997596741,
+      "learning_rate": 4.9171272401910504e-05,
+      "loss": 0.7602,
+      "step": 690
+    },
+    {
+      "epoch": 0.17664500662418775,
+      "grad_norm": 2.956536293029785,
+      "learning_rate": 4.9114080788979284e-05,
+      "loss": 0.8143,
+      "step": 700
+    },
+    {
+      "epoch": 0.179168506718819,
+      "grad_norm": 0.7164005041122437,
+      "learning_rate": 4.905501656070202e-05,
+      "loss": 0.8158,
+      "step": 710
+    },
+    {
+      "epoch": 0.18169200681345027,
+      "grad_norm": 0.6963945031166077,
+      "learning_rate": 4.8994084303800525e-05,
+      "loss": 0.8004,
+      "step": 720
+    },
+    {
+      "epoch": 0.1842155069080815,
+      "grad_norm": 0.7561419010162354,
+      "learning_rate": 4.89312887500612e-05,
+      "loss": 0.8064,
+      "step": 730
+    },
+    {
+      "epoch": 0.18673900700271276,
+      "grad_norm": 0.672164261341095,
+      "learning_rate": 4.8866634775967544e-05,
+      "loss": 0.8111,
+      "step": 740
+    },
+    {
+      "epoch": 0.18926250709734402,
+      "grad_norm": 0.6340550184249878,
+      "learning_rate": 4.880012740232154e-05,
+      "loss": 0.8009,
+      "step": 750
+    },
+    {
+      "epoch": 0.19178600719197528,
+      "grad_norm": 0.7655452489852905,
+      "learning_rate": 4.873177179385368e-05,
+      "loss": 0.7912,
+      "step": 760
+    },
+    {
+      "epoch": 0.19430950728660654,
+      "grad_norm": 0.7131490111351013,
+      "learning_rate": 4.866157325882192e-05,
+      "loss": 0.8019,
+      "step": 770
+    },
+    {
+      "epoch": 0.19683300738123777,
+      "grad_norm": 0.7386584877967834,
+      "learning_rate": 4.858953724859948e-05,
+      "loss": 0.7934,
+      "step": 780
+    },
+    {
+      "epoch": 0.19935650747586903,
+      "grad_norm": 0.6958081126213074,
+      "learning_rate": 4.851566935725147e-05,
+      "loss": 0.7727,
+      "step": 790
+    },
+    {
+      "epoch": 0.20188000757050029,
+      "grad_norm": 0.660095751285553,
+      "learning_rate": 4.843997532110051e-05,
+      "loss": 0.8002,
+      "step": 800
+    },
+    {
+      "epoch": 0.20440350766513155,
+      "grad_norm": 0.8626520037651062,
+      "learning_rate": 4.836246101828124e-05,
+      "loss": 0.8117,
+      "step": 810
+    },
+    {
+      "epoch": 0.2069270077597628,
+      "grad_norm": 0.608925998210907,
+      "learning_rate": 4.828313246828386e-05,
+      "loss": 0.8119,
+      "step": 820
+    },
+    {
+      "epoch": 0.20945050785439404,
+      "grad_norm": 0.6617856621742249,
+      "learning_rate": 4.820199583148667e-05,
+      "loss": 0.8057,
+      "step": 830
+    },
+    {
+      "epoch": 0.2119740079490253,
+      "grad_norm": 0.6674720644950867,
+      "learning_rate": 4.811905740867769e-05,
+      "loss": 0.815,
+      "step": 840
+    },
+    {
+      "epoch": 0.21449750804365655,
+      "grad_norm": 0.7636524438858032,
+      "learning_rate": 4.803432364056535e-05,
+      "loss": 0.8113,
+      "step": 850
+    },
+    {
+      "epoch": 0.2170210081382878,
+      "grad_norm": 0.6965427398681641,
+      "learning_rate": 4.794780110727832e-05,
+      "loss": 0.783,
+      "step": 860
+    },
+    {
+      "epoch": 0.21954450823291907,
+      "grad_norm": 0.7165718078613281,
+      "learning_rate": 4.785949652785453e-05,
+      "loss": 0.8162,
+      "step": 870
+    },
+    {
+      "epoch": 0.2220680083275503,
+      "grad_norm": 0.8301019072532654,
+      "learning_rate": 4.776941675971941e-05,
+      "loss": 0.7954,
+      "step": 880
+    },
+    {
+      "epoch": 0.22459150842218156,
+      "grad_norm": 0.6930355429649353,
+      "learning_rate": 4.767756879815334e-05,
+      "loss": 0.7955,
+      "step": 890
+    },
+    {
+      "epoch": 0.22711500851681282,
+      "grad_norm": 0.7135087847709656,
+      "learning_rate": 4.758395977574841e-05,
+      "loss": 0.8324,
+      "step": 900
+    },
+    {
+      "epoch": 0.22963850861144408,
+      "grad_norm": 0.6820036172866821,
+      "learning_rate": 4.748859696185458e-05,
+      "loss": 0.8207,
+      "step": 910
+    },
+    {
+      "epoch": 0.23216200870607534,
+      "grad_norm": 0.6496434807777405,
+      "learning_rate": 4.739148776201512e-05,
+      "loss": 0.7498,
+      "step": 920
+    },
+    {
+      "epoch": 0.23468550880070657,
+      "grad_norm": 0.692870020866394,
+      "learning_rate": 4.729263971739154e-05,
+      "loss": 0.8109,
+      "step": 930
+    },
+    {
+      "epoch": 0.23720900889533783,
+      "grad_norm": 0.6766488552093506,
+      "learning_rate": 4.719206050417796e-05,
+      "loss": 0.7938,
+      "step": 940
+    },
+    {
+      "epoch": 0.2397325089899691,
+      "grad_norm": 0.6905511617660522,
+      "learning_rate": 4.7089757933005016e-05,
+      "loss": 0.8036,
+      "step": 950
+    },
+    {
+      "epoch": 0.24225600908460035,
+      "grad_norm": 1.9566072225570679,
+      "learning_rate": 4.698573994833332e-05,
+      "loss": 0.7954,
+      "step": 960
+    },
+    {
+      "epoch": 0.2447795091792316,
+      "grad_norm": 0.7477098107337952,
+      "learning_rate": 4.688001462783648e-05,
+      "loss": 0.7862,
+      "step": 970
+    },
+    {
+      "epoch": 0.24730300927386284,
+      "grad_norm": 0.719272792339325,
+      "learning_rate": 4.6772590181773866e-05,
+      "loss": 0.7626,
+      "step": 980
+    },
+    {
+      "epoch": 0.2498265093684941,
+      "grad_norm": 0.6895411014556885,
+      "learning_rate": 4.6663474952353004e-05,
+      "loss": 0.7704,
+      "step": 990
+    },
+    {
+      "epoch": 0.25235000946312536,
+      "grad_norm": 3.369716167449951,
+      "learning_rate": 4.6552677413081756e-05,
+      "loss": 0.8274,
+      "step": 1000
+    },
+    {
+      "epoch": 0.2548735095577566,
+      "grad_norm": 1.2634356021881104,
+      "learning_rate": 4.644020616811029e-05,
+      "loss": 0.813,
+      "step": 1010
+    },
+    {
+      "epoch": 0.2573970096523879,
+      "grad_norm": 0.8203848600387573,
+      "learning_rate": 4.6326069951562924e-05,
+      "loss": 0.8331,
+      "step": 1020
+    },
+    {
+      "epoch": 0.25992050974701914,
+      "grad_norm": 0.6396493911743164,
+      "learning_rate": 4.6210277626859856e-05,
+      "loss": 0.7532,
+      "step": 1030
+    },
+    {
+      "epoch": 0.2624440098416504,
+      "grad_norm": 0.6259830594062805,
+      "learning_rate": 4.609283818602884e-05,
+      "loss": 0.8041,
+      "step": 1040
+    },
+    {
+      "epoch": 0.2649675099362816,
+      "grad_norm": 0.6914517283439636,
+      "learning_rate": 4.5973760749006963e-05,
+      "loss": 0.8101,
+      "step": 1050
+    },
+    {
+      "epoch": 0.26749101003091286,
+      "grad_norm": 0.652829110622406,
+      "learning_rate": 4.585305456293235e-05,
+      "loss": 0.8394,
+      "step": 1060
+    },
+    {
+      "epoch": 0.2700145101255441,
+      "grad_norm": 0.9240155220031738,
+      "learning_rate": 4.5730729001426083e-05,
+      "loss": 0.8135,
+      "step": 1070
+    },
+    {
+      "epoch": 0.2725380102201754,
+      "grad_norm": 0.6168021559715271,
+      "learning_rate": 4.5606793563864316e-05,
+      "loss": 0.7875,
+      "step": 1080
+    },
+    {
+      "epoch": 0.27506151031480663,
+      "grad_norm": 0.6858901381492615,
+      "learning_rate": 4.548125787464054e-05,
+      "loss": 0.7863,
+      "step": 1090
+    },
+    {
+      "epoch": 0.2775850104094379,
+      "grad_norm": 0.6377461552619934,
+      "learning_rate": 4.535413168241821e-05,
