Upload folder using huggingface_hub

.gitattributes

@@ -37,3 +37,4 @@ checkpoint-500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoint-1500/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-2500/tokenizer.json filter=lfs diff=lfs merge=lfs -text

checkpoint-2500/added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<image_soft_token>": 262144
+}

checkpoint-2500/chat_template.jinja

@@ -0,0 +1,47 @@
+{{ bos_token }}
+{%- if messages[0]['role'] == 'system' -%}
+    {%- if messages[0]['content'] is string -%}
+        {%- set first_user_prefix = messages[0]['content'] + '
+
+' -%}
+    {%- else -%}
+        {%- set first_user_prefix = messages[0]['content'][0]['text'] + '
+
+' -%}
+    {%- endif -%}
+    {%- set loop_messages = messages[1:] -%}
+{%- else -%}
+    {%- set first_user_prefix = "" -%}
+    {%- set loop_messages = messages -%}
+{%- endif -%}
+{%- for message in loop_messages -%}
+    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
+        {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
+    {%- endif -%}
+    {%- if (message['role'] == 'assistant') -%}
+        {%- set role = "model" -%}
+    {%- else -%}
+        {%- set role = message['role'] -%}
+    {%- endif -%}
+    {{ '<start_of_turn>' + role + '
+' + (first_user_prefix if loop.first else "") }}
+    {%- if message['content'] is string -%}
+        {{ message['content'] | trim }}
+    {%- elif message['content'] is iterable -%}
+        {%- for item in message['content'] -%}
+            {%- if item['type'] == 'image' -%}
+                {{ '<start_of_image>' }}
+            {%- elif item['type'] == 'text' -%}
+                {{ item['text'] | trim }}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- else -%}
+        {{ raise_exception("Invalid content type") }}
+    {%- endif -%}
+    {{ '<end_of_turn>
+' }}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+    {{'<start_of_turn>model
+'}}
+{%- endif -%}

checkpoint-2500/config.json

@@ -0,0 +1,86 @@
+{
+  "_sliding_window_pattern": 6,
+  "architectures": [
+    "Gemma3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_logit_softcapping": null,
+  "bos_token_id": 2,
+  "dtype": "bfloat16",
+  "eos_token_id": 1,
+  "final_logit_softcapping": null,
+  "head_dim": 256,
+  "hidden_activation": "gelu_pytorch_tanh",
+  "hidden_size": 3840,
+  "initializer_range": 0.02,
+  "intermediate_size": 15360,
+  "layer_types": [
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 131072,
+  "model_type": "gemma3_text",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 48,
+  "num_key_value_heads": 8,
+  "pad_token_id": 0,
+  "query_pre_attn_scalar": 256,
+  "rms_norm_eps": 1e-06,
+  "rope_local_base_freq": 10000.0,
+  "rope_scaling": {
+    "factor": 8.0,
+    "rope_type": "linear"
+  },
+  "rope_theta": 1000000.0,
+  "sliding_window": 1024,
+  "transformers_version": "4.56.1",
+  "use_cache": true,
+  "vocab_size": 262208
+}

checkpoint-2500/generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 2,
+  "do_sample": true,
+  "eos_token_id": [
+    1,
+    106
+  ],
+  "pad_token_id": 0,
+  "top_k": 64,
+  "top_p": 0.95,
+  "transformers_version": "4.56.1"
+}

checkpoint-2500/model-00001-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a4c14f38ffc9e23d873fa999eea1fc965dcedcec267e7a99f07c6fdbe8760d8
+size 4915892992

checkpoint-2500/model-00002-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0335ce336b31e38c581326acb77ff28330c2073764b1924ad7108d788b56ba68
+size 4931294472

checkpoint-2500/model-00003-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0d6a60cf6a1708be0613ab7c4d10457ad99b98c02e40ebd5f320fe02aa0d50c
+size 4931294528

checkpoint-2500/model-00004-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4bd9340bc368427bcc805e329f32ec7f6a73ab12274d00a8b9c03d49fe0853d0
+size 4931294528

checkpoint-2500/model-00005-of-00005.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d80ec30fa09df9cb78149b3af4da5e7f22bbe89087198871d4f36c70903e30a6
+size 3822364808

checkpoint-2500/model.safetensors.index.json

@@ -0,0 +1,634 @@
+{
+  "metadata": {
+    "total_parameters": 11766034176,
+    "total_size": 23532068352
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.self_attn.k_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.self_attn.q_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.self_attn.k_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.self_attn.q_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.17.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.17.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.17.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.17.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.self_attn.k_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.self_attn.q_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.post_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.pre_feedforward_layernorm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.28.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.28.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.28.self_attn.k_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.28.self_attn.q_norm.weight": "model-00003-of-00005.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00005.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.self_attn.k_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.self_attn.q_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.post_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.pre_feedforward_layernorm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.39.post_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.39.pre_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.39.self_attn.k_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.39.self_attn.q_norm.weight": "model-00004-of-00005.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00004-of-00005.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.self_attn.k_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.self_attn.q_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.mlp.down_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.mlp.gate_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.mlp.up_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.post_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.pre_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.self_attn.k_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.self_attn.q_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.mlp.down_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.mlp.gate_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.mlp.up_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.post_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.pre_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.self_attn.k_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.self_attn.q_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.mlp.down_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.mlp.gate_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.mlp.up_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.post_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.pre_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.self_attn.k_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.self_attn.q_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.input_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.mlp.down_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.mlp.gate_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.mlp.up_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.post_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.pre_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.self_attn.k_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.self_attn.q_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.43.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.input_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.mlp.down_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.mlp.gate_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.mlp.up_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.post_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.pre_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.self_attn.k_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.self_attn.q_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.44.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.input_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.mlp.down_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.mlp.gate_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.mlp.up_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.post_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.pre_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.self_attn.k_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.self_attn.q_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.45.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.input_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.mlp.down_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.mlp.gate_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.mlp.up_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.post_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.pre_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.self_attn.k_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.self_attn.o_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.self_attn.q_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.46.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.input_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.mlp.down_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.mlp.gate_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.mlp.up_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.post_attention_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.post_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.pre_feedforward_layernorm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.self_attn.k_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.self_attn.k_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.self_attn.o_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.self_attn.q_norm.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.self_attn.q_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.47.self_attn.v_proj.weight": "model-00005-of-00005.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.post_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.pre_feedforward_layernorm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.self_attn.k_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.self_attn.q_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.6.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.6.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.6.self_attn.k_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.6.self_attn.q_norm.weight": "model-00001-of-00005.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00005.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.post_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.pre_feedforward_layernorm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.self_attn.k_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.self_attn.q_norm.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00005.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00005.safetensors",
+    "model.norm.weight": "model-00005-of-00005.safetensors"
+  }
+}

