Upload folder using huggingface_hub

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image>": 32000
+}

chat_template.jinja

@@ -0,0 +1,2 @@
+{% for message in messages %}{% if message['role'] != 'system' %}{{ message['role'].upper() + ': '}}{% endif %}{# Render all images first #}{% for content in message['content'] | selectattr('type', 'equalto', 'image') %}{{ '<image>
+' }}{% endfor %}{# Render all text next #}{% if message['role'] != 'assistant' %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{{ content['text'] + ' '}}{% endfor %}{% else %}{% for content in message['content'] | selectattr('type', 'equalto', 'text') %}{% generation %}{{ content['text'] + ' '}}{% endgeneration %}{% endfor %}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ 'ASSISTANT:' }}{% endif %}

config.json

@@ -0,0 +1,87 @@
+{
+  "architectures": [
+    "LlavaNextForConditionalGeneration"
+  ],
+  "hidden_size": 4096,
+  "ignore_index": -100,
+  "image_grid_pinpoints": [
+    [
+      336,
+      672
+    ],
+    [
+      672,
+      336
+    ],
+    [
+      672,
+      672
+    ],
+    [
+      1008,
+      336
+    ],
+    [
+      336,
+      1008
+    ]
+  ],
+  "image_seq_length": 576,
+  "image_token_index": 32000,
+  "model_type": "llava_next",
+  "multimodal_projector_bias": true,
+  "projector_hidden_act": "gelu",
+  "text_config": {
+    "_name_or_path": "lmsys/vicuna-7b-v1.5",
+    "architectures": [
+      "LlamaForCausalLM"
+    ],
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "head_dim": 128,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "initializer_range": 0.02,
+    "intermediate_size": 11008,
+    "max_position_embeddings": 4096,
+    "mlp_bias": false,
+    "model_type": "llama",
+    "num_attention_heads": 32,
+    "num_hidden_layers": 32,
+    "num_key_value_heads": 32,
+    "pad_token_id": 0,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-05,
+    "rope_scaling": null,
+    "rope_theta": 10000.0,
+    "torch_dtype": "float32",
+    "use_cache": false,
+    "vocab_size": 32064
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.55.0",
+  "use_cache": false,
+  "use_image_newline_parameter": true,
+  "vision_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "quick_gelu",
+    "hidden_size": 1024,
+    "image_size": 336,
+    "initializer_factor": 1.0,
+    "initializer_range": 0.02,
+    "intermediate_size": 4096,
+    "layer_norm_eps": 1e-05,
+    "model_type": "clip_vision_model",
+    "num_attention_heads": 16,
+    "num_channels": 3,
+    "num_hidden_layers": 24,
+    "patch_size": 14,
+    "projection_dim": 768,
+    "torch_dtype": "float32",
+    "vocab_size": 32000
+  },
+  "vision_feature_layer": -2,
+  "vision_feature_select_strategy": "default",
+  "vocab_size": 32064
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 0,
+  "transformers_version": "4.55.0"
+}

global_step1200/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efd1910a7d31c5205df4c90f0bf432e9df227398e822226d027a77c2fd7237ac
+size 10108550145

global_step1200/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92dcddd38d7209cba3cc4be8523f63e6532f195f7b1662ab56ba6c1b72990912
+size 10108550145

global_step1200/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee1dbbbef0558bd2948744e740f5ea86a248290bf3f83c43765a1d191fda0b7a
+size 10108550145

global_step1200/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df95d66aa7fd3ba82336fe5c8edcb1bba82069834f60af436949b05d21f41227
+size 10108550145

global_step1200/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8c5701e6f5012d9809357529557669f4a69aea781bc88207740a5ca7330b6dd
+size 10108550145

global_step1200/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48c54c08643d4a488786ba6c4fa6622739e40a77177ca8370b3be55f04bcb382
+size 10108550145

global_step1200/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2906a11b1f6be2357e5408a7729e79c83d96f22cd58f262e03d3fd4714e5fe8c
+size 10108550145

global_step1200/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17849fa819a3d801a1179ffdd656fa3c5d60a1ba21e44ae907067e888fbbac83
+size 10108550145

global_step1200/zero_pp_rank_0_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47422e20fa4e9199e0802bcf826eff1166b80bb4d32978cc3dbcc98100bd09e1
+size 81627229

global_step1200/zero_pp_rank_1_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96c5cf5535f2a1cf8125af09e7efbf85c5d991b7756a6b20a1a12c9d70dcf9ca
+size 81627229

global_step1200/zero_pp_rank_2_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dfe3e900d8b72b279ae82c8502abcd06609dbf2225491bdc31d48e382586bcfb
+size 81627229

global_step1200/zero_pp_rank_3_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52be3ebdd0f7f5e78b2241ccad500495537f642a8fcba960b7c9777911a203b5
+size 81627229

global_step1200/zero_pp_rank_4_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ff375e72ab832529c1a1e825686d7a60bf8b095e1614d065ae142d7a159b01a
+size 81627229

global_step1200/zero_pp_rank_5_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f32e28cc5be7e7ee4f557c3ccebfaa627b24afa54413ea5e28d56e05da4a261
+size 81627229

global_step1200/zero_pp_rank_6_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f8fe819cfd335078899c38dfe972daa64fbb239a58d188911be524d1825b263
+size 81627229

global_step1200/zero_pp_rank_7_mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6322f2de3af599969670c3aa01b4ec0b04a54fe634870aa650056f850c3e5966
+size 81627229

latest

@@ -0,0 +1 @@
+global_step1200

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62e08f8dd5d393e52b1cd61a21853aaf6a11da7f9e172eacbb59fd7adbdc87d4
+size 4992938952

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:190724d9dfdab814ca60728dfcf543cb24cd0ec95f50a9d040f5737d31eaaa32
+size 4957878552

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0aab246b50f6199e78f865a610bee5693571a1cbccb9f0451424ac74915385fc
+size 4176137496

model.safetensors.index.json

@@ -0,0 +1,695 @@
+{
+  "metadata": {
+    "total_parameters": 603136,
+    "total_size": 14126862336
+  },
+  "weight_map": {
+    "image_newline": "model-00001-of-00003.safetensors",
+    "language_model.lm_head.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.10.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.10.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.10.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.10.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.10.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.10.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.10.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.10.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.11.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.11.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.11.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.11.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.11.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.22.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.22.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.22.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.22.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.22.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "language_model.model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.23.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.23.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.23.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.23.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.23.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.23.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.23.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.24.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.24.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.24.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.25.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.25.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.25.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.26.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.26.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.26.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.27.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.27.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.27.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.28.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.28.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.28.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.29.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.29.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.29.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "language_model.model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "language_model.model.norm.weight": "model-00003-of-00003.safetensors",
+    "multi_modal_projector.linear_1.bias": "model-00001-of-00003.safetensors",
+    "multi_modal_projector.linear_1.weight": "model-00001-of-00003.safetensors",
+    "multi_modal_projector.linear_2.bias": "model-00001-of-00003.safetensors",
+    "multi_modal_projector.linear_2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.embeddings.class_embedding": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.embeddings.patch_embedding.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.embeddings.position_embedding.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.12.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.13.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.14.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.15.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.16.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.17.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.18.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.19.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.20.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.21.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.22.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.23.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.layer_norm2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc1.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc1.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc2.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.mlp.fc2.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.post_layernorm.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.post_layernorm.weight": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.pre_layrnorm.bias": "model-00001-of-00003.safetensors",
+    "vision_tower.vision_model.pre_layrnorm.weight": "model-00001-of-00003.safetensors"
+  }
+}

preprocessor_config.json

@@ -0,0 +1,58 @@
+{
+  "aspect_ratio_setting": "anyres",
+  "crop_size": {
+    "height": 336,
+    "width": 336
+  },
+  "data_format": "channels_first",
+  "default_to_square": false,
+  "device": null,
+  "disable_grouping": null,
+  "do_center_crop": true,
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_pad": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_grid_pinpoints": [
+    [
+      336,
+      672
+    ],
+    [
+      672,
+      336
+    ],
+    [
+      672,
+      672
+    ],
+    [
+      1008,
+      336
+    ],
+    [
+      336,
+      1008
+    ]
+  ],
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "LlavaNextImageProcessorFast",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "input_data_format": null,
+  "processor_class": "LlavaNextProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "return_tensors": null,
+  "size": {
+    "shortest_edge": 336
+  }
+}

processor_config.json

@@ -0,0 +1,7 @@
+{
+  "image_token": "<image>",
+  "num_additional_image_tokens": 1,
+  "patch_size": 14,
+  "processor_class": "LlavaNextProcessor",
+  "vision_feature_select_strategy": "default"
+}

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:478b41e9f26d338fd8f896e08cad1adab7c423b61f1b45754113bc78d256a3f9
+size 16389

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce29a8767a7d907dd24987aa2c3e654d4317f3042fbc13b5b72cadb46d43311a
+size 16389

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61a48db011646b4e9a867bf12f4a233cad5dfbfe309686f8996c250196d3783a
+size 16389

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b9562ee822472a4f01dcd6349ab3d1ef42a48915fe3b92e843a0c37db53c8421
+size 16389

rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7d2767d83c3bf27f12db022b0632e2c4f8c164274ba75e380cf18f9d5f21819
+size 16389

rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76816358d4e5db8149d60d55234db658d67a13c0c1ce05d7404cf7125a676a5c
+size 16389

rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1562e7520c977d178183d641f70abcf3f57da2489938756cfbebf9b6e6c1a9fd
+size 16389

rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6b6cabaed045c5398cd1b732f7ec48bd363f3b43cd24e0e70e641a42bd00c28
+size 16389

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:96bb682ac8cbef13e4c6a0c008b6a7394d027af747629d85c154e1d8b52c3f50
+size 1465

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image>",
+  "pad_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,57 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": true,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32000": {
+      "content": "<image>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {
+    "image_token": "<image>"
+  },
+  "image_token": "<image>",
+  "legacy": false,
+  "model_max_length": 8192,
+  "pad_token": "<unk>",
+  "padding_side": "right",
+  "processor_class": "LlavaNextProcessor",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