+      "loss": 0.7945,
+      "step": 1100
+    },
+    {
+      "epoch": 0.28010851050406915,
+      "grad_norm": 0.644579291343689,
+      "learning_rate": 4.522542485937369e-05,
+      "loss": 0.8294,
+      "step": 1110
+    },
+    {
+      "epoch": 0.2826320105987004,
+      "grad_norm": 0.6372377872467041,
+      "learning_rate": 4.509514740042962e-05,
+      "loss": 0.7961,
+      "step": 1120
+    },
+    {
+      "epoch": 0.28515551069333167,
+      "grad_norm": 0.7171216011047363,
+      "learning_rate": 4.496330942247873e-05,
+      "loss": 0.7968,
+      "step": 1130
+    },
+    {
+      "epoch": 0.28767901078796293,
+      "grad_norm": 0.6972722411155701,
+      "learning_rate": 4.482992116359824e-05,
+      "loss": 0.7841,
+      "step": 1140
+    },
+    {
+      "epoch": 0.29020251088259413,
+      "grad_norm": 0.6313626170158386,
+      "learning_rate": 4.469499298225473e-05,
+      "loss": 0.7567,
+      "step": 1150
+    },
+    {
+      "epoch": 0.2927260109772254,
+      "grad_norm": 0.6984760761260986,
+      "learning_rate": 4.455853535649984e-05,
+      "loss": 0.7877,
+      "step": 1160
+    },
+    {
+      "epoch": 0.29524951107185665,
+      "grad_norm": 0.6628227829933167,
+      "learning_rate": 4.442055888315646e-05,
+      "loss": 0.7953,
+      "step": 1170
+    },
+    {
+      "epoch": 0.2977730111664879,
+      "grad_norm": 0.6865362524986267,
+      "learning_rate": 4.4281074276995936e-05,
+      "loss": 0.7553,
+      "step": 1180
+    },
+    {
+      "epoch": 0.30029651126111917,
+      "grad_norm": 0.7026821970939636,
+      "learning_rate": 4.4140092369905914e-05,
+      "loss": 0.7683,
+      "step": 1190
+    },
+    {
+      "epoch": 0.30282001135575043,
+      "grad_norm": 0.8070855736732483,
+      "learning_rate": 4.399762411004922e-05,
+      "loss": 0.8004,
+      "step": 1200
+    },
+    {
+      "epoch": 0.3053435114503817,
+      "grad_norm": 0.7692244648933411,
+      "learning_rate": 4.3853680561013647e-05,
+      "loss": 0.8224,
+      "step": 1210
+    },
+    {
+      "epoch": 0.30786701154501295,
+      "grad_norm": 0.6959588527679443,
+      "learning_rate": 4.370827290095277e-05,
+      "loss": 0.792,
+      "step": 1220
+    },
+    {
+      "epoch": 0.3103905116396442,
+      "grad_norm": 0.6718862652778625,
+      "learning_rate": 4.356141242171795e-05,
+      "loss": 0.7937,
+      "step": 1230
+    },
+    {
+      "epoch": 0.31291401173427547,
+      "grad_norm": 0.6850164532661438,
+      "learning_rate": 4.3413110527981406e-05,
+      "loss": 0.7705,
+      "step": 1240
+    },
+    {
+      "epoch": 0.31543751182890667,
+      "grad_norm": 0.7577234506607056,
+      "learning_rate": 4.3263378736350566e-05,
+      "loss": 0.7971,
+      "step": 1250
+    },
+    {
+      "epoch": 0.31796101192353793,
+      "grad_norm": 0.7119166851043701,
+      "learning_rate": 4.311222867447375e-05,
+      "loss": 0.8302,
+      "step": 1260
+    },
+    {
+      "epoch": 0.3204845120181692,
+      "grad_norm": 0.7407262921333313,
+      "learning_rate": 4.295967208013717e-05,
+      "loss": 0.7944,
+      "step": 1270
+    },
+    {
+      "epoch": 0.32300801211280045,
+      "grad_norm": 0.6641649007797241,
+      "learning_rate": 4.280572080035348e-05,
+      "loss": 0.7934,
+      "step": 1280
+    },
+    {
+      "epoch": 0.3255315122074317,
+      "grad_norm": 0.6886960864067078,
+      "learning_rate": 4.2650386790441696e-05,
+      "loss": 0.7839,
+      "step": 1290
+    },
+    {
+      "epoch": 0.32805501230206296,
+      "grad_norm": 0.6761602163314819,
+      "learning_rate": 4.2493682113098855e-05,
+      "loss": 0.7943,
+      "step": 1300
+    },
+    {
+      "epoch": 0.3305785123966942,
+      "grad_norm": 0.6564992666244507,
+      "learning_rate": 4.233561893746323e-05,
+      "loss": 0.8184,
+      "step": 1310
+    },
+    {
+      "epoch": 0.3331020124913255,
+      "grad_norm": 0.6928525567054749,
+      "learning_rate": 4.217620953816935e-05,
+      "loss": 0.7758,
+      "step": 1320
+    },
+    {
+      "epoch": 0.33562551258595674,
+      "grad_norm": 0.6335028409957886,
+      "learning_rate": 4.2015466294394756e-05,
+      "loss": 0.8091,
+      "step": 1330
+    },
+    {
+      "epoch": 0.338149012680588,
+      "grad_norm": 1.0897998809814453,
+      "learning_rate": 4.185340168889868e-05,
+      "loss": 0.7807,
+      "step": 1340
+    },
+    {
+      "epoch": 0.3406725127752192,
+      "grad_norm": 0.6639924645423889,
+      "learning_rate": 4.169002830705274e-05,
+      "loss": 0.7803,
+      "step": 1350
+    },
+    {
+      "epoch": 0.34319601286985046,
+      "grad_norm": 0.6235373020172119,
+      "learning_rate": 4.152535883586352e-05,
+      "loss": 0.7651,
+      "step": 1360
+    },
+    {
+      "epoch": 0.3457195129644817,
+      "grad_norm": 0.6909653544425964,
+      "learning_rate": 4.135940606298738e-05,
+      "loss": 0.7748,
+      "step": 1370
+    },
+    {
+      "epoch": 0.348243013059113,
+      "grad_norm": 0.6260821223258972,
+      "learning_rate": 4.119218287573743e-05,
+      "loss": 0.7624,
+      "step": 1380
+    },
+    {
+      "epoch": 0.35076651315374424,
+      "grad_norm": 0.6416032910346985,
+      "learning_rate": 4.102370226008271e-05,
+      "loss": 0.813,
+      "step": 1390
+    },
+    {
+      "epoch": 0.3532900132483755,
+      "grad_norm": 0.6173393726348877,
+      "learning_rate": 4.085397729963976e-05,
+      "loss": 0.7767,
+      "step": 1400
+    },
+    {
+      "epoch": 0.35581351334300676,
+      "grad_norm": 0.7046598792076111,
+      "learning_rate": 4.06830211746566e-05,
+      "loss": 0.8095,
+      "step": 1410
+    },
+    {
+      "epoch": 0.358337013437638,
+      "grad_norm": 0.7577833533287048,
+      "learning_rate": 4.051084716098921e-05,
+      "loss": 0.7859,
+      "step": 1420
+    },
+    {
+      "epoch": 0.3608605135322693,
+      "grad_norm": 0.6137785911560059,
+      "learning_rate": 4.0337468629070496e-05,
+      "loss": 0.7879,
+      "step": 1430
+    },
+    {
+      "epoch": 0.36338401362690054,
+      "grad_norm": 0.6560728549957275,
+      "learning_rate": 4.016289904287212e-05,
+      "loss": 0.7811,
+      "step": 1440
+    },
+    {
+      "epoch": 0.36590751372153174,
+      "grad_norm": 0.6869454383850098,
+      "learning_rate": 3.9987151958858794e-05,
+      "loss": 0.7954,
+      "step": 1450
+    },
+    {
+      "epoch": 0.368431013816163,
+      "grad_norm": 0.6819769144058228,
+      "learning_rate": 3.981024102493566e-05,
+      "loss": 0.7607,
+      "step": 1460
+    },
+    {
+      "epoch": 0.37095451391079426,
+      "grad_norm": 0.694311797618866,
+      "learning_rate": 3.963217997938834e-05,
+      "loss": 0.7926,
+      "step": 1470
+    },
+    {
+      "epoch": 0.3734780140054255,
+      "grad_norm": 0.7280906438827515,
+      "learning_rate": 3.945298264981614e-05,
+      "loss": 0.7749,
+      "step": 1480
+    },
+    {
+      "epoch": 0.3760015141000568,
+      "grad_norm": 0.6853066682815552,
+      "learning_rate": 3.927266295205818e-05,
+      "loss": 0.7927,
+      "step": 1490
+    },
+    {
+      "epoch": 0.37852501419468804,
+      "grad_norm": 0.6501105427742004,
+      "learning_rate": 3.9091234889112815e-05,
+      "loss": 0.7895,
+      "step": 1500
+    },
+    {
+      "epoch": 0.3810485142893193,
+      "grad_norm": 0.649364173412323,
+      "learning_rate": 3.8908712550050154e-05,