checkpoint-2500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f34edcc5f81ef6c480018a65dbccfba11298e0ce4b40c15ae8e462a9b004e264
+size 18355

checkpoint-2500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ca6dfb5217fc8f636627b7f6d5261930661582cc3f8cd6c427127b2bf7aef5db
+size 14645

checkpoint-2500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd200143f139327b4d41bcb99ffeaa13b2acc11ded5b3dbab181a6ff2ce855d7
+size 1465

checkpoint-2500/special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-2500/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
+size 33384568

checkpoint-2500/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

checkpoint-2500/trainer_state.json

@@ -0,0 +1,2809 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.448648047179259,
+  "eval_steps": 100,
+  "global_step": 2500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "entropy": 2.4441180236637594,
+      "epoch": 0.017803493935684877,
+      "grad_norm": 0.0079345703125,
+      "learning_rate": 1.8e-07,
+      "loss": 3.382,
+      "mean_token_accuracy": 0.4423037525266409,
+      "num_tokens": 257189.0,
+      "step": 10
+    },
+    {
+      "entropy": 2.45082361549139,
+      "epoch": 0.035606987871369754,
+      "grad_norm": 0.0086669921875,
+      "learning_rate": 3.8e-07,
+      "loss": 3.375,
+      "mean_token_accuracy": 0.4396079422906041,
+      "num_tokens": 528285.0,
+      "step": 20
+    },
+    {
+      "entropy": 2.461306857317686,
+      "epoch": 0.053410481807054634,
+      "grad_norm": 0.00860595703125,
+      "learning_rate": 5.800000000000001e-07,
+      "loss": 3.3839,
+      "mean_token_accuracy": 0.43654753603041174,
+      "num_tokens": 801635.0,
+      "step": 30
+    },
+    {
+      "entropy": 2.486989914625883,
+      "epoch": 0.07121397574273951,
+      "grad_norm": 0.00860595703125,
+      "learning_rate": 7.8e-07,
+      "loss": 3.4182,
+      "mean_token_accuracy": 0.435793649405241,
+      "num_tokens": 1060812.0,
+      "step": 40
+    },
+    {
+      "entropy": 2.4777335457503797,
+      "epoch": 0.08901746967842439,
+      "grad_norm": 0.007049560546875,
+      "learning_rate": 9.800000000000001e-07,
+      "loss": 3.407,
+      "mean_token_accuracy": 0.4357369150966406,
+      "num_tokens": 1327380.0,
+      "step": 50
+    },
+    {
+      "entropy": 2.478399809449911,
+      "epoch": 0.10682096361410927,
+      "grad_norm": 0.00726318359375,
+      "learning_rate": 1.1800000000000001e-06,
+      "loss": 3.4405,
+      "mean_token_accuracy": 0.43458456825464964,
+      "num_tokens": 1597405.0,
+      "step": 60
+    },
+    {
+      "entropy": 2.5134010180830955,
+      "epoch": 0.12462445754979415,
+      "grad_norm": 0.006683349609375,
+      "learning_rate": 1.3800000000000001e-06,
+      "loss": 3.4613,
+      "mean_token_accuracy": 0.4291722310706973,
+      "num_tokens": 1859684.0,
+      "step": 70
+    },
+    {
+      "entropy": 2.5495446130633352,
+      "epoch": 0.14242795148547902,
+      "grad_norm": 0.00787353515625,
+      "learning_rate": 1.5800000000000001e-06,
+      "loss": 3.5072,
+      "mean_token_accuracy": 0.42735045179724696,
+      "num_tokens": 2118068.0,
+      "step": 80
+    },
+    {
+      "entropy": 2.4757557071745397,
+      "epoch": 0.1602314454211639,
+      "grad_norm": 0.007659912109375,
+      "learning_rate": 1.7800000000000001e-06,
+      "loss": 3.4228,
+      "mean_token_accuracy": 0.43469055872410534,
+      "num_tokens": 2388824.0,
+      "step": 90
+    },
+    {
+      "entropy": 2.5207040145993234,
+      "epoch": 0.17803493935684878,
+      "grad_norm": 0.00994873046875,
+      "learning_rate": 1.98e-06,
+      "loss": 3.506,
+      "mean_token_accuracy": 0.4275117186829448,
+      "num_tokens": 2644330.0,
+      "step": 100
+    },
+    {
+      "epoch": 0.17803493935684878,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 2644330.0,
+      "eval_chemistry_runtime": 42.5715,
+      "eval_chemistry_samples_per_second": 11.745,
+      "eval_chemistry_steps_per_second": 2.936,
+      "step": 100
+    },
+    {
+      "entropy": 2.460470324009657,
+      "epoch": 0.19583843329253367,
+      "grad_norm": 0.0084228515625,
+      "learning_rate": 2.1800000000000003e-06,
+      "loss": 3.3945,
+      "mean_token_accuracy": 0.4391466261819005,
+      "num_tokens": 2913700.0,
+      "step": 110
+    },
+    {
+      "entropy": 2.478592600673437,