trainer_state.json

@@ -0,0 +1,1834 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.6910452058738843,
+  "eval_steps": 500,
+  "global_step": 1200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.005758710048949035,
+      "grad_norm": 19.853458404541016,
+      "learning_rate": 5.172413793103448e-08,
+      "logits/chosen": -0.562769889831543,
+      "logits/rejected": -0.5616950988769531,
+      "logps/chosen": -127.0331802368164,
+      "logps/rejected": -83.81842041015625,
+      "loss": 0.6929,
+      "rewards/accuracies": 0.4124999940395355,
+      "rewards/chosen": -0.00019152756431140006,
+      "rewards/margins": 0.000841214437969029,
+      "rewards/rejected": -0.0010327422060072422,
+      "step": 10
+    },
+    {
+      "epoch": 0.01151742009789807,
+      "grad_norm": 19.53203582763672,
+      "learning_rate": 1.0919540229885057e-07,
+      "logits/chosen": -0.6015105247497559,
+      "logits/rejected": -0.6022548079490662,
+      "logps/chosen": -143.99452209472656,
+      "logps/rejected": -89.27932739257812,
+      "loss": 0.6925,
+      "rewards/accuracies": 0.5062500238418579,
+      "rewards/chosen": 0.00048787007108330727,
+      "rewards/margins": 0.001671066740527749,
+      "rewards/rejected": -0.0011831964366137981,
+      "step": 20
+    },
+    {
+      "epoch": 0.017276130146847105,
+      "grad_norm": 22.230405807495117,
+      "learning_rate": 1.6666666666666665e-07,
+      "logits/chosen": -0.582757830619812,
+      "logits/rejected": -0.5794022679328918,
+      "logps/chosen": -128.6601104736328,
+      "logps/rejected": -76.13624572753906,
+      "loss": 0.6944,
+      "rewards/accuracies": 0.4812500476837158,
+      "rewards/chosen": -0.0016581722302362323,
+      "rewards/margins": -0.0020952033810317516,
+      "rewards/rejected": 0.0004370305105112493,
+      "step": 30
+    },
+    {
+      "epoch": 0.02303484019579614,
+      "grad_norm": 19.54130744934082,
+      "learning_rate": 2.2413793103448274e-07,
+      "logits/chosen": -0.5358110666275024,
+      "logits/rejected": -0.5239226222038269,
+      "logps/chosen": -148.56979370117188,
+      "logps/rejected": -91.35775756835938,
+      "loss": 0.6916,
+      "rewards/accuracies": 0.609375,
+      "rewards/chosen": 0.0023662634193897247,
+      "rewards/margins": 0.0035909225698560476,
+      "rewards/rejected": -0.0012246592668816447,
+      "step": 40
+    },
+    {
+      "epoch": 0.028793550244745177,
+      "grad_norm": 29.174854278564453,
+      "learning_rate": 2.816091954022988e-07,
+      "logits/chosen": -0.6022522449493408,
+      "logits/rejected": -0.6001952886581421,
+      "logps/chosen": -165.06558227539062,
+      "logps/rejected": -83.33187866210938,
+      "loss": 0.6882,
+      "rewards/accuracies": 0.6281249523162842,
+      "rewards/chosen": 0.004078409168869257,
+      "rewards/margins": 0.011784590780735016,
+      "rewards/rejected": -0.007706179283559322,
+      "step": 50
+    },
+    {
+      "epoch": 0.03455226029369421,
+      "grad_norm": 22.835206985473633,
+      "learning_rate": 3.390804597701149e-07,
+      "logits/chosen": -0.5758532881736755,
+      "logits/rejected": -0.5722157955169678,
+      "logps/chosen": -136.6300811767578,
+      "logps/rejected": -82.68758392333984,
+      "loss": 0.6806,
+      "rewards/accuracies": 0.6656249761581421,
+      "rewards/chosen": 0.014781510457396507,
+      "rewards/margins": 0.026332611218094826,
+      "rewards/rejected": -0.01155109889805317,
+      "step": 60
+    },
+    {
+      "epoch": 0.04031097034264325,
+      "grad_norm": 18.888277053833008,
+      "learning_rate": 3.9655172413793105e-07,
+      "logits/chosen": -0.584574282169342,
+      "logits/rejected": -0.5734848380088806,
+      "logps/chosen": -112.57417297363281,
+      "logps/rejected": -81.75566864013672,
+      "loss": 0.673,
+      "rewards/accuracies": 0.659375011920929,
+      "rewards/chosen": 0.026554793119430542,
+      "rewards/margins": 0.04325523227453232,
+      "rewards/rejected": -0.016700439155101776,
+      "step": 70
+    },
+    {
+      "epoch": 0.04606968039159228,
+      "grad_norm": 18.912086486816406,
+      "learning_rate": 4.540229885057471e-07,
+      "logits/chosen": -0.5466264486312866,
+      "logits/rejected": -0.5328729152679443,
+      "logps/chosen": -150.10618591308594,
+      "logps/rejected": -81.80156707763672,
+      "loss": 0.6557,
+      "rewards/accuracies": 0.7187500596046448,
+      "rewards/chosen": 0.05700210481882095,
+      "rewards/margins": 0.08444017171859741,
+      "rewards/rejected": -0.02743806689977646,
+      "step": 80
+    },
+    {
+      "epoch": 0.05182839044054132,
+      "grad_norm": 20.304170608520508,
+      "learning_rate": 5.114942528735632e-07,
+      "logits/chosen": -0.5730519890785217,
+      "logits/rejected": -0.5601622462272644,
+      "logps/chosen": -135.96995544433594,
+      "logps/rejected": -83.88729858398438,
+      "loss": 0.6352,
+      "rewards/accuracies": 0.7312500476837158,
+      "rewards/chosen": 0.07092135399580002,
+      "rewards/margins": 0.13441398739814758,
+      "rewards/rejected": -0.06349264085292816,
+      "step": 90
+    },
+    {
+      "epoch": 0.05758710048949035,
+      "grad_norm": 19.214027404785156,
+      "learning_rate": 5.689655172413793e-07,
+      "logits/chosen": -0.5703257322311401,
+      "logits/rejected": -0.5486276745796204,
+      "logps/chosen": -132.86158752441406,
+      "logps/rejected": -84.33039093017578,
+      "loss": 0.6198,
+      "rewards/accuracies": 0.768750011920929,
+      "rewards/chosen": 0.0864553153514862,
+      "rewards/margins": 0.18071594834327698,
+      "rewards/rejected": -0.09426065534353256,
+      "step": 100
+    },
+    {
+      "epoch": 0.06334581053843939,
+      "grad_norm": 18.2548885345459,
+      "learning_rate": 6.264367816091954e-07,
+      "logits/chosen": -0.5512282848358154,
+      "logits/rejected": -0.5419508814811707,
+      "logps/chosen": -128.60459899902344,
+      "logps/rejected": -88.85281372070312,
+      "loss": 0.6071,
+      "rewards/accuracies": 0.7250000238418579,
+      "rewards/chosen": 0.08900328725576401,
+      "rewards/margins": 0.23027661442756653,
+      "rewards/rejected": -0.1412733495235443,
+      "step": 110
+    },
+    {
+      "epoch": 0.06910452058738842,
+      "grad_norm": 16.20355987548828,
+      "learning_rate": 6.839080459770114e-07,
+      "logits/chosen": -0.5748304724693298,
+      "logits/rejected": -0.5648065805435181,
+      "logps/chosen": -131.20220947265625,
+      "logps/rejected": -83.58260345458984,
+      "loss": 0.588,
+      "rewards/accuracies": 0.7406250238418579,
+      "rewards/chosen": 0.1262008100748062,
+      "rewards/margins": 0.2918751537799835,
+      "rewards/rejected": -0.1656743437051773,
+      "step": 120
+    },
+    {
+      "epoch": 0.07486323063633746,
+      "grad_norm": 14.335201263427734,
+      "learning_rate": 7.413793103448276e-07,
+      "logits/chosen": -0.5980964303016663,
+      "logits/rejected": -0.5801858901977539,
+      "logps/chosen": -129.68930053710938,
+      "logps/rejected": -77.69696807861328,
+      "loss": 0.562,
+      "rewards/accuracies": 0.7406250238418579,
+      "rewards/chosen": 0.13151288032531738,
+      "rewards/margins": 0.3828701972961426,
+      "rewards/rejected": -0.2513573467731476,
+      "step": 130
+    },
+    {
+      "epoch": 0.0806219406852865,
+      "grad_norm": 17.238740921020508,
+      "learning_rate": 7.988505747126436e-07,
+      "logits/chosen": -0.5309160351753235,
+      "logits/rejected": -0.5147740840911865,
+      "logps/chosen": -136.6770477294922,
+      "logps/rejected": -84.51583862304688,
+      "loss": 0.5519,
+      "rewards/accuracies": 0.753125011920929,
+      "rewards/chosen": 0.19663727283477783,
+      "rewards/margins": 0.4776872396469116,
+      "rewards/rejected": -0.2810499668121338,
+      "step": 140
+    },
+    {
+      "epoch": 0.08638065073423554,
+      "grad_norm": 15.70281982421875,
+      "learning_rate": 8.563218390804597e-07,
+      "logits/chosen": -0.5557988286018372,
+      "logits/rejected": -0.540660560131073,
+      "logps/chosen": -144.00198364257812,
+      "logps/rejected": -100.53820037841797,
+      "loss": 0.5486,
+      "rewards/accuracies": 0.7875000238418579,
+      "rewards/chosen": 0.23025350272655487,
+      "rewards/margins": 0.5131940245628357,
+      "rewards/rejected": -0.28294044733047485,
+      "step": 150
+    },
+    {
+      "epoch": 0.09213936078318456,
+      "grad_norm": 17.085052490234375,
+      "learning_rate": 9.137931034482759e-07,
+      "logits/chosen": -0.5505703687667847,
+      "logits/rejected": -0.5197522640228271,
+      "logps/chosen": -136.61239624023438,
+      "logps/rejected": -96.98216247558594,
+      "loss": 0.5496,
+      "rewards/accuracies": 0.7375000715255737,
+      "rewards/chosen": 0.1979924887418747,
+      "rewards/margins": 0.5241073966026306,
+      "rewards/rejected": -0.3261149227619171,
+      "step": 160
+    },
+    {
+      "epoch": 0.0978980708321336,
+      "grad_norm": 16.59015464782715,
+      "learning_rate": 9.712643678160918e-07,
+      "logits/chosen": -0.5620574355125427,
+      "logits/rejected": -0.5433012247085571,
+      "logps/chosen": -130.6116943359375,
+      "logps/rejected": -88.13215637207031,
+      "loss": 0.5364,
+      "rewards/accuracies": 0.7500000596046448,
+      "rewards/chosen": 0.16394342482089996,
+      "rewards/margins": 0.577468752861023,
+      "rewards/rejected": -0.4135252833366394,
+      "step": 170
+    },
+    {
+      "epoch": 0.10365678088108264,
+      "grad_norm": 14.182692527770996,
+      "learning_rate": 9.99974750187855e-07,
+      "logits/chosen": -0.6073015332221985,
+      "logits/rejected": -0.5830385088920593,
+      "logps/chosen": -134.97865295410156,
+      "logps/rejected": -82.4712905883789,
+      "loss": 0.5313,
+      "rewards/accuracies": 0.7562500238418579,
+      "rewards/chosen": 0.26480597257614136,
+      "rewards/margins": 0.6398300528526306,
+      "rewards/rejected": -0.37502405047416687,
+      "step": 180
+    },
+    {
+      "epoch": 0.10941549093003167,
+      "grad_norm": 15.201092720031738,
+      "learning_rate": 9.997727669917109e-07,
+      "logits/chosen": -0.5887178182601929,
+      "logits/rejected": -0.5629587769508362,
+      "logps/chosen": -134.81166076660156,
+      "logps/rejected": -91.30838775634766,
+      "loss": 0.5264,
+      "rewards/accuracies": 0.734375,
+      "rewards/chosen": 0.15517549216747284,
+      "rewards/margins": 0.6431481242179871,
+      "rewards/rejected": -0.4879727065563202,
+      "step": 190
+    },
+    {
+      "epoch": 0.1151742009789807,
+      "grad_norm": 13.136405944824219,
+      "learning_rate": 9.993688821979663e-07,
+      "logits/chosen": -0.5851372480392456,
+      "logits/rejected": -0.552108645439148,
+      "logps/chosen": -143.36703491210938,
+      "logps/rejected": -102.49083709716797,
+      "loss": 0.4568,
+      "rewards/accuracies": 0.8062500357627869,