+      "loss": 0.7952,
+      "step": 1510
+    },
+    {
+      "epoch": 0.38357201438395055,
+      "grad_norm": 0.6459169387817383,
+      "learning_rate": 3.8725110108917975e-05,
+      "loss": 0.7585,
+      "step": 1520
+    },
+    {
+      "epoch": 0.3860955144785818,
+      "grad_norm": 0.6606280207633972,
+      "learning_rate": 3.854044182364098e-05,
+      "loss": 0.7851,
+      "step": 1530
+    },
+    {
+      "epoch": 0.3886190145732131,
+      "grad_norm": 0.6776377558708191,
+      "learning_rate": 3.835472203491367e-05,
+      "loss": 0.7768,
+      "step": 1540
+    },
+    {
+      "epoch": 0.3911425146678443,
+      "grad_norm": 0.6148844361305237,
+      "learning_rate": 3.816796516508658e-05,
+      "loss": 0.777,
+      "step": 1550
+    },
+    {
+      "epoch": 0.39366601476247554,
+      "grad_norm": 0.5975064039230347,
+      "learning_rate": 3.798018571704638e-05,
+      "loss": 0.7677,
+      "step": 1560
+    },
+    {
+      "epoch": 0.3961895148571068,
+      "grad_norm": 0.646436333656311,
+      "learning_rate": 3.779139827308956e-05,
+      "loss": 0.8021,
+      "step": 1570
+    },
+    {
+      "epoch": 0.39871301495173805,
+      "grad_norm": 0.7195472121238708,
+      "learning_rate": 3.760161749379008e-05,
+      "loss": 0.7915,
+      "step": 1580
+    },
+    {
+      "epoch": 0.4012365150463693,
+      "grad_norm": 0.647221565246582,
+      "learning_rate": 3.7410858116860836e-05,
+      "loss": 0.7696,
+      "step": 1590
+    },
+    {
+      "epoch": 0.40376001514100057,
+      "grad_norm": 0.6950120329856873,
+      "learning_rate": 3.721913495600923e-05,
+      "loss": 0.7946,
+      "step": 1600
+    },
+    {
+      "epoch": 0.40628351523563183,
+      "grad_norm": 0.5791555047035217,
+      "learning_rate": 3.7026462899786726e-05,
+      "loss": 0.7469,
+      "step": 1610
+    },
+    {
+      "epoch": 0.4088070153302631,
+      "grad_norm": 0.6242396235466003,
+      "learning_rate": 3.683285691043272e-05,
+      "loss": 0.7894,
+      "step": 1620
+    },
+    {
+      "epoch": 0.41133051542489435,
+      "grad_norm": 0.6625512838363647,
+      "learning_rate": 3.663833202271257e-05,
+      "loss": 0.7977,
+      "step": 1630
+    },
+    {
+      "epoch": 0.4138540155195256,
+      "grad_norm": 0.653052806854248,
+      "learning_rate": 3.6442903342750084e-05,
+      "loss": 0.7394,
+      "step": 1640
+    },
+    {
+      "epoch": 0.4163775156141568,
+      "grad_norm": 0.6416037678718567,
+      "learning_rate": 3.624658604685443e-05,
+      "loss": 0.7624,
+      "step": 1650
+    },
+    {
+      "epoch": 0.41890101570878807,
+      "grad_norm": 0.7198112607002258,
+      "learning_rate": 3.604939538034158e-05,
+      "loss": 0.7779,
+      "step": 1660
+    },
+    {
+      "epoch": 0.42142451580341933,
+      "grad_norm": 0.6925454139709473,
+      "learning_rate": 3.585134665635041e-05,
+      "loss": 0.7746,
+      "step": 1670
+    },
+    {
+      "epoch": 0.4239480158980506,
+      "grad_norm": 0.6133943200111389,
+      "learning_rate": 3.565245525465355e-05,
+      "loss": 0.8115,
+      "step": 1680
+    },
+    {
+      "epoch": 0.42647151599268185,
+      "grad_norm": 0.627083957195282,
+      "learning_rate": 3.5452736620463064e-05,
+      "loss": 0.7745,
+      "step": 1690
+    },
+    {
+      "epoch": 0.4289950160873131,
+      "grad_norm": 0.7416812777519226,
+      "learning_rate": 3.525220626323097e-05,
+      "loss": 0.7565,
+      "step": 1700
+    },
+    {
+      "epoch": 0.43151851618194437,
+      "grad_norm": 0.6600239276885986,
+      "learning_rate": 3.5050879755444877e-05,
+      "loss": 0.8086,
+      "step": 1710
+    },
+    {
+      "epoch": 0.4340420162765756,
+      "grad_norm": 0.6868900060653687,
+      "learning_rate": 3.484877273141866e-05,
+      "loss": 0.7782,
+      "step": 1720
+    },
+    {
+      "epoch": 0.4365655163712069,
+      "grad_norm": 0.6533142328262329,
+      "learning_rate": 3.464590088607839e-05,
+      "loss": 0.7917,
+      "step": 1730
+    },
+    {
+      "epoch": 0.43908901646583814,
+      "grad_norm": 0.7071284055709839,
+      "learning_rate": 3.444227997374345e-05,
+      "loss": 0.7987,
+      "step": 1740
+    },
+    {
+      "epoch": 0.44161251656046935,
+      "grad_norm": 0.7069833874702454,
+      "learning_rate": 3.4237925806903184e-05,
+      "loss": 0.782,
+      "step": 1750
+    },
+    {
+      "epoch": 0.4441360166551006,
+      "grad_norm": 0.5870257616043091,
+      "learning_rate": 3.403285425498889e-05,
+      "loss": 0.7802,
+      "step": 1760
+    },
+    {
+      "epoch": 0.44665951674973187,
+      "grad_norm": 0.7443609237670898,
+      "learning_rate": 3.3827081243141534e-05,
+      "loss": 0.751,
+      "step": 1770
+    },
+    {
+      "epoch": 0.4491830168443631,
+      "grad_norm": 0.64288729429245,
+      "learning_rate": 3.362062275097496e-05,
+      "loss": 0.8028,
+      "step": 1780
+    },
+    {
+      "epoch": 0.4517065169389944,
+      "grad_norm": 0.7132259607315063,
+      "learning_rate": 3.341349481133507e-05,
+      "loss": 0.7483,
+      "step": 1790
+    },
+    {
+      "epoch": 0.45423001703362564,
+      "grad_norm": 0.6608093976974487,
+      "learning_rate": 3.320571350905466e-05,
+      "loss": 0.7861,
+      "step": 1800
+    },
+    {
+      "epoch": 0.4567535171282569,
+      "grad_norm": 0.6749939322471619,
+      "learning_rate": 3.299729497970444e-05,
+      "loss": 0.8,
+      "step": 1810
+    },
+    {
+      "epoch": 0.45927701722288816,
+      "grad_norm": 0.6308214664459229,
+      "learning_rate": 3.278825540833995e-05,
+      "loss": 0.7682,
+      "step": 1820
+    },
+    {
+      "epoch": 0.4618005173175194,
+      "grad_norm": 0.6652118563652039,
+      "learning_rate": 3.2578611028244656e-05,
+      "loss": 0.7581,
+      "step": 1830
+    },
+    {
+      "epoch": 0.4643240174121507,
+      "grad_norm": 0.6874011158943176,
+      "learning_rate": 3.2368378119669363e-05,
+      "loss": 0.7395,
+      "step": 1840
+    },
+    {
+      "epoch": 0.4668475175067819,
+      "grad_norm": 0.580640971660614,
+      "learning_rate": 3.215757300856796e-05,
+      "loss": 0.7635,
+      "step": 1850
+    },
+    {
+      "epoch": 0.46937101760141314,
+      "grad_norm": 0.6897201538085938,
+      "learning_rate": 3.194621206532957e-05,
+      "loss": 0.7999,
+      "step": 1860
+    },
+    {
+      "epoch": 0.4718945176960444,
+      "grad_norm": 0.6152743697166443,
+      "learning_rate": 3.173431170350732e-05,
+      "loss": 0.7652,
+      "step": 1870
+    },
+    {
+      "epoch": 0.47441801779067566,
+      "grad_norm": 0.7078354358673096,
+      "learning_rate": 3.152188837854369e-05,
+      "loss": 0.7985,
+      "step": 1880
+    },
+    {
+      "epoch": 0.4769415178853069,
+      "grad_norm": 0.6471546292304993,
+      "learning_rate": 3.130895858649264e-05,
+      "loss": 0.7502,
+      "step": 1890
+    },
+    {
+      "epoch": 0.4794650179799382,
+      "grad_norm": 0.7108110189437866,
+      "learning_rate": 3.109553886273863e-05,
+      "loss": 0.7777,
+      "step": 1900
+    },
+    {
+      "epoch": 0.48198851807456944,
+      "grad_norm": 0.6540038585662842,
+      "learning_rate": 3.088164578071246e-05,
+      "loss": 0.7585,
+      "step": 1910
+    },
+    {
+      "epoch": 0.4845120181692007,
+      "grad_norm": 0.6966577172279358,
+      "learning_rate": 3.066729595060431e-05,
+      "loss": 0.7459,