+      "epoch": 0.21364192722821854,
+      "grad_norm": 0.006866455078125,
+      "learning_rate": 2.38e-06,
+      "loss": 3.3703,
+      "mean_token_accuracy": 0.4405368799343705,
+      "num_tokens": 3185255.0,
+      "step": 120
+    },
+    {
+      "entropy": 2.4654680602252483,
+      "epoch": 0.2314454211639034,
+      "grad_norm": 0.00994873046875,
+      "learning_rate": 2.5800000000000003e-06,
+      "loss": 3.3981,
+      "mean_token_accuracy": 0.43972534593194723,
+      "num_tokens": 3454750.0,
+      "step": 130
+    },
+    {
+      "entropy": 2.5018775165081024,
+      "epoch": 0.2492489150995883,
+      "grad_norm": 0.0091552734375,
+      "learning_rate": 2.7800000000000005e-06,
+      "loss": 3.4682,
+      "mean_token_accuracy": 0.4316142167896032,
+      "num_tokens": 3719113.0,
+      "step": 140
+    },
+    {
+      "entropy": 2.568158584088087,
+      "epoch": 0.26705240903527316,
+      "grad_norm": 0.007659912109375,
+      "learning_rate": 2.9800000000000003e-06,
+      "loss": 3.5044,
+      "mean_token_accuracy": 0.4264186339452863,
+      "num_tokens": 3990505.0,
+      "step": 150
+    },
+    {
+      "entropy": 2.420287236571312,
+      "epoch": 0.28485590297095803,
+      "grad_norm": 0.007415771484375,
+      "learning_rate": 3.1800000000000005e-06,
+      "loss": 3.3513,
+      "mean_token_accuracy": 0.44414950646460055,
+      "num_tokens": 4267403.0,
+      "step": 160
+    },
+    {
+      "entropy": 2.4532286070287226,
+      "epoch": 0.30265939690664295,
+      "grad_norm": 0.0079345703125,
+      "learning_rate": 3.3800000000000007e-06,
+      "loss": 3.3478,
+      "mean_token_accuracy": 0.4404185781255364,
+      "num_tokens": 4535458.0,
+      "step": 170
+    },
+    {
+      "entropy": 2.4985749416053293,
+      "epoch": 0.3204628908423278,
+      "grad_norm": 0.007720947265625,
+      "learning_rate": 3.58e-06,
+      "loss": 3.4363,
+      "mean_token_accuracy": 0.4365795683115721,
+      "num_tokens": 4796815.0,
+      "step": 180
+    },
+    {
+      "entropy": 2.4519996128976347,
+      "epoch": 0.3382663847780127,
+      "grad_norm": 0.00701904296875,
+      "learning_rate": 3.7800000000000002e-06,
+      "loss": 3.3977,
+      "mean_token_accuracy": 0.440047624707222,
+      "num_tokens": 5066948.0,
+      "step": 190
+    },
+    {
+      "entropy": 2.521295293420553,
+      "epoch": 0.35606987871369755,
+      "grad_norm": 0.010009765625,
+      "learning_rate": 3.980000000000001e-06,
+      "loss": 3.4496,
+      "mean_token_accuracy": 0.4322214875370264,
+      "num_tokens": 5324751.0,
+      "step": 200
+    },
+    {
+      "epoch": 0.35606987871369755,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 5324751.0,
+      "eval_chemistry_runtime": 42.5086,
+      "eval_chemistry_samples_per_second": 11.762,
+      "eval_chemistry_steps_per_second": 2.941,
+      "step": 200
+    },
+    {
+      "entropy": 2.44460117071867,
+      "epoch": 0.3738733726493824,
+      "grad_norm": 0.00732421875,
+      "learning_rate": 4.18e-06,
+      "loss": 3.3862,
+      "mean_token_accuracy": 0.4415476618334651,
+      "num_tokens": 5585508.0,
+      "step": 210
+    },
+    {
+      "entropy": 2.480318387597799,
+      "epoch": 0.39167686658506734,
+      "grad_norm": 0.006591796875,
+      "learning_rate": 4.38e-06,
+      "loss": 3.4199,
+      "mean_token_accuracy": 0.43676935136318207,
+      "num_tokens": 5848889.0,
+      "step": 220
+    },
+    {
+      "entropy": 2.4665837228298186,
+      "epoch": 0.4094803605207522,
+      "grad_norm": 0.007049560546875,
+      "learning_rate": 4.58e-06,
+      "loss": 3.399,
+      "mean_token_accuracy": 0.4371540261432528,
+      "num_tokens": 6114855.0,
+      "step": 230
+    },
+    {
+      "entropy": 2.4934676766395567,
+      "epoch": 0.4272838544564371,
+      "grad_norm": 0.00787353515625,
+      "learning_rate": 4.78e-06,
+      "loss": 3.392,
+      "mean_token_accuracy": 0.4352115549147129,
+      "num_tokens": 6378152.0,
+      "step": 240
+    },
+    {
+      "entropy": 2.510968402773142,
+      "epoch": 0.44508734839212194,
+      "grad_norm": 0.00811767578125,
+      "learning_rate": 4.980000000000001e-06,
+      "loss": 3.4512,
+      "mean_token_accuracy": 0.43123594429343937,
+      "num_tokens": 6637273.0,
+      "step": 250
+    },
+    {
+      "entropy": 2.459284470230341,
+      "epoch": 0.4628908423278068,