+      "rewards/chosen": 0.3289705216884613,
+      "rewards/margins": 0.8849547505378723,
+      "rewards/rejected": -0.5559841394424438,
+      "step": 200
+    },
+    {
+      "epoch": 0.12093291102792975,
+      "grad_norm": 26.78643035888672,
+      "learning_rate": 9.98763258970744e-07,
+      "logits/chosen": -0.6051122546195984,
+      "logits/rejected": -0.5708233714103699,
+      "logps/chosen": -152.91024780273438,
+      "logps/rejected": -95.11014556884766,
+      "loss": 0.4741,
+      "rewards/accuracies": 0.7718750238418579,
+      "rewards/chosen": 0.2776795029640198,
+      "rewards/margins": 0.8509930372238159,
+      "rewards/rejected": -0.5733135342597961,
+      "step": 210
+    },
+    {
+      "epoch": 0.12669162107687879,
+      "grad_norm": 24.18678855895996,
+      "learning_rate": 9.979561419738296e-07,
+      "logits/chosen": -0.6134639978408813,
+      "logits/rejected": -0.5940448045730591,
+      "logps/chosen": -132.16368103027344,
+      "logps/rejected": -93.20537567138672,
+      "loss": 0.5273,
+      "rewards/accuracies": 0.7437499761581421,
+      "rewards/chosen": 0.22999629378318787,
+      "rewards/margins": 0.7246684432029724,
+      "rewards/rejected": -0.49467211961746216,
+      "step": 220
+    },
+    {
+      "epoch": 0.13245033112582782,
+      "grad_norm": 12.556412696838379,
+      "learning_rate": 9.969478572718307e-07,
+      "logits/chosen": -0.6136160492897034,
+      "logits/rejected": -0.5797239542007446,
+      "logps/chosen": -124.40892028808594,
+      "logps/rejected": -90.66249084472656,
+      "loss": 0.4829,
+      "rewards/accuracies": 0.778124988079071,
+      "rewards/chosen": 0.29396748542785645,
+      "rewards/margins": 0.8314955830574036,
+      "rewards/rejected": -0.5375280976295471,
+      "step": 230
+    },
+    {
+      "epoch": 0.13820904117477684,
+      "grad_norm": 17.433090209960938,
+      "learning_rate": 9.95738812198451e-07,
+      "logits/chosen": -0.6312894821166992,
+      "logits/rejected": -0.6113855838775635,
+      "logps/chosen": -122.38996124267578,
+      "logps/rejected": -82.93913269042969,
+      "loss": 0.5194,
+      "rewards/accuracies": 0.7437500357627869,
+      "rewards/chosen": 0.19915173947811127,
+      "rewards/margins": 0.7839306592941284,
+      "rewards/rejected": -0.5847789645195007,
+      "step": 240
+    },
+    {
+      "epoch": 0.14396775122372588,
+      "grad_norm": 16.78449058532715,
+      "learning_rate": 9.943294951919325e-07,
+      "logits/chosen": -0.6331308484077454,
+      "logits/rejected": -0.6054038405418396,
+      "logps/chosen": -131.5919647216797,
+      "logps/rejected": -96.62232971191406,
+      "loss": 0.4465,
+      "rewards/accuracies": 0.7999999523162842,
+      "rewards/chosen": 0.3666311800479889,
+      "rewards/margins": 0.9798641204833984,
+      "rewards/rejected": -0.6132329702377319,
+      "step": 250
+    },
+    {
+      "epoch": 0.14972646127267492,
+      "grad_norm": 18.236032485961914,
+      "learning_rate": 9.92720475597734e-07,
+      "logits/chosen": -0.6338821649551392,
+      "logits/rejected": -0.6092484593391418,
+      "logps/chosen": -120.25117492675781,
+      "logps/rejected": -93.84246063232422,
+      "loss": 0.4875,
+      "rewards/accuracies": 0.793749988079071,
+      "rewards/chosen": 0.22484159469604492,
+      "rewards/margins": 0.8752425312995911,
+      "rewards/rejected": -0.6504008769989014,
+      "step": 260
+    },
+    {
+      "epoch": 0.15548517132162396,
+      "grad_norm": 11.804591178894043,
+      "learning_rate": 9.909124034385224e-07,
+      "logits/chosen": -0.6439257860183716,
+      "logits/rejected": -0.6129547357559204,
+      "logps/chosen": -129.7813262939453,
+      "logps/rejected": -83.8338623046875,
+      "loss": 0.4296,
+      "rewards/accuracies": 0.7875000238418579,
+      "rewards/chosen": 0.3556497097015381,
+      "rewards/margins": 1.0146920680999756,
+      "rewards/rejected": -0.6590422987937927,
+      "step": 270
+    },
+    {
+      "epoch": 0.161243881370573,
+      "grad_norm": 13.709634780883789,
+      "learning_rate": 9.889060091515707e-07,
+      "logits/chosen": -0.6321221590042114,
+      "logits/rejected": -0.6142387986183167,
+      "logps/chosen": -123.8255844116211,
+      "logps/rejected": -100.52135467529297,
+      "loss": 0.4754,
+      "rewards/accuracies": 0.746874988079071,
+      "rewards/chosen": 0.21102119982242584,
+      "rewards/margins": 0.9996404647827148,
+      "rewards/rejected": -0.7886192798614502,
+      "step": 280
+    },
+    {
+      "epoch": 0.16700259141952203,
+      "grad_norm": 16.686607360839844,
+      "learning_rate": 9.86702103293674e-07,
+      "logits/chosen": -0.6492558717727661,
+      "logits/rejected": -0.6142836809158325,
+      "logps/chosen": -149.1907196044922,
+      "logps/rejected": -94.28280639648438,
+      "loss": 0.4284,
+      "rewards/accuracies": 0.84375,
+      "rewards/chosen": 0.2924773693084717,
+      "rewards/margins": 1.1958118677139282,
+      "rewards/rejected": -0.9033344984054565,
+      "step": 290
+    },
+    {
+      "epoch": 0.17276130146847107,
+      "grad_norm": 18.625837326049805,
+      "learning_rate": 9.843015762136925e-07,
+      "logits/chosen": -0.6587651371955872,
+      "logits/rejected": -0.6321001052856445,
+      "logps/chosen": -129.7299041748047,
+      "logps/rejected": -85.21316528320312,
+      "loss": 0.443,
+      "rewards/accuracies": 0.7999999523162842,
+      "rewards/chosen": 0.3374091684818268,
+      "rewards/margins": 1.1282662153244019,
+      "rewards/rejected": -0.7908569574356079,
+      "step": 300
+    },
+    {
+      "epoch": 0.17852001151742009,
+      "grad_norm": 19.16632843017578,
+      "learning_rate": 9.817053976928643e-07,
+      "logits/chosen": -0.6443223357200623,
+      "logits/rejected": -0.6130384206771851,
+      "logps/chosen": -137.69725036621094,
+      "logps/rejected": -92.34024047851562,
+      "loss": 0.4861,
+      "rewards/accuracies": 0.7875000238418579,
+      "rewards/chosen": 0.2451004534959793,
+      "rewards/margins": 1.0736701488494873,
+      "rewards/rejected": -0.8285696506500244,
+      "step": 310
+    },
+    {
+      "epoch": 0.18427872156636912,
+      "grad_norm": 21.944595336914062,
+      "learning_rate": 9.789146165530254e-07,
+      "logits/chosen": -0.6652050018310547,
+      "logits/rejected": -0.6271861791610718,
+      "logps/chosen": -127.1378173828125,
+      "logps/rejected": -81.23876953125,
+      "loss": 0.444,
+      "rewards/accuracies": 0.8156250715255737,
+      "rewards/chosen": 0.20780594646930695,
+      "rewards/margins": 1.1428509950637817,
+      "rewards/rejected": -0.9350449442863464,
+      "step": 320
+    },
+    {
+      "epoch": 0.19003743161531816,
+      "grad_norm": 15.266040802001953,
+      "learning_rate": 9.759303602328992e-07,
+      "logits/chosen": -0.6632432341575623,
+      "logits/rejected": -0.6333745718002319,
+      "logps/chosen": -146.56114196777344,
+      "logps/rejected": -89.61295318603516,
+      "loss": 0.4436,
+      "rewards/accuracies": 0.8093750476837158,
+      "rewards/chosen": 0.305823415517807,
+      "rewards/margins": 1.1472581624984741,
+      "rewards/rejected": -0.8414347171783447,
+      "step": 330
+    },
+    {
+      "epoch": 0.1957961416642672,
+      "grad_norm": 14.493436813354492,
+      "learning_rate": 9.727538343326278e-07,
+      "logits/chosen": -0.6710892915725708,
+      "logits/rejected": -0.6370099782943726,
+      "logps/chosen": -139.57029724121094,
+      "logps/rejected": -87.5551986694336,
+      "loss": 0.4235,
+      "rewards/accuracies": 0.8031250238418579,
+      "rewards/chosen": 0.4220184087753296,
+      "rewards/margins": 1.233292579650879,
+      "rewards/rejected": -0.8112741708755493,
+      "step": 340
+    },
+    {
+      "epoch": 0.20155485171321624,
+      "grad_norm": 33.87898635864258,
+      "learning_rate": 9.693863221267237e-07,
+      "logits/chosen": -0.6808286309242249,
+      "logits/rejected": -0.6564449071884155,
+      "logps/chosen": -127.93206024169922,
+      "logps/rejected": -95.1683578491211,
+      "loss": 0.4536,
+      "rewards/accuracies": 0.815625011920929,
+      "rewards/chosen": 0.3246816098690033,
+      "rewards/margins": 1.1423108577728271,
+      "rewards/rejected": -0.8176291584968567,
+      "step": 350
+    },
+    {
+      "epoch": 0.20731356176216528,
+      "grad_norm": 22.24034881591797,
+      "learning_rate": 9.658291840456452e-07,
+      "logits/chosen": -0.6467706561088562,
+      "logits/rejected": -0.6211069822311401,
+      "logps/chosen": -155.3069610595703,
+      "logps/rejected": -108.99771118164062,
+      "loss": 0.4981,
+      "rewards/accuracies": 0.7781249284744263,
+      "rewards/chosen": 0.42678219079971313,
+      "rewards/margins": 1.1551151275634766,
+      "rewards/rejected": -0.7283328175544739,
+      "step": 360
+    },
+    {
+      "epoch": 0.21307227181111432,
+      "grad_norm": 39.71334457397461,
+      "learning_rate": 9.620838571261993e-07,
+      "logits/chosen": -0.6781237125396729,
+      "logits/rejected": -0.6576768159866333,
+      "logps/chosen": -129.8390350341797,
+      "logps/rejected": -101.4920425415039,
+      "loss": 0.455,
+      "rewards/accuracies": 0.8093750476837158,
+      "rewards/chosen": 0.32165372371673584,
+      "rewards/margins": 1.1231375932693481,
+      "rewards/rejected": -0.8014839291572571,
+      "step": 370
+    },
+    {
+      "epoch": 0.21883098186006333,
+      "grad_norm": 18.181711196899414,
+      "learning_rate": 9.581518544309992e-07,
+      "logits/chosen": -0.6723761558532715,
+      "logits/rejected": -0.6443379521369934,
+      "logps/chosen": -118.37055969238281,
+      "logps/rejected": -89.48719787597656,
+      "loss": 0.4255,
+      "rewards/accuracies": 0.809374988079071,
+      "rewards/chosen": 0.2892196774482727,
+      "rewards/margins": 1.1490939855575562,
+      "rewards/rejected": -0.8598741888999939,
+      "step": 380
+    },
+    {
+      "epoch": 0.22458969190901237,
+      "grad_norm": 21.584299087524414,
+      "learning_rate": 9.540347644372052e-07,
+      "logits/chosen": -0.6740597486495972,
+      "logits/rejected": -0.6414741277694702,
+      "logps/chosen": -124.53128814697266,
+      "logps/rejected": -91.60765838623047,
+      "loss": 0.4104,
+      "rewards/accuracies": 0.8187500834465027,
+      "rewards/chosen": 0.2165013551712036,
+      "rewards/margins": 1.1881834268569946,
+      "rewards/rejected": -0.971682071685791,
+      "step": 390
+    },
+    {
+      "epoch": 0.2303484019579614,
+      "grad_norm": 24.072954177856445,
+      "learning_rate": 9.497342503948025e-07,
+      "logits/chosen": -0.656568169593811,
+      "logits/rejected": -0.6270979046821594,
+      "logps/chosen": -142.2841339111328,
+      "logps/rejected": -107.37252044677734,
+      "loss": 0.3876,
+      "rewards/accuracies": 0.8343750238418579,
+      "rewards/chosen": 0.4219643473625183,