+      "step": 1920
+    },
+    {
+      "epoch": 0.48703551826383196,
+      "grad_norm": 0.7039586901664734,
+      "learning_rate": 3.0452506018073833e-05,
+      "loss": 0.7829,
+      "step": 1930
+    },
+    {
+      "epoch": 0.4895590183584632,
+      "grad_norm": 0.7935706973075867,
+      "learning_rate": 3.0237292662957473e-05,
+      "loss": 0.7635,
+      "step": 1940
+    },
+    {
+      "epoch": 0.4920825184530944,
+      "grad_norm": 0.6353578567504883,
+      "learning_rate": 3.0021672597973207e-05,
+      "loss": 0.7573,
+      "step": 1950
+    },
+    {
+      "epoch": 0.4946060185477257,
+      "grad_norm": 0.6605427265167236,
+      "learning_rate": 2.9805662567422676e-05,
+      "loss": 0.7915,
+      "step": 1960
+    },
+    {
+      "epoch": 0.49712951864235694,
+      "grad_norm": 0.6967743635177612,
+      "learning_rate": 2.9589279345890895e-05,
+      "loss": 0.7787,
+      "step": 1970
+    },
+    {
+      "epoch": 0.4996530187369882,
+      "grad_norm": 0.566273033618927,
+      "learning_rate": 2.9372539736943577e-05,
+      "loss": 0.7719,
+      "step": 1980
+    },
+    {
+      "epoch": 0.5021765188316195,
+      "grad_norm": 0.596185028553009,
+      "learning_rate": 2.9155460571822245e-05,
+      "loss": 0.759,
+      "step": 1990
+    },
+    {
+      "epoch": 0.5047000189262507,
+      "grad_norm": 0.6854886412620544,
+      "learning_rate": 2.893805870813717e-05,
+      "loss": 0.7461,
+      "step": 2000
+    },
+    {
+      "epoch": 0.5072235190208819,
+      "grad_norm": 0.6499455571174622,
+      "learning_rate": 2.872035102855826e-05,
+      "loss": 0.7586,
+      "step": 2010
+    },
+    {
+      "epoch": 0.5097470191155132,
+      "grad_norm": 0.6669695377349854,
+      "learning_rate": 2.850235443950402e-05,
+      "loss": 0.7319,
+      "step": 2020
+    },
+    {
+      "epoch": 0.5122705192101444,
+      "grad_norm": 0.6255702972412109,
+      "learning_rate": 2.8284085869828665e-05,
+      "loss": 0.7687,
+      "step": 2030
+    },
+    {
+      "epoch": 0.5147940193047758,
+      "grad_norm": 0.645270824432373,
+      "learning_rate": 2.8065562269507463e-05,
+      "loss": 0.7764,
+      "step": 2040
+    },
+    {
+      "epoch": 0.517317519399407,
+      "grad_norm": 0.7418912053108215,
+      "learning_rate": 2.7846800608320485e-05,
+      "loss": 0.7571,
+      "step": 2050
+    },
+    {
+      "epoch": 0.5198410194940383,
+      "grad_norm": 0.6680746674537659,
+      "learning_rate": 2.7627817874534762e-05,
+      "loss": 0.7489,
+      "step": 2060
+    },
+    {
+      "epoch": 0.5223645195886695,
+      "grad_norm": 0.6707837581634521,
+      "learning_rate": 2.7408631073585068e-05,
+      "loss": 0.7345,
+      "step": 2070
+    },
+    {
+      "epoch": 0.5248880196833008,
+      "grad_norm": 0.664960503578186,
+      "learning_rate": 2.7189257226753305e-05,
+      "loss": 0.7602,
+      "step": 2080
+    },
+    {
+      "epoch": 0.527411519777932,
+      "grad_norm": 0.618732213973999,
+      "learning_rate": 2.696971336984672e-05,
+      "loss": 0.7831,
+      "step": 2090
+    },
+    {
+      "epoch": 0.5299350198725632,
+      "grad_norm": 0.7285112142562866,
+      "learning_rate": 2.6750016551874945e-05,
+      "loss": 0.7575,
+      "step": 2100
+    },
+    {
+      "epoch": 0.5324585199671945,
+      "grad_norm": 0.6303609013557434,
+      "learning_rate": 2.6530183833726025e-05,
+      "loss": 0.7567,
+      "step": 2110
+    },
+    {
+      "epoch": 0.5349820200618257,
+      "grad_norm": 0.7151260375976562,
+      "learning_rate": 2.6310232286841546e-05,
+      "loss": 0.7876,
+      "step": 2120
+    },
+    {
+      "epoch": 0.537505520156457,
+      "grad_norm": 0.6547929048538208,
+      "learning_rate": 2.609017899189092e-05,
+      "loss": 0.7723,
+      "step": 2130
+    },
+    {
+      "epoch": 0.5400290202510882,
+      "grad_norm": 0.6668947339057922,
+      "learning_rate": 2.587004103744495e-05,
+      "loss": 0.7261,
+      "step": 2140
+    },
+    {
+      "epoch": 0.5425525203457195,
+      "grad_norm": 0.5897696614265442,
+      "learning_rate": 2.564983551864882e-05,
+      "loss": 0.7386,
+      "step": 2150
+    },
+    {
+      "epoch": 0.5450760204403508,
+      "grad_norm": 0.7253137826919556,
+      "learning_rate": 2.54295795358945e-05,
+      "loss": 0.762,
+      "step": 2160
+    },
+    {
+      "epoch": 0.5475995205349821,
+      "grad_norm": 0.6665747761726379,
+      "learning_rate": 2.5209290193492834e-05,
+      "loss": 0.7558,
+      "step": 2170
+    },
+    {
+      "epoch": 0.5501230206296133,
+      "grad_norm": 0.6269178986549377,
+      "learning_rate": 2.4988984598345247e-05,
+      "loss": 0.7248,
+      "step": 2180
+    },
+    {
+      "epoch": 0.5526465207242445,
+      "grad_norm": 0.6443243622779846,
+      "learning_rate": 2.4768679858615304e-05,
+      "loss": 0.7699,
+      "step": 2190
+    },
+    {
+      "epoch": 0.5551700208188758,
+      "grad_norm": 0.7323073744773865,
+      "learning_rate": 2.454839308240014e-05,
+      "loss": 0.796,
+      "step": 2200
+    },
+    {
+      "epoch": 0.557693520913507,
+      "grad_norm": 0.6538411974906921,
+      "learning_rate": 2.4328141376401903e-05,
+      "loss": 0.7521,
+      "step": 2210
+    },
+    {
+      "epoch": 0.5602170210081383,
+      "grad_norm": 0.638306200504303,
+      "learning_rate": 2.4107941844599312e-05,
+      "loss": 0.736,
+      "step": 2220
+    },
+    {
+      "epoch": 0.5627405211027695,
+      "grad_norm": 0.6431130170822144,
+      "learning_rate": 2.3887811586919424e-05,
+      "loss": 0.7688,
+      "step": 2230
+    },
+    {
+      "epoch": 0.5652640211974008,
+      "grad_norm": 0.6741809844970703,
+      "learning_rate": 2.3667767697909694e-05,
+      "loss": 0.7445,
+      "step": 2240
+    },
+    {
+      "epoch": 0.567787521292032,
+      "grad_norm": 0.6638673543930054,
+      "learning_rate": 2.3447827265410517e-05,
+      "loss": 0.7898,
+      "step": 2250
+    },
+    {
+      "epoch": 0.5703110213866633,
+      "grad_norm": 0.708502471446991,
+      "learning_rate": 2.3228007369228178e-05,
+      "loss": 0.8121,
+      "step": 2260
+    },
+    {
+      "epoch": 0.5728345214812945,
+      "grad_norm": 0.6407279372215271,
+      "learning_rate": 2.3008325079808576e-05,
+      "loss": 0.7682,
+      "step": 2270
+    },
+    {
+      "epoch": 0.5753580215759259,
+      "grad_norm": 0.65432208776474,
+      "learning_rate": 2.2788797456911503e-05,
+      "loss": 0.7305,
+      "step": 2280
+    },
+    {
+      "epoch": 0.5778815216705571,
+      "grad_norm": 0.7159484624862671,
+      "learning_rate": 2.2569441548285934e-05,
+      "loss": 0.7618,
+      "step": 2290
+    },
+    {
+      "epoch": 0.5804050217651883,
+      "grad_norm": 0.5184557437896729,
+      "learning_rate": 2.2350274388346064e-05,
+      "loss": 0.7192,
+      "step": 2300
+    },
+    {
+      "epoch": 0.5829285218598196,
+      "grad_norm": 0.6150662302970886,
+      "learning_rate": 2.213131299684858e-05,
+      "loss": 0.7764,
+      "step": 2310
+    },
+    {
+      "epoch": 0.5854520219544508,
+      "grad_norm": 0.708900511264801,
+      "learning_rate": 2.191257437757086e-05,
+      "loss": 0.7408,
+      "step": 2320
+    },
+    {
+      "epoch": 0.5879755220490821,
+      "grad_norm": 0.5827603340148926,
+      "learning_rate": 2.16940755169906e-05,
+      "loss": 0.7701,
+      "step": 2330
+    },