+      "grad_norm": 0.0081787109375,
+      "learning_rate": 5.18e-06,
+      "loss": 3.4105,
+      "mean_token_accuracy": 0.43572237603366376,
+      "num_tokens": 6896684.0,
+      "step": 260
+    },
+    {
+      "entropy": 2.4737772688269617,
+      "epoch": 0.48069433626349173,
+      "grad_norm": 0.01324462890625,
+      "learning_rate": 5.380000000000001e-06,
+      "loss": 3.4275,
+      "mean_token_accuracy": 0.4344434389844537,
+      "num_tokens": 7166608.0,
+      "step": 270
+    },
+    {
+      "entropy": 2.393792650848627,
+      "epoch": 0.4984978301991766,
+      "grad_norm": 0.00714111328125,
+      "learning_rate": 5.580000000000001e-06,
+      "loss": 3.3174,
+      "mean_token_accuracy": 0.445861529558897,
+      "num_tokens": 7444923.0,
+      "step": 280
+    },
+    {
+      "entropy": 2.4808071456849574,
+      "epoch": 0.5163013241348615,
+      "grad_norm": 0.0125732421875,
+      "learning_rate": 5.78e-06,
+      "loss": 3.4216,
+      "mean_token_accuracy": 0.4335135340690613,
+      "num_tokens": 7706502.0,
+      "step": 290
+    },
+    {
+      "entropy": 2.4993100173771383,
+      "epoch": 0.5341048180705463,
+      "grad_norm": 0.01153564453125,
+      "learning_rate": 5.98e-06,
+      "loss": 3.4677,
+      "mean_token_accuracy": 0.43226035628467796,
+      "num_tokens": 7969704.0,
+      "step": 300
+    },
+    {
+      "epoch": 0.5341048180705463,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 7969704.0,
+      "eval_chemistry_runtime": 42.3808,
+      "eval_chemistry_samples_per_second": 11.798,
+      "eval_chemistry_steps_per_second": 2.949,
+      "step": 300
+    },
+    {
+      "entropy": 2.49663657695055,
+      "epoch": 0.5519083120062312,
+      "grad_norm": 0.0101318359375,
+      "learning_rate": 6.18e-06,
+      "loss": 3.452,
+      "mean_token_accuracy": 0.4326026676222682,
+      "num_tokens": 8242162.0,
+      "step": 310
+    },
+    {
+      "entropy": 2.491827418655157,
+      "epoch": 0.5697118059419161,
+      "grad_norm": 0.007476806640625,
+      "learning_rate": 6.380000000000001e-06,
+      "loss": 3.4701,
+      "mean_token_accuracy": 0.4347437607124448,
+      "num_tokens": 8497852.0,
+      "step": 320
+    },
+    {
+      "entropy": 2.526278664171696,
+      "epoch": 0.587515299877601,
+      "grad_norm": 0.00946044921875,
+      "learning_rate": 6.5800000000000005e-06,
+      "loss": 3.504,
+      "mean_token_accuracy": 0.4276865454390645,
+      "num_tokens": 8757753.0,
+      "step": 330
+    },
+    {
+      "entropy": 2.4302435807883738,
+      "epoch": 0.6053187938132859,
+      "grad_norm": 0.0189208984375,
+      "learning_rate": 6.780000000000001e-06,
+      "loss": 3.3655,
+      "mean_token_accuracy": 0.4372914243489504,
+      "num_tokens": 9024677.0,
+      "step": 340
+    },
+    {
+      "entropy": 2.4870374642312525,
+      "epoch": 0.6231222877489707,
+      "grad_norm": 0.00750732421875,
+      "learning_rate": 6.98e-06,
+      "loss": 3.4166,
+      "mean_token_accuracy": 0.4357576759532094,
+      "num_tokens": 9291760.0,
+      "step": 350
+    },
+    {
+      "entropy": 2.440358117222786,
+      "epoch": 0.6409257816846556,
+      "grad_norm": 0.007415771484375,
+      "learning_rate": 7.180000000000001e-06,
+      "loss": 3.3708,
+      "mean_token_accuracy": 0.43819774594157934,
+      "num_tokens": 9561091.0,
+      "step": 360
+    },
+    {
+      "entropy": 2.4518581442534924,
+      "epoch": 0.6587292756203404,
+      "grad_norm": 0.01177978515625,
+      "learning_rate": 7.3800000000000005e-06,
+      "loss": 3.39,
+      "mean_token_accuracy": 0.4373545182868838,
+      "num_tokens": 9827272.0,
+      "step": 370
+    },
+    {
+      "entropy": 2.4458557978272437,
+      "epoch": 0.6765327695560254,
+      "grad_norm": 0.01019287109375,
+      "learning_rate": 7.58e-06,
+      "loss": 3.3803,
+      "mean_token_accuracy": 0.44243213571608064,
+      "num_tokens": 10096065.0,
+      "step": 380
+    },
+    {
+      "entropy": 2.457463264465332,
+      "epoch": 0.6943362634917103,
+      "grad_norm": 0.00775146484375,
+      "learning_rate": 7.78e-06,
+      "loss": 3.3973,
+      "mean_token_accuracy": 0.43960350602865217,
+      "num_tokens": 10364601.0,
+      "step": 390
+    },
+    {
+      "entropy": 2.4410374239087105,
+      "epoch": 0.7121397574273951,
+      "grad_norm": 0.00860595703125,
+      "learning_rate": 7.980000000000002e-06,