+      "rewards/margins": 1.381853699684143,
+      "rewards/rejected": -0.9598893523216248,
+      "step": 400
+    },
+    {
+      "epoch": 0.23610711200691045,
+      "grad_norm": 21.61984634399414,
+      "learning_rate": 9.452520496546692e-07,
+      "logits/chosen": -0.6732717156410217,
+      "logits/rejected": -0.6305854916572571,
+      "logps/chosen": -149.52734375,
+      "logps/rejected": -107.50003814697266,
+      "loss": 0.3872,
+      "rewards/accuracies": 0.8187500238418579,
+      "rewards/chosen": 0.560570478439331,
+      "rewards/margins": 1.6208797693252563,
+      "rewards/rejected": -1.0603091716766357,
+      "step": 410
+    },
+    {
+      "epoch": 0.2418658220558595,
+      "grad_norm": 17.159812927246094,
+      "learning_rate": 9.405899729667103e-07,
+      "logits/chosen": -0.6888399720191956,
+      "logits/rejected": -0.663253903388977,
+      "logps/chosen": -126.42945861816406,
+      "logps/rejected": -115.0486068725586,
+      "loss": 0.5476,
+      "rewards/accuracies": 0.8031249642372131,
+      "rewards/chosen": 0.2148836851119995,
+      "rewards/margins": 1.1695863008499146,
+      "rewards/rejected": -0.954702615737915,
+      "step": 420
+    },
+    {
+      "epoch": 0.24762453210480853,
+      "grad_norm": 46.91326904296875,
+      "learning_rate": 9.357499037483376e-07,
+      "logits/chosen": -0.7161371111869812,
+      "logits/rejected": -0.684617817401886,
+      "logps/chosen": -141.0438232421875,
+      "logps/rejected": -95.7398910522461,
+      "loss": 0.407,
+      "rewards/accuracies": 0.846875011920929,
+      "rewards/chosen": 0.252617210149765,
+      "rewards/margins": 1.3679397106170654,
+      "rewards/rejected": -1.115322470664978,
+      "step": 430
+    },
+    {
+      "epoch": 0.25338324215375757,
+      "grad_norm": 12.841931343078613,
+      "learning_rate": 9.307337973235949e-07,
+      "logits/chosen": -0.708075225353241,
+      "logits/rejected": -0.6804291605949402,
+      "logps/chosen": -134.2688751220703,
+      "logps/rejected": -88.66293334960938,
+      "loss": 0.4054,
+      "rewards/accuracies": 0.8281249403953552,
+      "rewards/chosen": 0.3387967050075531,
+      "rewards/margins": 1.3938863277435303,
+      "rewards/rejected": -1.0550895929336548,
+      "step": 440
+    },
+    {
+      "epoch": 0.2591419522027066,
+      "grad_norm": 29.414453506469727,
+      "learning_rate": 9.255436801332324e-07,
+      "logits/chosen": -0.7155076265335083,
+      "logits/rejected": -0.6844953894615173,
+      "logps/chosen": -128.78543090820312,
+      "logps/rejected": -87.72601318359375,
+      "loss": 0.4451,
+      "rewards/accuracies": 0.8218749761581421,
+      "rewards/chosen": 0.26765358448028564,
+      "rewards/margins": 1.3901262283325195,
+      "rewards/rejected": -1.1224726438522339,
+      "step": 450
+    },
+    {
+      "epoch": 0.26490066225165565,
+      "grad_norm": 39.492164611816406,
+      "learning_rate": 9.201816489160516e-07,
+      "logits/chosen": -0.7237324118614197,
+      "logits/rejected": -0.705731987953186,
+      "logps/chosen": -134.49008178710938,
+      "logps/rejected": -108.36016845703125,
+      "loss": 0.4056,
+      "rewards/accuracies": 0.8375000357627869,
+      "rewards/chosen": 0.34749743342399597,
+      "rewards/margins": 1.4375081062316895,
+      "rewards/rejected": -1.090010643005371,
+      "step": 460
+    },
+    {
+      "epoch": 0.27065937230060466,
+      "grad_norm": 10.968271255493164,
+      "learning_rate": 9.146498698618506e-07,
+      "logits/chosen": -0.7146254777908325,
+      "logits/rejected": -0.6773239374160767,
+      "logps/chosen": -142.56112670898438,
+      "logps/rejected": -97.36111450195312,
+      "loss": 0.385,
+      "rewards/accuracies": 0.8343750238418579,
+      "rewards/chosen": 0.3780550956726074,
+      "rewards/margins": 1.5382754802703857,
+      "rewards/rejected": -1.1602205038070679,
+      "step": 470
+    },
+    {
+      "epoch": 0.2764180823495537,
+      "grad_norm": 20.968557357788086,
+      "learning_rate": 9.089505777363112e-07,
+      "logits/chosen": -0.7134730219841003,
+      "logits/rejected": -0.6850117444992065,
+      "logps/chosen": -139.8096923828125,
+      "logps/rejected": -98.57595825195312,
+      "loss": 0.4562,
+      "rewards/accuracies": 0.7906249761581421,
+      "rewards/chosen": 0.22018523514270782,
+      "rewards/margins": 1.3545597791671753,
+      "rewards/rejected": -1.1343746185302734,
+      "step": 480
+    },
+    {
+      "epoch": 0.28217679239850274,
+      "grad_norm": 19.068077087402344,
+      "learning_rate": 9.030860749781846e-07,
+      "logits/chosen": -0.7155986428260803,
+      "logits/rejected": -0.6849885582923889,
+      "logps/chosen": -138.48265075683594,
+      "logps/rejected": -111.72834014892578,
+      "loss": 0.4645,
+      "rewards/accuracies": 0.7906250357627869,
+      "rewards/chosen": 0.24146807193756104,
+      "rewards/margins": 1.4533673524856567,
+      "rewards/rejected": -1.2118991613388062,
+      "step": 490
+    },
+    {
+      "epoch": 0.28793550244745175,
+      "grad_norm": 24.932621002197266,
+      "learning_rate": 8.970587307691355e-07,
+      "logits/chosen": -0.6751305460929871,
+      "logits/rejected": -0.6510910987854004,
+      "logps/chosen": -137.3748321533203,
+      "logps/rejected": -99.03021240234375,
+      "loss": 0.5792,
+      "rewards/accuracies": 0.8187500238418579,
+      "rewards/chosen": 0.15273617208003998,
+      "rewards/margins": 1.3070764541625977,
+      "rewards/rejected": -1.1543402671813965,
+      "step": 500
+    },
+    {
+      "epoch": 0.2936942124964008,
+      "grad_norm": 27.17708969116211,
+      "learning_rate": 8.908709800766236e-07,
+      "logits/chosen": -0.6968642473220825,
+      "logits/rejected": -0.6686012744903564,
+      "logps/chosen": -134.98231506347656,
+      "logps/rejected": -92.57615661621094,
+      "loss": 0.4038,
+      "rewards/accuracies": 0.8156250715255737,
+      "rewards/chosen": 0.13865375518798828,
+      "rewards/margins": 1.3224716186523438,
+      "rewards/rejected": -1.1838178634643555,
+      "step": 510
+    },
+    {
+      "epoch": 0.29945292254534983,
+      "grad_norm": 14.130817413330078,
+      "learning_rate": 8.845253226702103e-07,
+      "logits/chosen": -0.6665252447128296,
+      "logits/rejected": -0.6443254947662354,
+      "logps/chosen": -136.849609375,
+      "logps/rejected": -116.30662536621094,
+      "loss": 0.4034,
+      "rewards/accuracies": 0.84375,
+      "rewards/chosen": 0.3671894669532776,
+      "rewards/margins": 1.4308054447174072,
+      "rewards/rejected": -1.0636159181594849,
+      "step": 520
+    },
+    {
+      "epoch": 0.3052116325942989,
+      "grad_norm": 19.757686614990234,
+      "learning_rate": 8.780243221116837e-07,
+      "logits/chosen": -0.7094990015029907,
+      "logits/rejected": -0.6770952939987183,
+      "logps/chosen": -124.83984375,
+      "logps/rejected": -87.22465515136719,
+      "loss": 0.3836,
+      "rewards/accuracies": 0.8437500596046448,
+      "rewards/chosen": 0.23798859119415283,
+      "rewards/margins": 1.3941946029663086,
+      "rewards/rejected": -1.1562058925628662,
+      "step": 530
+    },
+    {
+      "epoch": 0.3109703426432479,
+      "grad_norm": 24.308874130249023,
+      "learning_rate": 8.713706047194135e-07,
+      "logits/chosen": -0.6967591047286987,
+      "logits/rejected": -0.6650117635726929,
+      "logps/chosen": -136.66961669921875,
+      "logps/rejected": -92.86897277832031,
+      "loss": 0.4288,
+      "rewards/accuracies": 0.8031250238418579,
+      "rewards/chosen": 0.1572980284690857,
+      "rewards/margins": 1.3664262294769287,
+      "rewards/rejected": -1.2091283798217773,
+      "step": 540
+    },
+    {
+      "epoch": 0.3167290526921969,
+      "grad_norm": 22.169261932373047,
+      "learning_rate": 8.645668585073538e-07,
+      "logits/chosen": -0.7082847356796265,
+      "logits/rejected": -0.6733636260032654,
+      "logps/chosen": -123.37152099609375,
+      "logps/rejected": -85.6908187866211,
+      "loss": 0.3699,
+      "rewards/accuracies": 0.859375,
+      "rewards/chosen": 0.24349649250507355,
+      "rewards/margins": 1.6746883392333984,
+      "rewards/rejected": -1.4311916828155518,
+      "step": 550
+    },
+    {
+      "epoch": 0.322487762741146,
+      "grad_norm": 16.986181259155273,
+      "learning_rate": 8.576158320991204e-07,
+      "logits/chosen": -0.7158868908882141,
+      "logits/rejected": -0.6811345815658569,
+      "logps/chosen": -135.4944610595703,
+      "logps/rejected": -91.58667755126953,
+      "loss": 0.4072,
+      "rewards/accuracies": 0.7968749403953552,
+      "rewards/chosen": 0.31856077909469604,
+      "rewards/margins": 1.5791938304901123,
+      "rewards/rejected": -1.2606329917907715,
+      "step": 560
+    },
+    {
+      "epoch": 0.328246472790095,
+      "grad_norm": 17.154346466064453,
+      "learning_rate": 8.505203336175835e-07,
+      "logits/chosen": -0.70228111743927,
+      "logits/rejected": -0.6674139499664307,
+      "logps/chosen": -140.62643432617188,
+      "logps/rejected": -104.01993560791016,
+      "loss": 0.4011,
+      "rewards/accuracies": 0.809374988079071,
+      "rewards/chosen": 0.26829949021339417,
+      "rewards/margins": 1.6834442615509033,
+      "rewards/rejected": -1.415144920349121,
+      "step": 570
+    },
+    {
+      "epoch": 0.33400518283904407,
+      "grad_norm": 44.71592712402344,
+      "learning_rate": 8.432832295504223e-07,
+      "logits/chosen": -0.7484235763549805,
+      "logits/rejected": -0.7252366542816162,
+      "logps/chosen": -130.38858032226562,
+      "logps/rejected": -97.56017303466797,
+      "loss": 0.4027,
+      "rewards/accuracies": 0.824999988079071,
+      "rewards/chosen": 0.21075472235679626,
+      "rewards/margins": 1.5023763179779053,
+      "rewards/rejected": -1.2916215658187866,
+      "step": 580
+    },
+    {
+      "epoch": 0.3397638928879931,
+      "grad_norm": 18.29545021057129,
+      "learning_rate": 8.359074435921031e-07,
+      "logits/chosen": -0.7082387208938599,
+      "logits/rejected": -0.672591507434845,
+      "logps/chosen": -140.56137084960938,
+      "logps/rejected": -97.13379669189453,
+      "loss": 0.3661,
+      "rewards/accuracies": 0.8562499284744263,
+      "rewards/chosen": 0.36341509222984314,
+      "rewards/margins": 1.8082225322723389,
+      "rewards/rejected": -1.4448072910308838,
+      "step": 590
+    },
+    {
+      "epoch": 0.34552260293694215,
+      "grad_norm": 14.033991813659668,
+      "learning_rate": 8.283959554627446e-07,
+      "logits/chosen": -0.716846764087677,
+      "logits/rejected": -0.6769986748695374,
+      "logps/chosen": -142.1970977783203,
+      "logps/rejected": -102.24588775634766,
+      "loss": 0.3741,
+      "rewards/accuracies": 0.8625000715255737,
+      "rewards/chosen": 0.10098960995674133,
+      "rewards/margins": 1.6507571935653687,
+      "rewards/rejected": -1.5497674942016602,
+      "step": 600