+    {
+      "epoch": 0.5904990221437133,
+      "grad_norm": 0.7515769600868225,
+      "learning_rate": 2.1475833382966647e-05,
+      "loss": 0.805,
+      "step": 2340
+    },
+    {
+      "epoch": 0.5930225222383446,
+      "grad_norm": 0.6381510496139526,
+      "learning_rate": 2.1257864923421404e-05,
+      "loss": 0.7889,
+      "step": 2350
+    },
+    {
+      "epoch": 0.5955460223329758,
+      "grad_norm": 0.6384143829345703,
+      "learning_rate": 2.1040187065024605e-05,
+      "loss": 0.7111,
+      "step": 2360
+    },
+    {
+      "epoch": 0.5980695224276071,
+      "grad_norm": 0.6297293305397034,
+      "learning_rate": 2.0822816711878978e-05,
+      "loss": 0.7759,
+      "step": 2370
+    },
+    {
+      "epoch": 0.6005930225222383,
+      "grad_norm": 0.6517510414123535,
+      "learning_rate": 2.0605770744207413e-05,
+      "loss": 0.7514,
+      "step": 2380
+    },
+    {
+      "epoch": 0.6031165226168695,
+      "grad_norm": 0.6666356325149536,
+      "learning_rate": 2.0389066017042192e-05,
+      "loss": 0.7308,
+      "step": 2390
+    },
+    {
+      "epoch": 0.6056400227115009,
+      "grad_norm": 0.8132819533348083,
+      "learning_rate": 2.0172719358916042e-05,
+      "loss": 0.7363,
+      "step": 2400
+    },
+    {
+      "epoch": 0.6081635228061321,
+      "grad_norm": 0.7209652662277222,
+      "learning_rate": 1.9956747570555288e-05,
+      "loss": 0.7838,
+      "step": 2410
+    },
+    {
+      "epoch": 0.6106870229007634,
+      "grad_norm": 0.6177300214767456,
+      "learning_rate": 1.9741167423575186e-05,
+      "loss": 0.7153,
+      "step": 2420
+    },
+    {
+      "epoch": 0.6132105229953946,
+      "grad_norm": 0.6164495944976807,
+      "learning_rate": 1.9525995659177484e-05,
+      "loss": 0.7502,
+      "step": 2430
+    },
+    {
+      "epoch": 0.6157340230900259,
+      "grad_norm": 0.6001858115196228,
+      "learning_rate": 1.9311248986850365e-05,
+      "loss": 0.7396,
+      "step": 2440
+    },
+    {
+      "epoch": 0.6182575231846571,
+      "grad_norm": 0.661919116973877,
+      "learning_rate": 1.9096944083070866e-05,
+      "loss": 0.7557,
+      "step": 2450
+    },
+    {
+      "epoch": 0.6207810232792884,
+      "grad_norm": 0.6734655499458313,
+      "learning_rate": 1.8883097590009775e-05,
+      "loss": 0.7659,
+      "step": 2460
+    },
+    {
+      "epoch": 0.6233045233739196,
+      "grad_norm": 0.6683171391487122,
+      "learning_rate": 1.866972611423936e-05,
+      "loss": 0.7464,
+      "step": 2470
+    },
+    {
+      "epoch": 0.6258280234685509,
+      "grad_norm": 0.7079278826713562,
+      "learning_rate": 1.8456846225443648e-05,
+      "loss": 0.7051,
+      "step": 2480
+    },
+    {
+      "epoch": 0.6283515235631821,
+      "grad_norm": 0.6441357731819153,
+      "learning_rate": 1.8244474455131792e-05,
+      "loss": 0.7441,
+      "step": 2490
+    },
+    {
+      "epoch": 0.6308750236578133,
+      "grad_norm": 0.7199136018753052,
+      "learning_rate": 1.8032627295354183e-05,
+      "loss": 0.7419,
+      "step": 2500
+    },
+    {
+      "epoch": 0.6333985237524447,
+      "grad_norm": 0.7727274894714355,
+      "learning_rate": 1.7821321197421837e-05,
+      "loss": 0.785,
+      "step": 2510
+    },
+    {
+      "epoch": 0.6359220238470759,
+      "grad_norm": 0.6691886782646179,
+      "learning_rate": 1.761057257062876e-05,
+      "loss": 0.7576,
+      "step": 2520
+    },
+    {
+      "epoch": 0.6384455239417072,
+      "grad_norm": 0.5676743388175964,
+      "learning_rate": 1.740039778097772e-05,
+      "loss": 0.7151,
+      "step": 2530
+    },
+    {
+      "epoch": 0.6409690240363384,
+      "grad_norm": 0.6777900457382202,
+      "learning_rate": 1.7190813149909274e-05,
+      "loss": 0.7477,
+      "step": 2540
+    },
+    {
+      "epoch": 0.6434925241309697,
+      "grad_norm": 0.691109836101532,
+      "learning_rate": 1.6981834953034344e-05,
+      "loss": 0.7536,
+      "step": 2550
+    },
+    {
+      "epoch": 0.6460160242256009,
+      "grad_norm": 0.7271637916564941,
+      "learning_rate": 1.677347941887028e-05,
+      "loss": 0.7181,
+      "step": 2560
+    },
+    {
+      "epoch": 0.6485395243202322,
+      "grad_norm": 0.693095326423645,
+      "learning_rate": 1.656576272758061e-05,
+      "loss": 0.7703,
+      "step": 2570
+    },
+    {
+      "epoch": 0.6510630244148634,
+      "grad_norm": 0.6617145538330078,
+      "learning_rate": 1.6358701009718577e-05,
+      "loss": 0.7505,
+      "step": 2580
+    },
+    {
+      "epoch": 0.6535865245094946,
+      "grad_norm": 0.5836735963821411,
+      "learning_rate": 1.615231034497444e-05,
+      "loss": 0.7629,
+      "step": 2590
+    },
+    {
+      "epoch": 0.6561100246041259,
+      "grad_norm": 0.767706036567688,
+      "learning_rate": 1.5946606760926865e-05,
+      "loss": 0.7311,
+      "step": 2600
+    },
+    {
+      "epoch": 0.6586335246987571,
+      "grad_norm": 1.1599899530410767,
+      "learning_rate": 1.574160623179816e-05,
+      "loss": 0.7538,
+      "step": 2610
+    },
+    {
+      "epoch": 0.6611570247933884,
+      "grad_norm": 0.6588570475578308,
+      "learning_rate": 1.553732467721392e-05,
+      "loss": 0.7181,
+      "step": 2620
+    },
+    {
+      "epoch": 0.6636805248880197,
+      "grad_norm": 0.5837569832801819,
+      "learning_rate": 1.5333777960966616e-05,
+      "loss": 0.7218,
+      "step": 2630
+    },
+    {
+      "epoch": 0.666204024982651,
+      "grad_norm": 0.5983703136444092,
+      "learning_rate": 1.5130981889783795e-05,
+      "loss": 0.728,
+      "step": 2640
+    },
+    {
+      "epoch": 0.6687275250772822,
+      "grad_norm": 0.6551673412322998,
+      "learning_rate": 1.4928952212100483e-05,
+      "loss": 0.7561,
+      "step": 2650
+    },
+    {
+      "epoch": 0.6712510251719135,
+      "grad_norm": 0.5981004238128662,
+      "learning_rate": 1.4727704616836296e-05,
+      "loss": 0.7553,
+      "step": 2660
+    },
+    {
+      "epoch": 0.6737745252665447,
+      "grad_norm": 0.6593780517578125,
+      "learning_rate": 1.4527254732177043e-05,
+      "loss": 0.7428,
+      "step": 2670
+    },
+    {
+      "epoch": 0.676298025361176,
+      "grad_norm": 0.6018021702766418,
+      "learning_rate": 1.4327618124361114e-05,
+      "loss": 0.7489,
+      "step": 2680
+    },
+    {
+      "epoch": 0.6788215254558072,
+      "grad_norm": 0.629327118396759,
+      "learning_rate": 1.412881029647065e-05,
+      "loss": 0.7199,
+      "step": 2690
+    },
+    {
+      "epoch": 0.6813450255504384,
+      "grad_norm": 0.616880476474762,
+      "learning_rate": 1.3930846687227664e-05,
+      "loss": 0.7236,
+      "step": 2700
+    },
+    {
+      "epoch": 0.6838685256450697,
+      "grad_norm": 0.6667315363883972,
+      "learning_rate": 1.3733742669795049e-05,
+      "loss": 0.7679,
+      "step": 2710
+    },
+    {
+      "epoch": 0.6863920257397009,
+      "grad_norm": 0.7159212231636047,
+      "learning_rate": 1.3537513550582853e-05,
+      "loss": 0.7577,
+      "step": 2720
+    },
+    {
+      "epoch": 0.6889155258343322,
+      "grad_norm": 0.5977271199226379,
+      "learning_rate": 1.3342174568059527e-05,
+      "loss": 0.7289,
+      "step": 2730
+    },
+    {
+      "epoch": 0.6914390259289634,
+      "grad_norm": 0.6134091019630432,
+      "learning_rate": 1.3147740891568661e-05,
+      "loss": 0.7159,
+      "step": 2740
+    },
+    {