+      "loss": 3.3916,
+      "mean_token_accuracy": 0.4379643935710192,
+      "num_tokens": 10633325.0,
+      "step": 400
+    },
+    {
+      "epoch": 0.7121397574273951,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 10633325.0,
+      "eval_chemistry_runtime": 42.4526,
+      "eval_chemistry_samples_per_second": 11.778,
+      "eval_chemistry_steps_per_second": 2.944,
+      "step": 400
+    },
+    {
+      "entropy": 2.409457255154848,
+      "epoch": 0.72994325136308,
+      "grad_norm": 0.0079345703125,
+      "learning_rate": 8.18e-06,
+      "loss": 3.3536,
+      "mean_token_accuracy": 0.4413986885920167,
+      "num_tokens": 10897916.0,
+      "step": 410
+    },
+    {
+      "entropy": 2.4690888836979865,
+      "epoch": 0.7477467452987648,
+      "grad_norm": 0.009033203125,
+      "learning_rate": 8.380000000000001e-06,
+      "loss": 3.4202,
+      "mean_token_accuracy": 0.43675678241997956,
+      "num_tokens": 11165356.0,
+      "step": 420
+    },
+    {
+      "entropy": 2.456707588583231,
+      "epoch": 0.7655502392344498,
+      "grad_norm": 0.0074462890625,
+      "learning_rate": 8.580000000000001e-06,
+      "loss": 3.3849,
+      "mean_token_accuracy": 0.43797059506177904,
+      "num_tokens": 11436799.0,
+      "step": 430
+    },
+    {
+      "entropy": 2.481979151070118,
+      "epoch": 0.7833537331701347,
+      "grad_norm": 0.00927734375,
+      "learning_rate": 8.78e-06,
+      "loss": 3.3948,
+      "mean_token_accuracy": 0.4386452713981271,
+      "num_tokens": 11703496.0,
+      "step": 440
+    },
+    {
+      "entropy": 2.511328932642937,
+      "epoch": 0.8011572271058195,
+      "grad_norm": 0.00836181640625,
+      "learning_rate": 8.98e-06,
+      "loss": 3.4704,
+      "mean_token_accuracy": 0.4308528220281005,
+      "num_tokens": 11965530.0,
+      "step": 450
+    },
+    {
+      "entropy": 2.4488867104053496,
+      "epoch": 0.8189607210415044,
+      "grad_norm": 0.00897216796875,
+      "learning_rate": 9.180000000000002e-06,
+      "loss": 3.4121,
+      "mean_token_accuracy": 0.43401869516819713,
+      "num_tokens": 12224427.0,
+      "step": 460
+    },
+    {
+      "entropy": 2.41854225769639,
+      "epoch": 0.8367642149771892,
+      "grad_norm": 0.01202392578125,
+      "learning_rate": 9.38e-06,
+      "loss": 3.312,
+      "mean_token_accuracy": 0.4470580581575632,
+      "num_tokens": 12509124.0,
+      "step": 470
+    },
+    {
+      "entropy": 2.4362686768174173,
+      "epoch": 0.8545677089128741,
+      "grad_norm": 0.007232666015625,
+      "learning_rate": 9.58e-06,
+      "loss": 3.3794,
+      "mean_token_accuracy": 0.4400447830557823,
+      "num_tokens": 12778408.0,
+      "step": 480
+    },
+    {
+      "entropy": 2.4321289747953414,
+      "epoch": 0.8723712028485591,
+      "grad_norm": 0.00811767578125,
+      "learning_rate": 9.780000000000001e-06,
+      "loss": 3.3687,
+      "mean_token_accuracy": 0.44097492955625056,
+      "num_tokens": 13046473.0,
+      "step": 490
+    },
+    {
+      "entropy": 2.4888896882534026,
+      "epoch": 0.8901746967842439,
+      "grad_norm": 0.007720947265625,
+      "learning_rate": 9.980000000000001e-06,
+      "loss": 3.4261,
+      "mean_token_accuracy": 0.43556336853653194,
+      "num_tokens": 13301659.0,
+      "step": 500
+    },
+    {
+      "epoch": 0.8901746967842439,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 13301659.0,
+      "eval_chemistry_runtime": 42.4183,
+      "eval_chemistry_samples_per_second": 11.787,
+      "eval_chemistry_steps_per_second": 2.947,
+      "step": 500
+    },
+    {
+      "entropy": 2.4065550975501537,
+      "epoch": 0.9079781907199288,
+      "grad_norm": 0.007110595703125,
+      "learning_rate": 1.018e-05,
+      "loss": 3.3268,
+      "mean_token_accuracy": 0.44642978757619856,
+      "num_tokens": 13570986.0,
+      "step": 510
+    },
+    {
+      "entropy": 2.4483018897473814,
+      "epoch": 0.9257816846556136,
+      "grad_norm": 0.0087890625,
+      "learning_rate": 1.038e-05,
+      "loss": 3.4018,
+      "mean_token_accuracy": 0.4386210318654776,
+      "num_tokens": 13833901.0,
+      "step": 520
+    },
+    {
+      "entropy": 2.454428631067276,
+      "epoch": 0.9435851785912985,
+      "grad_norm": 0.00921630859375,
+      "learning_rate": 1.0580000000000002e-05,
+      "loss": 3.4197,
+      "mean_token_accuracy": 0.43513929657638073,