+    },
+    {
+      "epoch": 0.35128131298589116,
+      "grad_norm": 13.405789375305176,
+      "learning_rate": 8.207517997043504e-07,
+      "logits/chosen": -0.7140165567398071,
+      "logits/rejected": -0.6859545111656189,
+      "logps/chosen": -144.771240234375,
+      "logps/rejected": -104.20320892333984,
+      "loss": 0.3605,
+      "rewards/accuracies": 0.8343750238418579,
+      "rewards/chosen": 0.20804497599601746,
+      "rewards/margins": 1.7361663579940796,
+      "rewards/rejected": -1.5281215906143188,
+      "step": 610
+    },
+    {
+      "epoch": 0.35704002303484017,
+      "grad_norm": 15.424454689025879,
+      "learning_rate": 8.129780644548938e-07,
+      "logits/chosen": -0.7223767042160034,
+      "logits/rejected": -0.6883643269538879,
+      "logps/chosen": -130.39566040039062,
+      "logps/rejected": -103.71538543701172,
+      "loss": 0.3571,
+      "rewards/accuracies": 0.8375000357627869,
+      "rewards/chosen": 0.09431158006191254,
+      "rewards/margins": 1.7663012742996216,
+      "rewards/rejected": -1.6719896793365479,
+      "step": 620
+    },
+    {
+      "epoch": 0.36279873308378924,
+      "grad_norm": 15.335771560668945,
+      "learning_rate": 8.05077890200752e-07,
+      "logits/chosen": -0.7171313166618347,
+      "logits/rejected": -0.6881922483444214,
+      "logps/chosen": -151.4913787841797,
+      "logps/rejected": -98.9349594116211,
+      "loss": 0.3646,
+      "rewards/accuracies": 0.8312499523162842,
+      "rewards/chosen": 0.1370360106229782,
+      "rewards/margins": 1.744744062423706,
+      "rewards/rejected": -1.607708215713501,
+      "step": 630
+    },
+    {
+      "epoch": 0.36855744313273825,
+      "grad_norm": 17.60926628112793,
+      "learning_rate": 7.970544685079894e-07,
+      "logits/chosen": -0.6996970772743225,
+      "logits/rejected": -0.6725396513938904,
+      "logps/chosen": -138.51010131835938,
+      "logps/rejected": -98.81480407714844,
+      "loss": 0.3743,
+      "rewards/accuracies": 0.8375000357627869,
+      "rewards/chosen": 0.06485229730606079,
+      "rewards/margins": 1.6162770986557007,
+      "rewards/rejected": -1.5514247417449951,
+      "step": 640
+    },
+    {
+      "epoch": 0.3743161531816873,
+      "grad_norm": 14.755389213562012,
+      "learning_rate": 7.889110407330083e-07,
+      "logits/chosen": -0.7361860275268555,
+      "logits/rejected": -0.7013921737670898,
+      "logps/chosen": -141.9473876953125,
+      "logps/rejected": -101.18817138671875,
+      "loss": 0.3477,
+      "rewards/accuracies": 0.8343750238418579,
+      "rewards/chosen": 0.34291204810142517,
+      "rewards/margins": 1.7922308444976807,
+      "rewards/rejected": -1.4493186473846436,
+      "step": 650
+    },
+    {
+      "epoch": 0.38007486323063633,
+      "grad_norm": 20.350278854370117,
+      "learning_rate": 7.806508967130836e-07,
+      "logits/chosen": -0.7531692981719971,
+      "logits/rejected": -0.7267601490020752,
+      "logps/chosen": -134.09832763671875,
+      "logps/rejected": -99.92311096191406,
+      "loss": 0.3806,
+      "rewards/accuracies": 0.828125,
+      "rewards/chosen": 0.0986885130405426,
+      "rewards/margins": 1.6266651153564453,
+      "rewards/rejected": -1.5279765129089355,
+      "step": 660
+    },
+    {
+      "epoch": 0.3858335732795854,
+      "grad_norm": 20.015928268432617,
+      "learning_rate": 7.722773734373113e-07,
+      "logits/chosen": -0.7297524213790894,
+      "logits/rejected": -0.6996749639511108,
+      "logps/chosen": -136.6656951904297,
+      "logps/rejected": -103.38198852539062,
+      "loss": 0.3441,
+      "rewards/accuracies": 0.8687500357627869,
+      "rewards/chosen": 0.19155237078666687,
+      "rewards/margins": 1.881716012954712,
+      "rewards/rejected": -1.690163493156433,
+      "step": 670
+    },
+    {
+      "epoch": 0.3915922833285344,
+      "grad_norm": 15.462444305419922,
+      "learning_rate": 7.637938536985099e-07,
+      "logits/chosen": -0.753920316696167,
+      "logits/rejected": -0.7297205924987793,
+      "logps/chosen": -129.48484802246094,
+      "logps/rejected": -100.32254028320312,
+      "loss": 0.406,
+      "rewards/accuracies": 0.828125,
+      "rewards/chosen": 0.016594011336565018,
+      "rewards/margins": 1.5454182624816895,
+      "rewards/rejected": -1.5288242101669312,
+      "step": 680
+    },
+    {
+      "epoch": 0.3973509933774834,
+      "grad_norm": 11.136063575744629,
+      "learning_rate": 7.552037647266157e-07,
+      "logits/chosen": -0.7570206522941589,
+      "logits/rejected": -0.7179579734802246,
+      "logps/chosen": -144.1813507080078,
+      "logps/rejected": -103.70523834228516,
+      "loss": 0.3538,
+      "rewards/accuracies": 0.84375,
+      "rewards/chosen": 0.2709888219833374,
+      "rewards/margins": 1.8022199869155884,
+      "rewards/rejected": -1.5312312841415405,
+      "step": 690
+    },
+    {
+      "epoch": 0.4031097034264325,
+      "grad_norm": 29.570873260498047,
+      "learning_rate": 7.465105768041282e-07,
+      "logits/chosen": -0.7303261160850525,
+      "logits/rejected": -0.6878946423530579,
+      "logps/chosen": -146.9130401611328,
+      "logps/rejected": -101.49455261230469,
+      "loss": 0.3586,
+      "rewards/accuracies": 0.84375,
+      "rewards/chosen": 0.23865552246570587,
+      "rewards/margins": 1.850990891456604,
+      "rewards/rejected": -1.6123353242874146,
+      "step": 700
+    },
+    {
+      "epoch": 0.4088684134753815,
+      "grad_norm": 14.321788787841797,
+      "learning_rate": 7.377178018641613e-07,
+      "logits/chosen": -0.7625525593757629,
+      "logits/rejected": -0.7228215932846069,
+      "logps/chosen": -122.93366241455078,
+      "logps/rejected": -191.53184509277344,
+      "loss": 0.3407,
+      "rewards/accuracies": 0.84375,
+      "rewards/chosen": 0.17545101046562195,
+      "rewards/margins": 2.5035836696624756,
+      "rewards/rejected": -2.3281326293945312,
+      "step": 710
+    },
+    {
+      "epoch": 0.41462712352433057,
+      "grad_norm": 26.58182144165039,
+      "learning_rate": 7.288289920716685e-07,
+      "logits/chosen": -0.7637885808944702,
+      "logits/rejected": -0.7382663488388062,
+      "logps/chosen": -113.6340560913086,
+      "logps/rejected": -94.61100769042969,
+      "loss": 0.3549,
+      "rewards/accuracies": 0.8593750596046448,
+      "rewards/chosen": -0.030434802174568176,
+      "rewards/margins": 1.6987662315368652,
+      "rewards/rejected": -1.7292009592056274,
+      "step": 720
+    },
+    {
+      "epoch": 0.4203858335732796,
+      "grad_norm": 20.130050659179688,
+      "learning_rate": 7.198477383884161e-07,
+      "logits/chosen": -0.7368783950805664,
+      "logits/rejected": -0.7132025957107544,
+      "logps/chosen": -129.5439910888672,
+      "logps/rejected": -99.97601318359375,
+      "loss": 0.3928,
+      "rewards/accuracies": 0.8218749761581421,
+      "rewards/chosen": 0.08906654268503189,
+      "rewards/margins": 1.758063793182373,
+      "rewards/rejected": -1.6689971685409546,
+      "step": 730
+    },
+    {
+      "epoch": 0.42614454362222864,
+      "grad_norm": 15.920429229736328,
+      "learning_rate": 7.107776691222802e-07,
+      "logits/chosen": -0.7325925827026367,
+      "logits/rejected": -0.7019511461257935,
+      "logps/chosen": -119.99696350097656,
+      "logps/rejected": -93.21834564208984,
+      "loss": 0.4016,
+      "rewards/accuracies": 0.8187500238418579,
+      "rewards/chosen": 0.09440048038959503,
+      "rewards/margins": 1.7340251207351685,
+      "rewards/rejected": -1.639624834060669,
+      "step": 740
+    },
+    {
+      "epoch": 0.43190325367117766,
+      "grad_norm": 21.42776107788086,
+      "learning_rate": 7.016224484614608e-07,
+      "logits/chosen": -0.6936213970184326,
+      "logits/rejected": -0.6615394949913025,
+      "logps/chosen": -131.905029296875,
+      "logps/rejected": -94.2560806274414,
+      "loss": 0.4054,
+      "rewards/accuracies": 0.8187499642372131,
+      "rewards/chosen": 0.23300418257713318,
+      "rewards/margins": 1.7516708374023438,
+      "rewards/rejected": -1.5186666250228882,
+      "step": 750
+    },
+    {
+      "epoch": 0.43766196372012667,
+      "grad_norm": 15.192195892333984,
+      "learning_rate": 6.923857749941959e-07,
+      "logits/chosen": -0.7206861972808838,
+      "logits/rejected": -0.6929584741592407,
+      "logps/chosen": -129.8375701904297,
+      "logps/rejected": -100.96007537841797,
+      "loss": 0.3734,
+      "rewards/accuracies": 0.8156249523162842,
+      "rewards/chosen": 0.22579814493656158,
+      "rewards/margins": 1.7992628812789917,
+      "rewards/rejected": -1.5734646320343018,
+      "step": 760
+    },
+    {
+      "epoch": 0.44342067376907573,
+      "grad_norm": 10.402118682861328,
+      "learning_rate": 6.830713802145818e-07,
+      "logits/chosen": -0.6799423098564148,
+      "logits/rejected": -0.6525387763977051,
+      "logps/chosen": -154.219482421875,
+      "logps/rejected": -114.71685028076172,
+      "loss": 0.4382,
+      "rewards/accuracies": 0.8281250596046448,
+      "rewards/chosen": 0.4146007299423218,
+      "rewards/margins": 1.8844369649887085,
+      "rewards/rejected": -1.4698363542556763,
+      "step": 770
+    },
+    {
+      "epoch": 0.44917938381802475,
+      "grad_norm": 10.21727466583252,
+      "learning_rate": 6.736830270150991e-07,
+      "logits/chosen": -0.6829798221588135,
+      "logits/rejected": -0.6589241623878479,
+      "logps/chosen": -121.01963806152344,
+      "logps/rejected": -97.19430541992188,
+      "loss": 0.3348,
+      "rewards/accuracies": 0.8499999642372131,
+      "rewards/chosen": 0.27119049429893494,
+      "rewards/margins": 1.897121548652649,
+      "rewards/rejected": -1.6259310245513916,
+      "step": 780
+    },
+    {
+      "epoch": 0.4549380938669738,
+      "grad_norm": 20.992265701293945,
+      "learning_rate": 6.642245081664522e-07,
+      "logits/chosen": -0.6706294417381287,
+      "logits/rejected": -0.6368067860603333,
+      "logps/chosen": -137.04710388183594,
+      "logps/rejected": -98.4258041381836,
+      "loss": 0.386,
+      "rewards/accuracies": 0.8375000357627869,
+      "rewards/chosen": 0.18035945296287537,
+      "rewards/margins": 1.880584955215454,
+      "rewards/rejected": -1.7002257108688354,
+      "step": 790
+    },
+    {
+      "epoch": 0.4606968039159228,
+      "grad_norm": 18.026098251342773,
+      "learning_rate": 6.54699644785342e-07,
+      "logits/chosen": -0.6860750913619995,
+      "logits/rejected": -0.6613771319389343,
+      "logps/chosen": -136.12977600097656,
+      "logps/rejected": -94.88777160644531,
+      "loss": 0.374,
+      "rewards/accuracies": 0.840624988079071,
+      "rewards/chosen": 0.26585477590560913,
+      "rewards/margins": 1.7964979410171509,
+      "rewards/rejected": -1.5306432247161865,
+      "step": 800
+    },
+    {
+      "epoch": 0.4664555139648719,
+      "grad_norm": 12.507932662963867,