+      "epoch": 0.6939625260235948,
+      "grad_norm": 0.6235146522521973,
+      "learning_rate": 1.2954227620150904e-05,
+      "loss": 0.7587,
+      "step": 2750
+    },
+    {
+      "epoch": 0.696486026118226,
+      "grad_norm": 0.6821407079696655,
+      "learning_rate": 1.2761649781371479e-05,
+      "loss": 0.7366,
+      "step": 2760
+    },
+    {
+      "epoch": 0.6990095262128573,
+      "grad_norm": 0.6241364479064941,
+      "learning_rate": 1.257002233015318e-05,
+      "loss": 0.7587,
+      "step": 2770
+    },
+    {
+      "epoch": 0.7015330263074885,
+      "grad_norm": 0.6628735065460205,
+      "learning_rate": 1.2379360147614994e-05,
+      "loss": 0.7289,
+      "step": 2780
+    },
+    {
+      "epoch": 0.7040565264021197,
+      "grad_norm": 0.6374642252922058,
+      "learning_rate": 1.2189678039916532e-05,
+      "loss": 0.7036,
+      "step": 2790
+    },
+    {
+      "epoch": 0.706580026496751,
+      "grad_norm": 0.5973037481307983,
+      "learning_rate": 1.2000990737108225e-05,
+      "loss": 0.723,
+      "step": 2800
+    },
+    {
+      "epoch": 0.7091035265913822,
+      "grad_norm": 0.7454132437705994,
+      "learning_rate": 1.1813312891987392e-05,
+      "loss": 0.7631,
+      "step": 2810
+    },
+    {
+      "epoch": 0.7116270266860135,
+      "grad_norm": 0.6454845070838928,
+      "learning_rate": 1.1626659078960424e-05,
+      "loss": 0.7357,
+      "step": 2820
+    },
+    {
+      "epoch": 0.7141505267806447,
+      "grad_norm": 0.616027295589447,
+      "learning_rate": 1.1441043792910936e-05,
+      "loss": 0.7326,
+      "step": 2830
+    },
+    {
+      "epoch": 0.716674026875276,
+      "grad_norm": 0.7227725386619568,
+      "learning_rate": 1.1256481448074179e-05,
+      "loss": 0.7293,
+      "step": 2840
+    },
+    {
+      "epoch": 0.7191975269699072,
+      "grad_norm": 0.765404999256134,
+      "learning_rate": 1.1072986376917638e-05,
+      "loss": 0.7395,
+      "step": 2850
+    },
+    {
+      "epoch": 0.7217210270645386,
+      "grad_norm": 0.6478826403617859,
+      "learning_rate": 1.0890572829028087e-05,
+      "loss": 0.7734,
+      "step": 2860
+    },
+    {
+      "epoch": 0.7242445271591698,
+      "grad_norm": 0.6000937819480896,
+      "learning_rate": 1.0709254970004937e-05,
+      "loss": 0.7247,
+      "step": 2870
+    },
+    {
+      "epoch": 0.7267680272538011,
+      "grad_norm": 0.5255608558654785,
+      "learning_rate": 1.0529046880360263e-05,
+      "loss": 0.7219,
+      "step": 2880
+    },
+    {
+      "epoch": 0.7292915273484323,
+      "grad_norm": 0.5780526995658875,
+      "learning_rate": 1.034996255442529e-05,
+      "loss": 0.7658,
+      "step": 2890
+    },
+    {
+      "epoch": 0.7318150274430635,
+      "grad_norm": 0.5964454412460327,
+      "learning_rate": 1.0172015899263712e-05,
+      "loss": 0.7363,
+      "step": 2900
+    },
+    {
+      "epoch": 0.7343385275376948,
+      "grad_norm": 0.6541391015052795,
+      "learning_rate": 9.995220733591639e-06,
+      "loss": 0.7214,
+      "step": 2910
+    },
+    {
+      "epoch": 0.736862027632326,
+      "grad_norm": 0.572470486164093,
+      "learning_rate": 9.819590786704572e-06,
+      "loss": 0.7559,
+      "step": 2920
+    },
+    {
+      "epoch": 0.7393855277269573,
+      "grad_norm": 0.6316998600959778,
+      "learning_rate": 9.645139697411149e-06,
+      "loss": 0.7289,
+      "step": 2930
+    },
+    {
+      "epoch": 0.7419090278215885,
+      "grad_norm": 0.6283255815505981,
+      "learning_rate": 9.471881012974071e-06,
+      "loss": 0.754,
+      "step": 2940
+    },
+    {
+      "epoch": 0.7444325279162198,
+      "grad_norm": 0.6679040193557739,
+      "learning_rate": 9.299828188058013e-06,
+      "loss": 0.7417,
+      "step": 2950
+    },
+    {
+      "epoch": 0.746956028010851,
+      "grad_norm": 0.6345584988594055,
+      "learning_rate": 9.128994583684838e-06,
+      "loss": 0.7222,
+      "step": 2960
+    },
+    {
+      "epoch": 0.7494795281054824,
+      "grad_norm": 0.6412184238433838,
+      "learning_rate": 8.959393466195972e-06,
+      "loss": 0.7713,
+      "step": 2970
+    },
+    {
+      "epoch": 0.7520030282001136,
+      "grad_norm": 0.6120537519454956,
+      "learning_rate": 8.791038006222233e-06,
+      "loss": 0.7494,
+      "step": 2980
+    },
+    {
+      "epoch": 0.7545265282947448,
+      "grad_norm": 0.6384350657463074,
+      "learning_rate": 8.623941277660994e-06,
+      "loss": 0.7241,
+      "step": 2990
+    },
+    {
+      "epoch": 0.7570500283893761,
+      "grad_norm": 0.6902556419372559,
+      "learning_rate": 8.458116256660981e-06,
+      "loss": 0.7677,
+      "step": 3000
+    },
+    {
+      "epoch": 0.7595735284840073,
+      "grad_norm": 0.6172592639923096,
+      "learning_rate": 8.293575820614508e-06,
+      "loss": 0.741,
+      "step": 3010
+    },
+    {
+      "epoch": 0.7620970285786386,
+      "grad_norm": 0.6649114489555359,
+      "learning_rate": 8.130332747157542e-06,
+      "loss": 0.6986,
+      "step": 3020
+    },
+    {
+      "epoch": 0.7646205286732698,
+      "grad_norm": 0.8120855689048767,
+      "learning_rate": 7.968399713177366e-06,
+      "loss": 0.7496,
+      "step": 3030
+    },
+    {
+      "epoch": 0.7671440287679011,
+      "grad_norm": 0.6434275507926941,
+      "learning_rate": 7.807789293828204e-06,
+      "loss": 0.7617,
+      "step": 3040
+    },
+    {
+      "epoch": 0.7696675288625323,
+      "grad_norm": 0.6495864987373352,
+      "learning_rate": 7.648513961554607e-06,
+      "loss": 0.6993,
+      "step": 3050
+    },
+    {
+      "epoch": 0.7721910289571636,
+      "grad_norm": 0.5868855714797974,
+      "learning_rate": 7.4905860851229605e-06,
+      "loss": 0.7572,
+      "step": 3060
+    },
+    {
+      "epoch": 0.7747145290517948,
+      "grad_norm": 0.689618706703186,
+      "learning_rate": 7.334017928660902e-06,
+      "loss": 0.7454,
+      "step": 3070
+    },
+    {
+      "epoch": 0.7772380291464261,
+      "grad_norm": 0.6041099429130554,
+      "learning_rate": 7.1788216507049865e-06,
+      "loss": 0.7309,
+      "step": 3080
+    },
+    {
+      "epoch": 0.7797615292410573,
+      "grad_norm": 0.6858187317848206,
+      "learning_rate": 7.0250093032564494e-06,
+      "loss": 0.6921,
+      "step": 3090
+    },
+    {
+      "epoch": 0.7822850293356886,
+      "grad_norm": 0.625678539276123,
+      "learning_rate": 6.872592830845339e-06,
+      "loss": 0.7257,
+      "step": 3100
+    },
+    {
+      "epoch": 0.7848085294303199,
+      "grad_norm": 0.574103593826294,
+      "learning_rate": 6.72158406960289e-06,
+      "loss": 0.7532,
+      "step": 3110
+    },
+    {
+      "epoch": 0.7873320295249511,
+      "grad_norm": 0.5581086874008179,
+      "learning_rate": 6.571994746342439e-06,
+      "loss": 0.7354,
+      "step": 3120
+    },
+    {
+      "epoch": 0.7898555296195824,
+      "grad_norm": 0.6243663430213928,
+      "learning_rate": 6.4238364776486785e-06,
+      "loss": 0.7154,
+      "step": 3130
+    },
+    {
+      "epoch": 0.7923790297142136,
+      "grad_norm": 0.6376320719718933,
+      "learning_rate": 6.277120768975644e-06,
+      "loss": 0.7647,
+      "step": 3140
+    },
+    {
+      "epoch": 0.7949025298088449,
+      "grad_norm": 0.7120991945266724,
+      "learning_rate": 6.131859013753155e-06,
+      "loss": 0.7199,
+      "step": 3150
+    },
+    {
+      "epoch": 0.7974260299034761,