+      "num_tokens": 14093888.0,
+      "step": 530
+    },
+    {
+      "entropy": 2.4978294670581818,
+      "epoch": 0.9613886725269835,
+      "grad_norm": 0.00750732421875,
+      "learning_rate": 1.0780000000000002e-05,
+      "loss": 3.43,
+      "mean_token_accuracy": 0.43464562948793173,
+      "num_tokens": 14354918.0,
+      "step": 540
+    },
+    {
+      "entropy": 2.46180320456624,
+      "epoch": 0.9791921664626683,
+      "grad_norm": 0.00811767578125,
+      "learning_rate": 1.0980000000000002e-05,
+      "loss": 3.4206,
+      "mean_token_accuracy": 0.43483266811817883,
+      "num_tokens": 14620244.0,
+      "step": 550
+    },
+    {
+      "entropy": 2.4592869602143765,
+      "epoch": 0.9969956603983532,
+      "grad_norm": 0.00787353515625,
+      "learning_rate": 1.1180000000000001e-05,
+      "loss": 3.4471,
+      "mean_token_accuracy": 0.43371057212352754,
+      "num_tokens": 14886593.0,
+      "step": 560
+    },
+    {
+      "entropy": 2.3787012269420007,
+      "epoch": 1.014242795148548,
+      "grad_norm": 0.00830078125,
+      "learning_rate": 1.138e-05,
+      "loss": 3.2918,
+      "mean_token_accuracy": 0.44634086931905437,
+      "num_tokens": 15146469.0,
+      "step": 570
+    },
+    {
+      "entropy": 2.524292258173227,
+      "epoch": 1.0320462890842328,
+      "grad_norm": 0.0128173828125,
+      "learning_rate": 1.1580000000000001e-05,
+      "loss": 3.4816,
+      "mean_token_accuracy": 0.43193594105541705,
+      "num_tokens": 15404569.0,
+      "step": 580
+    },
+    {
+      "entropy": 2.4150695733726026,
+      "epoch": 1.0498497830199176,
+      "grad_norm": 0.00921630859375,
+      "learning_rate": 1.178e-05,
+      "loss": 3.3678,
+      "mean_token_accuracy": 0.4405944043770432,
+      "num_tokens": 15681709.0,
+      "step": 590
+    },
+    {
+      "entropy": 2.441706320643425,
+      "epoch": 1.0676532769556026,
+      "grad_norm": 0.007659912109375,
+      "learning_rate": 1.198e-05,
+      "loss": 3.3742,
+      "mean_token_accuracy": 0.4395477233454585,
+      "num_tokens": 15949421.0,
+      "step": 600
+    },
+    {
+      "epoch": 1.0676532769556026,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 15949421.0,
+      "eval_chemistry_runtime": 42.5959,
+      "eval_chemistry_samples_per_second": 11.738,
+      "eval_chemistry_steps_per_second": 2.935,
+      "step": 600
+    },
+    {
+      "entropy": 2.4119591929018496,
+      "epoch": 1.0854567708912874,
+      "grad_norm": 0.0076904296875,
+      "learning_rate": 1.218e-05,
+      "loss": 3.367,
+      "mean_token_accuracy": 0.4415699910372496,
+      "num_tokens": 16221916.0,
+      "step": 610
+    },
+    {
+      "entropy": 2.51780494004488,
+      "epoch": 1.1032602648269723,
+      "grad_norm": 0.007537841796875,
+      "learning_rate": 1.2380000000000002e-05,
+      "loss": 3.4485,
+      "mean_token_accuracy": 0.4301307639107108,
+      "num_tokens": 16488154.0,
+      "step": 620
+    },
+    {
+      "entropy": 2.4647247321903705,
+      "epoch": 1.121063758762657,
+      "grad_norm": 0.007354736328125,
+      "learning_rate": 1.2580000000000002e-05,
+      "loss": 3.4027,
+      "mean_token_accuracy": 0.4368906727060676,
+      "num_tokens": 16756828.0,
+      "step": 630
+    },
+    {
+      "entropy": 2.4489255413413047,
+      "epoch": 1.138867252698342,
+      "grad_norm": 0.00799560546875,
+      "learning_rate": 1.2780000000000001e-05,
+      "loss": 3.3816,
+      "mean_token_accuracy": 0.43627936094999314,
+      "num_tokens": 17023681.0,
+      "step": 640
+    },
+    {
+      "entropy": 2.4757089667022227,
+      "epoch": 1.156670746634027,
+      "grad_norm": 0.01190185546875,
+      "learning_rate": 1.2980000000000001e-05,
+      "loss": 3.4233,
+      "mean_token_accuracy": 0.4360200822353363,
+      "num_tokens": 17286672.0,
+      "step": 650
+    },
+    {
+      "entropy": 2.542103400081396,
+      "epoch": 1.1744742405697117,
+      "grad_norm": 0.00726318359375,
+      "learning_rate": 1.3180000000000001e-05,
+      "loss": 3.4708,
+      "mean_token_accuracy": 0.42764721568673847,
+      "num_tokens": 17551581.0,
+      "step": 660
+    },
+    {
+      "entropy": 2.4632993809878827,
+      "epoch": 1.1922777345053968,
+      "grad_norm": 0.00946044921875,
+      "learning_rate": 1.3380000000000002e-05,
+      "loss": 3.3879,
+      "mean_token_accuracy": 0.43964554853737353,