+      "learning_rate": 6.451122847907842e-07,
+      "logits/chosen": -0.6504511833190918,
+      "logits/rejected": -0.6260837912559509,
+      "logps/chosen": -130.69212341308594,
+      "logps/rejected": -99.03807830810547,
+      "loss": 0.3886,
+      "rewards/accuracies": 0.8468750715255737,
+      "rewards/chosen": 0.2847888469696045,
+      "rewards/margins": 1.795453667640686,
+      "rewards/rejected": -1.5106645822525024,
+      "step": 810
+    },
+    {
+      "epoch": 0.4722142240138209,
+      "grad_norm": 21.029708862304688,
+      "learning_rate": 6.354663013496005e-07,
+      "logits/chosen": -0.6501750946044922,
+      "logits/rejected": -0.6221209764480591,
+      "logps/chosen": -147.17929077148438,
+      "logps/rejected": -126.30306243896484,
+      "loss": 0.38,
+      "rewards/accuracies": 0.8562500476837158,
+      "rewards/chosen": 0.4969615340232849,
+      "rewards/margins": 1.870727777481079,
+      "rewards/rejected": -1.373766303062439,
+      "step": 820
+    },
+    {
+      "epoch": 0.4779729340627699,
+      "grad_norm": 15.563918113708496,
+      "learning_rate": 6.257655913117117e-07,
+      "logits/chosen": -0.6645382642745972,
+      "logits/rejected": -0.6453496217727661,
+      "logps/chosen": -128.99713134765625,
+      "logps/rejected": -102.27412414550781,
+      "loss": 0.3668,
+      "rewards/accuracies": 0.8531250357627869,
+      "rewards/chosen": 0.3891734480857849,
+      "rewards/margins": 1.8439801931381226,
+      "rewards/rejected": -1.4548066854476929,
+      "step": 830
+    },
+    {
+      "epoch": 0.483731644111719,
+      "grad_norm": 9.265098571777344,
+      "learning_rate": 6.160140736358599e-07,
+      "logits/chosen": -0.7049685120582581,
+      "logits/rejected": -0.6709137558937073,
+      "logps/chosen": -139.44696044921875,
+      "logps/rejected": -107.42019653320312,
+      "loss": 0.3237,
+      "rewards/accuracies": 0.8656250238418579,
+      "rewards/chosen": 0.42667171359062195,
+      "rewards/margins": 1.9013292789459229,
+      "rewards/rejected": -1.4746575355529785,
+      "step": 840
+    },
+    {
+      "epoch": 0.489490354160668,
+      "grad_norm": 15.329190254211426,
+      "learning_rate": 6.062156878064025e-07,
+      "logits/chosen": -0.6818621158599854,
+      "logits/rejected": -0.6550964713096619,
+      "logps/chosen": -122.67804718017578,
+      "logps/rejected": -90.76681518554688,
+      "loss": 0.3436,
+      "rewards/accuracies": 0.8437500596046448,
+      "rewards/chosen": 0.35221126675605774,
+      "rewards/margins": 1.8415284156799316,
+      "rewards/rejected": -1.4893171787261963,
+      "step": 850
+    },
+    {
+      "epoch": 0.49524906420961706,
+      "grad_norm": 11.879908561706543,
+      "learning_rate": 5.963743922418121e-07,
+      "logits/chosen": -0.6924838423728943,
+      "logits/rejected": -0.6654370427131653,
+      "logps/chosen": -131.25909423828125,
+      "logps/rejected": -95.65048217773438,
+      "loss": 0.3591,
+      "rewards/accuracies": 0.846875011920929,
+      "rewards/chosen": 0.36284345388412476,
+      "rewards/margins": 1.8446120023727417,
+      "rewards/rejected": -1.4817683696746826,
+      "step": 860
+    },
+    {
+      "epoch": 0.5010077742585661,
+      "grad_norm": 24.157817840576172,
+      "learning_rate": 5.864941626955274e-07,
+      "logits/chosen": -0.7024100422859192,
+      "logits/rejected": -0.6722136735916138,
+      "logps/chosen": -128.6143035888672,
+      "logps/rejected": -91.6454849243164,
+      "loss": 0.3274,
+      "rewards/accuracies": 0.856249988079071,
+      "rewards/chosen": 0.4442964494228363,
+      "rewards/margins": 1.9124963283538818,
+      "rewards/rejected": -1.4681998491287231,
+      "step": 870
+    },
+    {
+      "epoch": 0.5067664843075151,
+      "grad_norm": 24.65102767944336,
+      "learning_rate": 5.765789906498015e-07,
+      "logits/chosen": -0.6772369146347046,
+      "logits/rejected": -0.6518146991729736,
+      "logps/chosen": -120.58905792236328,
+      "logps/rejected": -93.23050689697266,
+      "loss": 0.4348,
+      "rewards/accuracies": 0.7968749403953552,
+      "rewards/chosen": 0.1704983413219452,
+      "rewards/margins": 1.723240852355957,
+      "rewards/rejected": -1.5527423620224,
+      "step": 880
+    },
+    {
+      "epoch": 0.5125251943564642,
+      "grad_norm": 21.338768005371094,
+      "learning_rate": 5.666328817031957e-07,
+      "logits/chosen": -0.6858566403388977,
+      "logits/rejected": -0.6644458770751953,
+      "logps/chosen": -143.04592895507812,
+      "logps/rejected": -112.9994125366211,
+      "loss": 0.3478,
+      "rewards/accuracies": 0.8374999761581421,
+      "rewards/chosen": 0.39247894287109375,
+      "rewards/margins": 1.9909181594848633,
+      "rewards/rejected": -1.598439335823059,
+      "step": 890
+    },
+    {
+      "epoch": 0.5182839044054132,
+      "grad_norm": 22.700468063354492,
+      "learning_rate": 5.56659853952371e-07,
+      "logits/chosen": -0.6734040975570679,
+      "logits/rejected": -0.6462678909301758,
+      "logps/chosen": -126.01319885253906,
+      "logps/rejected": -101.61735534667969,
+      "loss": 0.3346,
+      "rewards/accuracies": 0.8406249284744263,
+      "rewards/chosen": 0.2884215712547302,
+      "rewards/margins": 1.878092646598816,
+      "rewards/rejected": -1.589671015739441,
+      "step": 900
+    },
+    {
+      "epoch": 0.5240426144543622,
+      "grad_norm": 14.079261779785156,
+      "learning_rate": 5.466639363688295e-07,
+      "logits/chosen": -0.6971774101257324,
+      "logits/rejected": -0.6686007976531982,
+      "logps/chosen": -136.96533203125,
+      "logps/rejected": -98.08868408203125,
+      "loss": 0.3788,
+      "rewards/accuracies": 0.856249988079071,
+      "rewards/chosen": 0.458240270614624,
+      "rewards/margins": 1.9593617916107178,
+      "rewards/rejected": -1.5011215209960938,
+      "step": 910
+    },
+    {
+      "epoch": 0.5298013245033113,
+      "grad_norm": 9.099288940429688,
+      "learning_rate": 5.366491671712641e-07,
+      "logits/chosen": -0.6674658060073853,
+      "logits/rejected": -0.6430867910385132,
+      "logps/chosen": -194.9149169921875,
+      "logps/rejected": -105.85720825195312,
+      "loss": 0.4739,
+      "rewards/accuracies": 0.8687500357627869,
+      "rewards/chosen": 0.2768912613391876,
+      "rewards/margins": 1.7589211463928223,
+      "rewards/rejected": -1.482029914855957,
+      "step": 920
+    },
+    {
+      "epoch": 0.5355600345522603,
+      "grad_norm": 15.008882522583008,
+      "learning_rate": 5.266195921941696e-07,
+      "logits/chosen": -0.6785784363746643,
+      "logits/rejected": -0.6494132280349731,
+      "logps/chosen": -132.4415740966797,
+      "logps/rejected": -93.6052017211914,
+      "loss": 0.2967,
+      "rewards/accuracies": 0.890625,
+      "rewards/chosen": 0.5966256260871887,
+      "rewards/margins": 2.0661702156066895,
+      "rewards/rejected": -1.469544768333435,
+      "step": 930
+    },
+    {
+      "epoch": 0.5413187446012093,
+      "grad_norm": 15.97289752960205,
+      "learning_rate": 5.165792632533811e-07,
+      "logits/chosen": -0.6924957036972046,
+      "logits/rejected": -0.6638644337654114,
+      "logps/chosen": -139.21484375,
+      "logps/rejected": -106.23296356201172,
+      "loss": 0.375,
+      "rewards/accuracies": 0.8312499523162842,
+      "rewards/chosen": 0.31263190507888794,
+      "rewards/margins": 1.9060099124908447,
+      "rewards/rejected": -1.593377947807312,
+      "step": 940
+    },
+    {
+      "epoch": 0.5470774546501583,
+      "grad_norm": 12.529922485351562,
+      "learning_rate": 5.065322365091928e-07,
+      "logits/chosen": -0.6531215310096741,
+      "logits/rejected": -0.6270167231559753,
+      "logps/chosen": -126.1975326538086,
+      "logps/rejected": -94.7371826171875,
+      "loss": 0.3378,
+      "rewards/accuracies": 0.8687499761581421,
+      "rewards/chosen": 0.16207355260849,
+      "rewards/margins": 1.917543649673462,
+      "rewards/rejected": -1.7554700374603271,
+      "step": 950
+    },
+    {
+      "epoch": 0.5528361646991073,
+      "grad_norm": 12.316908836364746,
+      "learning_rate": 4.964825708277229e-07,
+      "logits/chosen": -0.6698362231254578,
+      "logits/rejected": -0.6397368311882019,
+      "logps/chosen": -142.1855926513672,
+      "logps/rejected": -107.25379943847656,
+      "loss": 0.3183,
+      "rewards/accuracies": 0.856249988079071,
+      "rewards/chosen": 0.49424219131469727,
+      "rewards/margins": 2.0777647495269775,
+      "rewards/rejected": -1.5835226774215698,
+      "step": 960
+    },
+    {
+      "epoch": 0.5585948747480565,
+      "grad_norm": 20.065715789794922,
+      "learning_rate": 4.864343261411856e-07,
+      "logits/chosen": -0.706594705581665,
+      "logits/rejected": -0.6800837516784668,
+      "logps/chosen": -136.40191650390625,
+      "logps/rejected": -96.95333862304688,
+      "loss": 0.338,
+      "rewards/accuracies": 0.8562500476837158,
+      "rewards/chosen": 0.36437052488327026,
+      "rewards/margins": 2.0314860343933105,
+      "rewards/rejected": -1.6671154499053955,
+      "step": 970
+    },
+    {
+      "epoch": 0.5643535847970055,
+      "grad_norm": 15.975777626037598,
+      "learning_rate": 4.763915618077319e-07,
+      "logits/chosen": -0.7158640623092651,
+      "logits/rejected": -0.6942066550254822,
+      "logps/chosen": -131.8379669189453,
+      "logps/rejected": -107.44449615478516,
+      "loss": 0.3128,
+      "rewards/accuracies": 0.856249988079071,
+      "rewards/chosen": 0.2980409562587738,
+      "rewards/margins": 2.096712827682495,
+      "rewards/rejected": -1.798672080039978,
+      "step": 980
+    },
+    {
+      "epoch": 0.5701122948459545,
+      "grad_norm": 17.336626052856445,
+      "learning_rate": 4.6635833497152217e-07,
+      "logits/chosen": -0.7176687121391296,
+      "logits/rejected": -0.6980822086334229,
+      "logps/chosen": -134.8371124267578,
+      "logps/rejected": -111.04306030273438,
+      "loss": 0.3319,
+      "rewards/accuracies": 0.859375,
+      "rewards/chosen": 0.4131332337856293,
+      "rewards/margins": 2.0368969440460205,
+      "rewards/rejected": -1.6237636804580688,
+      "step": 990
+    },
+    {
+      "epoch": 0.5758710048949035,
+      "grad_norm": 23.855201721191406,
+      "learning_rate": 4.5633869892369436e-07,
+      "logits/chosen": -0.6962485313415527,
+      "logits/rejected": -0.6787732839584351,
+      "logps/chosen": -142.33580017089844,
+      "logps/rejected": -106.88719940185547,
+      "loss": 0.3622,
+      "rewards/accuracies": 0.8437499403953552,
+      "rewards/chosen": 0.36459341645240784,
+      "rewards/margins": 1.871263027191162,
+      "rewards/rejected": -1.5066696405410767,
+      "step": 1000
+    },
+    {
+      "epoch": 0.5816297149438526,
+      "grad_norm": 21.980791091918945,
+      "learning_rate": 4.463367014648871e-07,
+      "logits/chosen": -0.6853346824645996,