+      "grad_norm": 0.6970154643058777,
+      "learning_rate": 5.988062492502117e-06,
+      "loss": 0.7067,
+      "step": 3160
+    },
+    {
+      "epoch": 0.7999495299981074,
+      "grad_norm": 0.5977396368980408,
+      "learning_rate": 5.8457423719584435e-06,
+      "loss": 0.7244,
+      "step": 3170
+    },
+    {
+      "epoch": 0.8024730300927386,
+      "grad_norm": 0.6535403728485107,
+      "learning_rate": 5.704909704205949e-06,
+      "loss": 0.7529,
+      "step": 3180
+    },
+    {
+      "epoch": 0.8049965301873698,
+      "grad_norm": 0.6335867643356323,
+      "learning_rate": 5.565575425818054e-06,
+      "loss": 0.7417,
+      "step": 3190
+    },
+    {
+      "epoch": 0.8075200302820011,
+      "grad_norm": 0.6783677339553833,
+      "learning_rate": 5.427750357008468e-06,
+      "loss": 0.7037,
+      "step": 3200
+    },
+    {
+      "epoch": 0.8100435303766323,
+      "grad_norm": 0.6452818512916565,
+      "learning_rate": 5.291445200790982e-06,
+      "loss": 0.7491,
+      "step": 3210
+    },
+    {
+      "epoch": 0.8125670304712637,
+      "grad_norm": 0.6002216935157776,
+      "learning_rate": 5.156670542148267e-06,
+      "loss": 0.7501,
+      "step": 3220
+    },
+    {
+      "epoch": 0.8150905305658949,
+      "grad_norm": 0.7028263211250305,
+      "learning_rate": 5.023436847209887e-06,
+      "loss": 0.741,
+      "step": 3230
+    },
+    {
+      "epoch": 0.8176140306605262,
+      "grad_norm": 0.6382498145103455,
+      "learning_rate": 4.891754462439557e-06,
+      "loss": 0.7066,
+      "step": 3240
+    },
+    {
+      "epoch": 0.8201375307551574,
+      "grad_norm": 0.6723865270614624,
+      "learning_rate": 4.761633613831645e-06,
+      "loss": 0.7426,
+      "step": 3250
+    },
+    {
+      "epoch": 0.8226610308497887,
+      "grad_norm": 0.7076628804206848,
+      "learning_rate": 4.6330844061170914e-06,
+      "loss": 0.7148,
+      "step": 3260
+    },
+    {
+      "epoch": 0.8251845309444199,
+      "grad_norm": 0.6604047417640686,
+      "learning_rate": 4.506116821978662e-06,
+      "loss": 0.7268,
+      "step": 3270
+    },
+    {
+      "epoch": 0.8277080310390512,
+      "grad_norm": 0.5714926719665527,
+      "learning_rate": 4.380740721275786e-06,
+      "loss": 0.7508,
+      "step": 3280
+    },
+    {
+      "epoch": 0.8302315311336824,
+      "grad_norm": 0.6624537110328674,
+      "learning_rate": 4.25696584027882e-06,
+      "loss": 0.7313,
+      "step": 3290
+    },
+    {
+      "epoch": 0.8327550312283136,
+      "grad_norm": 0.7087505459785461,
+      "learning_rate": 4.134801790913006e-06,
+      "loss": 0.6936,
+      "step": 3300
+    },
+    {
+      "epoch": 0.8352785313229449,
+      "grad_norm": 0.7512862682342529,
+      "learning_rate": 4.014258060012005e-06,
+      "loss": 0.7525,
+      "step": 3310
+    },
+    {
+      "epoch": 0.8378020314175761,
+      "grad_norm": 0.5958043932914734,
+      "learning_rate": 3.895344008581222e-06,
+      "loss": 0.7235,
+      "step": 3320
+    },
+    {
+      "epoch": 0.8403255315122075,
+      "grad_norm": 0.7010710835456848,
+      "learning_rate": 3.7780688710708223e-06,
+      "loss": 0.7122,
+      "step": 3330
+    },
+    {
+      "epoch": 0.8428490316068387,
+      "grad_norm": 0.64909827709198,
+      "learning_rate": 3.6624417546586574e-06,
+      "loss": 0.6846,
+      "step": 3340
+    },
+    {
+      "epoch": 0.84537253170147,
+      "grad_norm": 0.7238036394119263,
+      "learning_rate": 3.548471638542991e-06,
+      "loss": 0.7514,
+      "step": 3350
+    },
+    {
+      "epoch": 0.8478960317961012,
+      "grad_norm": 0.6112589240074158,
+      "learning_rate": 3.436167373245247e-06,
+      "loss": 0.7297,
+      "step": 3360
+    },
+    {
+      "epoch": 0.8504195318907325,
+      "grad_norm": 0.6378879547119141,
+      "learning_rate": 3.325537679922672e-06,
+      "loss": 0.7429,
+      "step": 3370
+    },
+    {
+      "epoch": 0.8529430319853637,
+      "grad_norm": 0.7128148674964905,
+      "learning_rate": 3.2165911496911173e-06,
+      "loss": 0.7168,
+      "step": 3380
+    },
+    {
+      "epoch": 0.8554665320799949,
+      "grad_norm": 0.6448369026184082,
+      "learning_rate": 3.1093362429578414e-06,
+      "loss": 0.7358,
+      "step": 3390
+    },
+    {
+      "epoch": 0.8579900321746262,
+      "grad_norm": 0.5888521075248718,
+      "learning_rate": 3.0037812887645483e-06,
+      "loss": 0.7522,
+      "step": 3400
+    },
+    {
+      "epoch": 0.8605135322692574,
+      "grad_norm": 0.5976009368896484,
+      "learning_rate": 2.8999344841405373e-06,
+      "loss": 0.7222,
+      "step": 3410
+    },
+    {
+      "epoch": 0.8630370323638887,
+      "grad_norm": 0.6030629277229309,
+      "learning_rate": 2.7978038934662024e-06,
+      "loss": 0.7157,
+      "step": 3420
+    },
+    {
+      "epoch": 0.8655605324585199,
+      "grad_norm": 0.6424097418785095,
+      "learning_rate": 2.697397447846725e-06,
+      "loss": 0.7324,
+      "step": 3430
+    },
+    {
+      "epoch": 0.8680840325531513,
+      "grad_norm": 0.6089041829109192,
+      "learning_rate": 2.5987229444962237e-06,
+      "loss": 0.7447,
+      "step": 3440
+    },
+    {
+      "epoch": 0.8706075326477825,
+      "grad_norm": 0.6227433681488037,
+      "learning_rate": 2.501788046132203e-06,
+      "loss": 0.7552,
+      "step": 3450
+    },
+    {
+      "epoch": 0.8731310327424138,
+      "grad_norm": 0.5971252918243408,
+      "learning_rate": 2.4066002803805386e-06,
+      "loss": 0.7276,
+      "step": 3460
+    },
+    {
+      "epoch": 0.875654532837045,
+      "grad_norm": 0.6682915091514587,
+      "learning_rate": 2.313167039190861e-06,
+      "loss": 0.7312,
+      "step": 3470
+    },
+    {
+      "epoch": 0.8781780329316763,
+      "grad_norm": 0.6341772079467773,
+      "learning_rate": 2.2214955782625752e-06,
+      "loss": 0.7303,
+      "step": 3480
+    },
+    {
+      "epoch": 0.8807015330263075,
+      "grad_norm": 0.6950182914733887,
+      "learning_rate": 2.1315930164813507e-06,
+      "loss": 0.757,
+      "step": 3490
+    },
+    {
+      "epoch": 0.8832250331209387,
+      "grad_norm": 0.6251794099807739,
+      "learning_rate": 2.0434663353663536e-06,
+      "loss": 0.7293,
+      "step": 3500
+    },
+    {
+      "epoch": 0.88574853321557,
+      "grad_norm": 0.6550153493881226,
+      "learning_rate": 1.9571223785280314e-06,
+      "loss": 0.7323,
+      "step": 3510
+    },
+    {
+      "epoch": 0.8882720333102012,
+      "grad_norm": 0.6896747350692749,
+      "learning_rate": 1.8725678511367001e-06,
+      "loss": 0.7429,
+      "step": 3520
+    },
+    {
+      "epoch": 0.8907955334048325,
+      "grad_norm": 0.6337852478027344,
+      "learning_rate": 1.789809319401825e-06,
+      "loss": 0.743,
+      "step": 3530
+    },
+    {
+      "epoch": 0.8933190334994637,
+      "grad_norm": 0.6342365145683289,
+      "learning_rate": 1.7088532100621224e-06,
+      "loss": 0.7093,
+      "step": 3540
+    },
+    {
+      "epoch": 0.895842533594095,
+      "grad_norm": 0.6229385137557983,
+      "learning_rate": 1.629705809886467e-06,
+      "loss": 0.7111,
+      "step": 3550
+    },
+    {
+      "epoch": 0.8983660336887263,
+      "grad_norm": 0.5831491947174072,
+      "learning_rate": 1.5523732651857082e-06,
+      "loss": 0.7267,
+      "step": 3560
+    },
+    {
+      "epoch": 0.9008895337833576,
+      "grad_norm": 0.6794403791427612,