+      "num_tokens": 17821205.0,
+      "step": 670
+    },
+    {
+      "entropy": 2.4974757246673107,
+      "epoch": 1.2100812284410816,
+      "grad_norm": 0.007415771484375,
+      "learning_rate": 1.3580000000000002e-05,
+      "loss": 3.4339,
+      "mean_token_accuracy": 0.4323098760098219,
+      "num_tokens": 18087759.0,
+      "step": 680
+    },
+    {
+      "entropy": 2.485871135443449,
+      "epoch": 1.2278847223767664,
+      "grad_norm": 0.0084228515625,
+      "learning_rate": 1.378e-05,
+      "loss": 3.4259,
+      "mean_token_accuracy": 0.4338374109938741,
+      "num_tokens": 18355732.0,
+      "step": 690
+    },
+    {
+      "entropy": 2.4948798827826977,
+      "epoch": 1.2456882163124514,
+      "grad_norm": 0.0069580078125,
+      "learning_rate": 1.398e-05,
+      "loss": 3.4297,
+      "mean_token_accuracy": 0.435987046174705,
+      "num_tokens": 18626245.0,
+      "step": 700
+    },
+    {
+      "epoch": 1.2456882163124514,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 18626245.0,
+      "eval_chemistry_runtime": 42.5841,
+      "eval_chemistry_samples_per_second": 11.741,
+      "eval_chemistry_steps_per_second": 2.935,
+      "step": 700
+    },
+    {
+      "entropy": 2.408124291151762,
+      "epoch": 1.2634917102481362,
+      "grad_norm": 0.00927734375,
+      "learning_rate": 1.418e-05,
+      "loss": 3.3266,
+      "mean_token_accuracy": 0.4450059160590172,
+      "num_tokens": 18899632.0,
+      "step": 710
+    },
+    {
+      "entropy": 2.4984724432229997,
+      "epoch": 1.281295204183821,
+      "grad_norm": 0.01214599609375,
+      "learning_rate": 1.4380000000000001e-05,
+      "loss": 3.4488,
+      "mean_token_accuracy": 0.4306912997737527,
+      "num_tokens": 19173635.0,
+      "step": 720
+    },
+    {
+      "entropy": 2.407220109552145,
+      "epoch": 1.299098698119506,
+      "grad_norm": 0.01129150390625,
+      "learning_rate": 1.4580000000000001e-05,
+      "loss": 3.3575,
+      "mean_token_accuracy": 0.44124190472066405,
+      "num_tokens": 19453495.0,
+      "step": 730
+    },
+    {
+      "entropy": 2.426318255066872,
+      "epoch": 1.3169021920551909,
+      "grad_norm": 0.007049560546875,
+      "learning_rate": 1.478e-05,
+      "loss": 3.366,
+      "mean_token_accuracy": 0.43959415052086115,
+      "num_tokens": 19719086.0,
+      "step": 740
+    },
+    {
+      "entropy": 2.4674886889755725,
+      "epoch": 1.3347056859908757,
+      "grad_norm": 0.009033203125,
+      "learning_rate": 1.498e-05,
+      "loss": 3.4065,
+      "mean_token_accuracy": 0.4346128342673182,
+      "num_tokens": 19981272.0,
+      "step": 750
+    },
+    {
+      "entropy": 2.4601897999644278,
+      "epoch": 1.3525091799265607,
+      "grad_norm": 0.007354736328125,
+      "learning_rate": 1.5180000000000002e-05,
+      "loss": 3.4277,
+      "mean_token_accuracy": 0.43639134224504234,
+      "num_tokens": 20243034.0,
+      "step": 760
+    },
+    {
+      "entropy": 2.4810195587575437,
+      "epoch": 1.3703126738622455,
+      "grad_norm": 0.0084228515625,
+      "learning_rate": 1.5380000000000002e-05,
+      "loss": 3.4245,
+      "mean_token_accuracy": 0.43384860958904026,
+      "num_tokens": 20509028.0,
+      "step": 770
+    },
+    {
+      "entropy": 2.5357875250279904,
+      "epoch": 1.3881161677979303,
+      "grad_norm": 0.006988525390625,
+      "learning_rate": 1.5580000000000003e-05,
+      "loss": 3.5276,
+      "mean_token_accuracy": 0.42721649557352065,
+      "num_tokens": 20762668.0,
+      "step": 780
+    },
+    {
+      "entropy": 2.573459583520889,
+      "epoch": 1.4059196617336152,
+      "grad_norm": 0.00787353515625,
+      "learning_rate": 1.578e-05,
+      "loss": 3.5041,
+      "mean_token_accuracy": 0.42504764050245286,
+      "num_tokens": 21025277.0,
+      "step": 790
+    },
+    {
+      "entropy": 2.5099629506468775,
+      "epoch": 1.4237231556693,
+      "grad_norm": 0.0091552734375,
+      "learning_rate": 1.5980000000000003e-05,
+      "loss": 3.4735,
+      "mean_token_accuracy": 0.42990608550608156,
+      "num_tokens": 21292320.0,
+      "step": 800
+    },
+    {
+      "epoch": 1.4237231556693,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 21292320.0,
+      "eval_chemistry_runtime": 42.58,
+      "eval_chemistry_samples_per_second": 11.743,
+      "eval_chemistry_steps_per_second": 2.936,
+      "step": 800