+      "logits/rejected": -0.6605929732322693,
+      "logps/chosen": -126.45109558105469,
+      "logps/rejected": -87.18302154541016,
+      "loss": 0.3155,
+      "rewards/accuracies": 0.8687500357627869,
+      "rewards/chosen": 0.4707247316837311,
+      "rewards/margins": 2.1418490409851074,
+      "rewards/rejected": -1.6711241006851196,
+      "step": 1010
+    },
+    {
+      "epoch": 0.5873884249928016,
+      "grad_norm": 18.08453369140625,
+      "learning_rate": 4.3635638326998194e-07,
+      "logits/chosen": -0.7209752202033997,
+      "logits/rejected": -0.6910983920097351,
+      "logps/chosen": -144.9892120361328,
+      "logps/rejected": -90.99137115478516,
+      "loss": 0.3438,
+      "rewards/accuracies": 0.8812500238418579,
+      "rewards/chosen": 0.30342555046081543,
+      "rewards/margins": 2.0735397338867188,
+      "rewards/rejected": -1.7701140642166138,
+      "step": 1020
+    },
+    {
+      "epoch": 0.5931471350417507,
+      "grad_norm": 19.45905876159668,
+      "learning_rate": 4.264017762557244e-07,
+      "logits/chosen": -0.6922816634178162,
+      "logits/rejected": -0.6663939952850342,
+      "logps/chosen": -129.85174560546875,
+      "logps/rejected": -93.9735336303711,
+      "loss": 0.3056,
+      "rewards/accuracies": 0.8718750476837158,
+      "rewards/chosen": 0.280218243598938,
+      "rewards/margins": 2.1616480350494385,
+      "rewards/rejected": -1.8814294338226318,
+      "step": 1030
+    },
+    {
+      "epoch": 0.5989058450906997,
+      "grad_norm": 23.51837730407715,
+      "learning_rate": 4.1647690195188225e-07,
+      "logits/chosen": -0.7207853198051453,
+      "logits/rejected": -0.6952850818634033,
+      "logps/chosen": -129.96766662597656,
+      "logps/rejected": -103.2422103881836,
+      "loss": 0.3515,
+      "rewards/accuracies": 0.84375,
+      "rewards/chosen": 0.19336190819740295,
+      "rewards/margins": 2.06461238861084,
+      "rewards/rejected": -1.8712507486343384,
+      "step": 1040
+    },
+    {
+      "epoch": 0.6046645551396487,
+      "grad_norm": 10.306842803955078,
+      "learning_rate": 4.0658576987660154e-07,
+      "logits/chosen": -0.7156558036804199,
+      "logits/rejected": -0.6903259754180908,
+      "logps/chosen": -128.18174743652344,
+      "logps/rejected": -92.14160919189453,
+      "loss": 0.2842,
+      "rewards/accuracies": 0.90625,
+      "rewards/chosen": 0.19515244662761688,
+      "rewards/margins": 2.040672779083252,
+      "rewards/rejected": -1.845520257949829,
+      "step": 1050
+    },
+    {
+      "epoch": 0.6104232651885978,
+      "grad_norm": 30.55817413330078,
+      "learning_rate": 3.9673237591661265e-07,
+      "logits/chosen": -0.7145802974700928,
+      "logits/rejected": -0.6979396343231201,
+      "logps/chosen": -132.6157989501953,
+      "logps/rejected": -100.8826675415039,
+      "loss": 0.3722,
+      "rewards/accuracies": 0.856249988079071,
+      "rewards/chosen": 0.2976403832435608,
+      "rewards/margins": 2.033444881439209,
+      "rewards/rejected": -1.735804557800293,
+      "step": 1060
+    },
+    {
+      "epoch": 0.6161819752375468,
+      "grad_norm": 15.014007568359375,
+      "learning_rate": 3.8692070071294623e-07,
+      "logits/chosen": -0.6951994299888611,
+      "logits/rejected": -0.6723592281341553,
+      "logps/chosen": -138.485107421875,
+      "logps/rejected": -107.72498321533203,
+      "loss": 0.3226,
+      "rewards/accuracies": 0.8812500238418579,
+      "rewards/chosen": 0.159155011177063,
+      "rewards/margins": 2.102383613586426,
+      "rewards/rejected": -1.9432284832000732,
+      "step": 1070
+    },
+    {
+      "epoch": 0.6219406852864958,
+      "grad_norm": 17.77401351928711,
+      "learning_rate": 3.7715470805280694e-07,
+      "logits/chosen": -0.7224422097206116,
+      "logits/rejected": -0.7026802897453308,
+      "logps/chosen": -128.47036743164062,
+      "logps/rejected": -101.08393859863281,
+      "loss": 0.3732,
+      "rewards/accuracies": 0.8499999642372131,
+      "rewards/chosen": 0.11434885859489441,
+      "rewards/margins": 1.996582269668579,
+      "rewards/rejected": -1.8822333812713623,
+      "step": 1080
+    },
+    {
+      "epoch": 0.6276993953354448,
+      "grad_norm": 17.239843368530273,
+      "learning_rate": 3.6743834326825384e-07,
+      "logits/chosen": -0.7175098657608032,
+      "logits/rejected": -0.6943240761756897,
+      "logps/chosen": -146.55691528320312,
+      "logps/rejected": -115.63310241699219,
+      "loss": 0.34,
+      "rewards/accuracies": 0.859375,
+      "rewards/chosen": 0.2911383807659149,
+      "rewards/margins": 2.0382208824157715,
+      "rewards/rejected": -1.7470825910568237,
+      "step": 1090
+    },
+    {
+      "epoch": 0.6334581053843938,
+      "grad_norm": 14.190447807312012,
+      "learning_rate": 3.577755316423411e-07,
+      "logits/chosen": -0.7407609820365906,
+      "logits/rejected": -0.7160014510154724,
+      "logps/chosen": -133.70611572265625,
+      "logps/rejected": -110.30499267578125,
+      "loss": 0.3647,
+      "rewards/accuracies": 0.84375,
+      "rewards/chosen": 0.23839513957500458,
+      "rewards/margins": 1.9629977941513062,
+      "rewards/rejected": -1.7246026992797852,
+      "step": 1100
+    },
+    {
+      "epoch": 0.639216815433343,
+      "grad_norm": 20.399852752685547,
+      "learning_rate": 3.481701768233532e-07,
+      "logits/chosen": -0.715621829032898,
+      "logits/rejected": -0.6828228235244751,
+      "logps/chosen": -127.63819885253906,
+      "logps/rejected": -189.13275146484375,
+      "loss": 0.5705,
+      "rewards/accuracies": 0.84375,
+      "rewards/chosen": 0.16459399461746216,
+      "rewards/margins": 1.7133777141571045,
+      "rewards/rejected": -1.5487836599349976,
+      "step": 1110
+    },
+    {
+      "epoch": 0.644975525482292,
+      "grad_norm": 34.342124938964844,
+      "learning_rate": 3.386261592477832e-07,
+      "logits/chosen": -0.7274751663208008,
+      "logits/rejected": -0.7126445770263672,
+      "logps/chosen": -124.43698120117188,
+      "logps/rejected": -102.45167541503906,
+      "loss": 0.342,
+      "rewards/accuracies": 0.875,
+      "rewards/chosen": 0.3266334533691406,
+      "rewards/margins": 2.0077967643737793,
+      "rewards/rejected": -1.6811633110046387,
+      "step": 1120
+    },
+    {
+      "epoch": 0.650734235531241,
+      "grad_norm": 26.605039596557617,
+      "learning_rate": 3.2914733457268874e-07,
+      "logits/chosen": -0.7203111052513123,
+      "logits/rejected": -0.7033373713493347,
+      "logps/chosen": -120.60269165039062,
+      "logps/rejected": -101.74288940429688,
+      "loss": 0.3712,
+      "rewards/accuracies": 0.8250000476837158,
+      "rewards/chosen": 0.2722548544406891,
+      "rewards/margins": 1.9117896556854248,
+      "rewards/rejected": -1.6395349502563477,
+      "step": 1130
+    },
+    {
+      "epoch": 0.65649294558019,
+      "grad_norm": 12.683929443359375,
+      "learning_rate": 3.1973753211805593e-07,
+      "logits/chosen": -0.7631603479385376,
+      "logits/rejected": -0.7429041862487793,
+      "logps/chosen": -118.06796264648438,
+      "logps/rejected": -101.36824035644531,
+      "loss": 0.346,
+      "rewards/accuracies": 0.8718750476837158,
+      "rewards/chosen": 0.1378428041934967,
+      "rewards/margins": 1.9998619556427002,
+      "rewards/rejected": -1.8620191812515259,
+      "step": 1140
+    },
+    {
+      "epoch": 0.6622516556291391,
+      "grad_norm": 15.436881065368652,
+      "learning_rate": 3.1040055331980573e-07,
+      "logits/chosen": -0.7278124690055847,
+      "logits/rejected": -0.7010136842727661,
+      "logps/chosen": -131.94236755371094,
+      "logps/rejected": -107.63276672363281,
+      "loss": 0.3427,
+      "rewards/accuracies": 0.871874988079071,
+      "rewards/chosen": 0.30168789625167847,
+      "rewards/margins": 1.9916753768920898,
+      "rewards/rejected": -1.6899876594543457,
+      "step": 1150
+    },
+    {
+      "epoch": 0.6680103656780881,
+      "grad_norm": 16.586223602294922,
+      "learning_rate": 3.0114017019406355e-07,
+      "logits/chosen": -0.7338711023330688,
+      "logits/rejected": -0.7031553387641907,
+      "logps/chosen": -135.208984375,
+      "logps/rejected": -95.02485656738281,
+      "loss": 0.3435,
+      "rewards/accuracies": 0.8687500357627869,
+      "rewards/chosen": 0.28808894753456116,
+      "rewards/margins": 1.994765043258667,
+      "rewards/rejected": -1.7066762447357178,
+      "step": 1160
+    },
+    {
+      "epoch": 0.6737690757270371,
+      "grad_norm": 15.553061485290527,
+      "learning_rate": 2.9196012381331447e-07,
+      "logits/chosen": -0.7527965307235718,
+      "logits/rejected": -0.7202584147453308,
+      "logps/chosen": -138.75991821289062,
+      "logps/rejected": -97.48345947265625,
+      "loss": 0.3684,
+      "rewards/accuracies": 0.8687500357627869,
+      "rewards/chosen": 0.2913173735141754,
+      "rewards/margins": 2.0032336711883545,
+      "rewards/rejected": -1.711916208267212,
+      "step": 1170
+    },
+    {
+      "epoch": 0.6795277857759862,
+      "grad_norm": 17.600046157836914,
+      "learning_rate": 2.8286412279506e-07,
+      "logits/chosen": -0.7335983514785767,
+      "logits/rejected": -0.709388792514801,
+      "logps/chosen": -194.87017822265625,
+      "logps/rejected": -102.54981231689453,
+      "loss": 0.345,
+      "rewards/accuracies": 0.8500000238418579,
+      "rewards/chosen": 1.1104973554611206,
+      "rewards/margins": 2.8248836994171143,
+      "rewards/rejected": -1.7143861055374146,
+      "step": 1180
+    },
+    {
+      "epoch": 0.6852864958249352,
+      "grad_norm": 16.01340103149414,
+      "learning_rate": 2.7385584180358454e-07,
+      "logits/chosen": -0.7293104529380798,
+      "logits/rejected": -0.7005465626716614,
+      "logps/chosen": -125.95570373535156,
+      "logps/rejected": -99.11387634277344,
+      "loss": 0.2968,
+      "rewards/accuracies": 0.878125011920929,
+      "rewards/chosen": 0.33182141184806824,
+      "rewards/margins": 2.125437021255493,
+      "rewards/rejected": -1.793615460395813,
+      "step": 1190
+    },
+    {
+      "epoch": 0.6910452058738843,
+      "grad_norm": 18.85894203186035,
+      "learning_rate": 2.6493892006544117e-07,
+      "logits/chosen": -0.743747889995575,
+      "logits/rejected": -0.7257972955703735,
+      "logps/chosen": -144.69972229003906,
+      "logps/rejected": -113.71995544433594,
+      "loss": 0.354,
+      "rewards/accuracies": 0.8437500596046448,
+      "rewards/chosen": 0.2043149769306183,
+      "rewards/margins": 1.9560476541519165,
+      "rewards/rejected": -1.7517324686050415,
+      "step": 1200
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1737,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 400,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 653223570964480.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3626af6279f425d38151e52760e6109e7265905cd46dd2c5bf648259f31ba398
+size 8657