+      "learning_rate": 1.4768615813353398e-06,
+      "loss": 0.7302,
+      "step": 3570
+    },
+    {
+      "epoch": 0.9034130338779888,
+      "grad_norm": 0.5871224403381348,
+      "learning_rate": 1.4031766223091603e-06,
+      "loss": 0.7284,
+      "step": 3580
+    },
+    {
+      "epoch": 0.90593653397262,
+      "grad_norm": 0.5899478197097778,
+      "learning_rate": 1.3313241102239054e-06,
+      "loss": 0.7055,
+      "step": 3590
+    },
+    {
+      "epoch": 0.9084600340672513,
+      "grad_norm": 0.625042200088501,
+      "learning_rate": 1.261309624894863e-06,
+      "loss": 0.7455,
+      "step": 3600
+    },
+    {
+      "epoch": 0.9109835341618825,
+      "grad_norm": 0.5997599363327026,
+      "learning_rate": 1.1931386034025882e-06,
+      "loss": 0.73,
+      "step": 3610
+    },
+    {
+      "epoch": 0.9135070342565138,
+      "grad_norm": 0.5365213751792908,
+      "learning_rate": 1.1268163396706583e-06,
+      "loss": 0.74,
+      "step": 3620
+    },
+    {
+      "epoch": 0.916030534351145,
+      "grad_norm": 0.5916723012924194,
+      "learning_rate": 1.0623479840545874e-06,
+      "loss": 0.6867,
+      "step": 3630
+    },
+    {
+      "epoch": 0.9185540344457763,
+      "grad_norm": 0.660201370716095,
+      "learning_rate": 9.997385429418555e-07,
+      "loss": 0.7505,
+      "step": 3640
+    },
+    {
+      "epoch": 0.9210775345404075,
+      "grad_norm": 0.6315280795097351,
+      "learning_rate": 9.389928783631207e-07,
+      "loss": 0.7543,
+      "step": 3650
+    },
+    {
+      "epoch": 0.9236010346350388,
+      "grad_norm": 0.5867049694061279,
+      "learning_rate": 8.801157076146705e-07,
+      "loss": 0.7151,
+      "step": 3660
+    },
+    {
+      "epoch": 0.92612453472967,
+      "grad_norm": 0.6637170910835266,
+      "learning_rate": 8.231116028920765e-07,
+      "loss": 0.7189,
+      "step": 3670
+    },
+    {
+      "epoch": 0.9286480348243014,
+      "grad_norm": 0.5963181853294373,
+      "learning_rate": 7.679849909351472e-07,
+      "loss": 0.708,
+      "step": 3680
+    },
+    {
+      "epoch": 0.9311715349189326,
+      "grad_norm": 0.677011251449585,
+      "learning_rate": 7.147401526841485e-07,
+      "loss": 0.7049,
+      "step": 3690
+    },
+    {
+      "epoch": 0.9336950350135638,
+      "grad_norm": 0.6534477472305298,
+      "learning_rate": 6.633812229473791e-07,
+      "loss": 0.7607,
+      "step": 3700
+    },
+    {
+      "epoch": 0.9362185351081951,
+      "grad_norm": 0.6873798370361328,
+      "learning_rate": 6.139121900800515e-07,
+      "loss": 0.7044,
+      "step": 3710
+    },
+    {
+      "epoch": 0.9387420352028263,
+      "grad_norm": 0.6045345067977905,
+      "learning_rate": 5.663368956745963e-07,
+      "loss": 0.7141,
+      "step": 3720
+    },
+    {
+      "epoch": 0.9412655352974576,
+      "grad_norm": 0.6731101274490356,
+      "learning_rate": 5.206590342623164e-07,
+      "loss": 0.7156,
+      "step": 3730
+    },
+    {
+      "epoch": 0.9437890353920888,
+      "grad_norm": 0.5900946855545044,
+      "learning_rate": 4.768821530264977e-07,
+      "loss": 0.7481,
+      "step": 3740
+    },
+    {
+      "epoch": 0.9463125354867201,
+      "grad_norm": 0.6716229915618896,
+      "learning_rate": 4.350096515269325e-07,
+      "loss": 0.7438,
+      "step": 3750
+    },
+    {
+      "epoch": 0.9488360355813513,
+      "grad_norm": 0.6293070912361145,
+      "learning_rate": 3.950447814359409e-07,
+      "loss": 0.7449,
+      "step": 3760
+    },
+    {
+      "epoch": 0.9513595356759826,
+      "grad_norm": 0.6327414512634277,
+      "learning_rate": 3.5699064628583745e-07,
+      "loss": 0.7241,
+      "step": 3770
+    },
+    {
+      "epoch": 0.9538830357706138,
+      "grad_norm": 0.5867704749107361,
+      "learning_rate": 3.2085020122793186e-07,
+      "loss": 0.7247,
+      "step": 3780
+    },
+    {
+      "epoch": 0.956406535865245,
+      "grad_norm": 0.6275327801704407,
+      "learning_rate": 2.8662625280304613e-07,
+      "loss": 0.7166,
+      "step": 3790
+    },
+    {
+      "epoch": 0.9589300359598764,
+      "grad_norm": 0.634965717792511,
+      "learning_rate": 2.5432145872355816e-07,
+      "loss": 0.713,
+      "step": 3800
+    },
+    {
+      "epoch": 0.9614535360545076,
+      "grad_norm": 0.685463011264801,
+      "learning_rate": 2.2393832766701706e-07,
+      "loss": 0.7328,
+      "step": 3810
+    },
+    {
+      "epoch": 0.9639770361491389,
+      "grad_norm": 0.5693522095680237,
+      "learning_rate": 1.9547921908133483e-07,
+      "loss": 0.7333,
+      "step": 3820
+    },
+    {
+      "epoch": 0.9665005362437701,
+      "grad_norm": 0.6611155271530151,
+      "learning_rate": 1.689463430015442e-07,
+      "loss": 0.7102,
+      "step": 3830
+    },
+    {
+      "epoch": 0.9690240363384014,
+      "grad_norm": 0.6425820589065552,
+      "learning_rate": 1.443417598781971e-07,
+      "loss": 0.7294,
+      "step": 3840
+    },
+    {
+      "epoch": 0.9715475364330326,
+      "grad_norm": 0.66698157787323,
+      "learning_rate": 1.2166738041733684e-07,
+      "loss": 0.728,
+      "step": 3850
+    },
+    {
+      "epoch": 0.9740710365276639,
+      "grad_norm": 0.6152459383010864,
+      "learning_rate": 1.0092496543212814e-07,
+      "loss": 0.7201,
+      "step": 3860
+    },
+    {
+      "epoch": 0.9765945366222951,
+      "grad_norm": 0.613070547580719,
+      "learning_rate": 8.211612570611926e-08,
+      "loss": 0.7148,
+      "step": 3870
+    },
+    {
+      "epoch": 0.9791180367169264,
+      "grad_norm": 0.6403253674507141,
+      "learning_rate": 6.524232186815305e-08,
+      "loss": 0.7129,
+      "step": 3880
+    },
+    {
+      "epoch": 0.9816415368115576,
+      "grad_norm": 0.6211004257202148,
+      "learning_rate": 5.03048642789411e-08,
+      "loss": 0.7068,
+      "step": 3890
+    },
+    {
+      "epoch": 0.9841650369061888,
+      "grad_norm": 0.7897922992706299,
+      "learning_rate": 3.730491292930072e-08,
+      "loss": 0.7184,
+      "step": 3900
+    },
+    {
+      "epoch": 0.9866885370008202,
+      "grad_norm": 0.756610631942749,
+      "learning_rate": 2.624347735007693e-08,
+      "loss": 0.738,
+      "step": 3910
+    },
+    {
+      "epoch": 0.9892120370954514,
+      "grad_norm": 0.6681756973266602,
+      "learning_rate": 1.7121416533749658e-08,
+      "loss": 0.7298,
+      "step": 3920
+    },
+    {
+      "epoch": 0.9917355371900827,
+      "grad_norm": 0.630942702293396,
+      "learning_rate": 9.939438867723194e-09,
+      "loss": 0.7322,
+      "step": 3930
+    },
+    {
+      "epoch": 0.9942590372847139,
+      "grad_norm": 0.6125476360321045,
+      "learning_rate": 4.6981020793118725e-09,
+      "loss": 0.7414,
+      "step": 3940
+    },
+    {
+      "epoch": 0.9967825373793452,
+      "grad_norm": 0.6152609586715698,
+      "learning_rate": 1.3978131924385906e-09,
+      "loss": 0.7124,
+      "step": 3950
+    },
+    {
+      "epoch": 0.9993060374739764,
+      "grad_norm": 0.6308910846710205,
+      "learning_rate": 3.88284960184393e-11,
+      "loss": 0.7419,
+      "step": 3960
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 3962,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.315842275872604e+18,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:521f433a3f3ed9e53b6435fbdd695b6a70697bf5603a54f87df385d0426108e2
+size 7160

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)