zero_to_fp32.py

@@ -0,0 +1,760 @@
+#!/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 . output_dir/
+#   or
+#   python zero_to_fp32.py . output_dir/ --safe_serialization
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+import gc
+import json
+import numpy as np
+from tqdm import tqdm
+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, weights_only=False)
+
+        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 tqdm(files, desc='Loading checkpoint shards'):
+        state_dict = torch.load(f, map_location=device, mmap=True, weights_only=False)
+        # 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}")
+
+    fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] 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")
+
+
+class GatheredTensor:
+    """
+    A pseudo tensor that collects partitioned weights.
+    It is more memory efficient when there are multiple groups.
+    """
+
+    def __init__(self, flat_groups, flat_groups_offset, offset, partitioned_numel, shape):
+        self.flat_groups = flat_groups
+        self.flat_groups_offset = flat_groups_offset
+        self.offset = offset
+        self.partitioned_numel = partitioned_numel
+        self.shape = shape
+        self.dtype = self.flat_groups[0][0].dtype
+
+    def contiguous(self):
+        """
+        Merge partitioned weights from flat_groups into a single tensor.
+        """
+        end_idx = self.offset + self.partitioned_numel
+        world_size = len(self.flat_groups)
+        pad_flat_param_chunks = []
+
+        for rank_i in range(world_size):
+            # for each rank, we need to collect weights from related group/groups
+            flat_groups_at_rank_i = self.flat_groups[rank_i]
+            start_group_id = None
+            end_group_id = None
+            for group_id in range(len(self.flat_groups_offset)):
+                if self.flat_groups_offset[group_id] <= self.offset < self.flat_groups_offset[group_id + 1]:
+                    start_group_id = group_id
+                if self.flat_groups_offset[group_id] < end_idx <= self.flat_groups_offset[group_id + 1]:
+                    end_group_id = group_id
+                    break
+            # collect weights from related group/groups
+            for group_id in range(start_group_id, end_group_id + 1):
+                flat_tensor = flat_groups_at_rank_i[group_id]
+                start_offset = self.offset - self.flat_groups_offset[group_id]
+                end_offset = min(end_idx, self.flat_groups_offset[group_id + 1]) - self.flat_groups_offset[group_id]
+                pad_flat_param_chunks.append(flat_tensor[start_offset:end_offset])
+
+        # collect weights from all ranks
+        pad_flat_param = torch.cat(pad_flat_param_chunks, dim=0)
+        param = pad_flat_param[:self.shape.numel()].view(self.shape).contiguous()
+        return param
+
+
+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 = sum([flat_group.numel() for flat_group in fp32_flat_groups[0]]) * 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
+    flat_groups_offset = [0] + list(np.cumsum([flat_tensor.numel() for flat_tensor in fp32_flat_groups[0]]))
+    for name, shape in tqdm(param_shapes.items(), desc='Gathering sharded weights'):
+        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}"
+            )
+
+        # memory efficient tensor
+        tensor = GatheredTensor(fp32_flat_groups, flat_groups_offset, offset, partitioned_numel, shape)
+        state_dict[name] = tensor
+        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 to_torch_tensor(state_dict, return_empty_tensor=False):
+    """
+    Convert state_dict of GatheredTensor to torch tensor
+    """
+    torch_state_dict = {}
+    converted_tensors = {}
+    for name, tensor in state_dict.items():
+        tensor_id = id(tensor)
+        if tensor_id in converted_tensors:  # shared tensors
+            shared_tensor = torch_state_dict[converted_tensors[tensor_id]]
+            torch_state_dict[name] = shared_tensor
+        else:
+            converted_tensors[tensor_id] = name
+            if return_empty_tensor:
+                torch_state_dict[name] = torch.empty(tensor.shape, dtype=tensor.dtype)
+            else:
+                torch_state_dict[name] = tensor.contiguous()
+    return torch_state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                             tag=None,
+                                             exclude_frozen_parameters=False,
+                                             lazy_mode=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
+        - ``lazy_mode``: get state_dict in lazy mode. It returns a dict of pesduo tensor instead of torch tensor, which is more memory efficient.
+          Convert the pesduo tensor to torch tensor by ``.contiguous()``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    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.
+
+    Note: the above usage may not work if your application doesn't have sufficient free CPU memory.
+    You may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint. Or you can load state_dict in lazy mode ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, lazy_mode=True) # not on cpu
+        for name, lazy_tensor in state_dict.item():
+            tensor = lazy_tensor.contiguous()  # to cpu
+            print(name, tensor)
+            # del tensor to release memory if it no longer in use
+    """
+    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")
+
+    state_dict = _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+    if lazy_mode:
+        return state_dict
+    else:
+        return to_torch_tensor(state_dict)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir,
+                                               output_dir,
+                                               max_shard_size="5GB",
+                                               safe_serialization=False,
+                                               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_dir``: directory to the pytorch fp32 state_dict output files
+        - ``max_shard_size``: the maximum size for a checkpoint before being sharded, default value is 5GB
+        - ``safe_serialization``:  whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).
+        - ``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
+    """
+
+    # Dependency pre-check
+    if safe_serialization:
+        try:
+            from safetensors.torch import save_file
+        except ImportError:
+            print('If you want to use `safe_serialization`, please `pip install safetensors`')
+            raise
+    if max_shard_size is not None:
+        try:
+            from huggingface_hub import split_torch_state_dict_into_shards
+        except ImportError:
+            print('If you want to use `max_shard_size`, please `pip install huggingface_hub`')
+            raise
+
+    # Convert zero checkpoint to state_dict
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir,
+                                                          tag,
+                                                          exclude_frozen_parameters,
+                                                          lazy_mode=True)
+
+    # Shard the model if it is too big.
+    weights_name = "model.safetensors" if safe_serialization else "pytorch_model.bin"
+    if max_shard_size is not None:
+        filename_pattern = weights_name.replace(".bin", "{suffix}.bin").replace(".safetensors", "{suffix}.safetensors")
+        # an memory-efficient approach for sharding
+        empty_state_dict = to_torch_tensor(state_dict, return_empty_tensor=True)
+        state_dict_split = split_torch_state_dict_into_shards(empty_state_dict,
+                                                              filename_pattern=filename_pattern,
+                                                              max_shard_size=max_shard_size)
+    else:
+        from collections import namedtuple
+        StateDictSplit = namedtuple("StateDictSplit", ["is_sharded", "filename_to_tensors"])
+        state_dict_split = StateDictSplit(is_sharded=False,
+                                          filename_to_tensors={weights_name: list(state_dict.keys())})
+
+    # Save the model by shard
+    os.makedirs(output_dir, exist_ok=True)
+    filename_to_tensors = state_dict_split.filename_to_tensors.items()
+    for shard_file, tensors in tqdm(filename_to_tensors, desc="Saving checkpoint shards"):
+        shard_state_dict = {tensor_name: state_dict[tensor_name] for tensor_name in tensors}
+        shard_state_dict = to_torch_tensor(shard_state_dict)
+        output_path = os.path.join(output_dir, shard_file)
+        if safe_serialization:
+            save_file(shard_state_dict, output_path, metadata={"format": "pt"})
+        else:
+            torch.save(shard_state_dict, output_path)
+        # release the memory of current shard
+        for tensor_name in list(shard_state_dict.keys()):
+            del state_dict[tensor_name]
+            del shard_state_dict[tensor_name]
+        del shard_state_dict
+        gc.collect()
+
+    # Save index if sharded
+    if state_dict_split.is_sharded:
+        index = {
+            "metadata": state_dict_split.metadata,
+            "weight_map": state_dict_split.tensor_to_filename,
+        }
+        save_index_file = "model.safetensors.index.json" if safe_serialization else "pytorch_model.bin.index.json"
+        save_index_file = os.path.join(output_dir, save_index_file)
+        with open(save_index_file, "w", encoding="utf-8") as f:
+            content = json.dumps(index, indent=2, sort_keys=True) + "\n"
+            f.write(content)
+
+
+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_dir",
+                        type=str,
+                        help="directory to the pytorch fp32 state_dict output files"
+                        "(e.g. path/checkpoint-12-output/)")
+    parser.add_argument(
+        "--max_shard_size",
+        type=str,
+        default="5GB",
+        help="The maximum size for a checkpoint before being sharded. Checkpoints shard will then be each of size"
+        "lower than this size. If expressed as a string, needs to be digits followed by a unit (like `5MB`"
+        "We default it to 5GB in order for models to be able to run easily on free-tier google colab instances"
+        "without CPU OOM issues.")
+    parser.add_argument(
+        "--safe_serialization",
+        default=False,
+        action='store_true',
+        help="Whether to save the model using `safetensors` or the traditional PyTorch way (that uses `pickle`).")
+    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_dir,
+                                               max_shard_size=args.max_shard_size,
+                                               safe_serialization=args.safe_serialization,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)