+    },
+    {
+      "entropy": 2.5165034629404546,
+      "epoch": 1.441526649604985,
+      "grad_norm": 0.00732421875,
+      "learning_rate": 1.618e-05,
+      "loss": 3.4497,
+      "mean_token_accuracy": 0.4366149786859751,
+      "num_tokens": 21563668.0,
+      "step": 810
+    },
+    {
+      "entropy": 2.4973253421485424,
+      "epoch": 1.4593301435406698,
+      "grad_norm": 0.007232666015625,
+      "learning_rate": 1.638e-05,
+      "loss": 3.4814,
+      "mean_token_accuracy": 0.4312220415100455,
+      "num_tokens": 21817298.0,
+      "step": 820
+    },
+    {
+      "entropy": 2.52367008626461,
+      "epoch": 1.4771336374763546,
+      "grad_norm": 0.007598876953125,
+      "learning_rate": 1.658e-05,
+      "loss": 3.468,
+      "mean_token_accuracy": 0.4320263473317027,
+      "num_tokens": 22083010.0,
+      "step": 830
+    },
+    {
+      "entropy": 2.50519383251667,
+      "epoch": 1.4949371314120397,
+      "grad_norm": 0.0074462890625,
+      "learning_rate": 1.6780000000000002e-05,
+      "loss": 3.4575,
+      "mean_token_accuracy": 0.4309014746919274,
+      "num_tokens": 22350046.0,
+      "step": 840
+    },
+    {
+      "entropy": 2.458015527576208,
+      "epoch": 1.5127406253477245,
+      "grad_norm": 0.0081787109375,
+      "learning_rate": 1.698e-05,
+      "loss": 3.4087,
+      "mean_token_accuracy": 0.4401903055608273,
+      "num_tokens": 22615933.0,
+      "step": 850
+    },
+    {
+      "entropy": 2.4192430078983307,
+      "epoch": 1.5305441192834093,
+      "grad_norm": 0.00787353515625,
+      "learning_rate": 1.718e-05,
+      "loss": 3.3493,
+      "mean_token_accuracy": 0.4408345725387335,
+      "num_tokens": 22886843.0,
+      "step": 860
+    },
+    {
+      "entropy": 2.4866544254124165,
+      "epoch": 1.5483476132190943,
+      "grad_norm": 0.006927490234375,
+      "learning_rate": 1.7380000000000003e-05,
+      "loss": 3.4323,
+      "mean_token_accuracy": 0.4359192430973053,
+      "num_tokens": 23140244.0,
+      "step": 870
+    },
+    {
+      "entropy": 2.472956708818674,
+      "epoch": 1.5661511071547791,
+      "grad_norm": 0.0076904296875,
+      "learning_rate": 1.758e-05,
+      "loss": 3.4127,
+      "mean_token_accuracy": 0.4368757115676999,
+      "num_tokens": 23405121.0,
+      "step": 880
+    },
+    {
+      "entropy": 2.4871600806713103,
+      "epoch": 1.583954601090464,
+      "grad_norm": 0.006378173828125,
+      "learning_rate": 1.7780000000000003e-05,
+      "loss": 3.425,
+      "mean_token_accuracy": 0.43275546226650474,
+      "num_tokens": 23671826.0,
+      "step": 890
+    },
+    {
+      "entropy": 2.5024729229509832,
+      "epoch": 1.601758095026149,
+      "grad_norm": 0.01068115234375,
+      "learning_rate": 1.798e-05,
+      "loss": 3.4415,
+      "mean_token_accuracy": 0.43367660194635393,
+      "num_tokens": 23933374.0,
+      "step": 900
+    },
+    {
+      "epoch": 1.601758095026149,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 23933374.0,
+      "eval_chemistry_runtime": 42.5199,
+      "eval_chemistry_samples_per_second": 11.759,
+      "eval_chemistry_steps_per_second": 2.94,
+      "step": 900
+    },
+    {
+      "entropy": 2.4241936951875687,
+      "epoch": 1.6195615889618338,
+      "grad_norm": 0.00836181640625,
+      "learning_rate": 1.8180000000000002e-05,
+      "loss": 3.3926,
+      "mean_token_accuracy": 0.4393082341179252,
+      "num_tokens": 24210156.0,
+      "step": 910
+    },
+    {
+      "entropy": 2.501750738918781,
+      "epoch": 1.6373650828975186,
+      "grad_norm": 0.01043701171875,
+      "learning_rate": 1.8380000000000004e-05,
+      "loss": 3.428,
+      "mean_token_accuracy": 0.4345035368576646,
+      "num_tokens": 24462726.0,
+      "step": 920
+    },
+    {
+      "entropy": 2.4093470610678196,
+      "epoch": 1.6551685768332036,
+      "grad_norm": 0.0118408203125,
+      "learning_rate": 1.858e-05,
+      "loss": 3.3407,
+      "mean_token_accuracy": 0.44576158542186023,
+      "num_tokens": 24732815.0,
+      "step": 930
+    },
+    {
+      "entropy": 2.469248706102371,
+      "epoch": 1.6729720707688884,
+      "grad_norm": 0.0128173828125,
+      "learning_rate": 1.878e-05,
+      "loss": 3.3903,
+      "mean_token_accuracy": 0.4362227080389857,
+      "num_tokens": 25002091.0,
+      "step": 940
+    },
+    {
+      "entropy": 2.4393011182546616,
+      "epoch": 1.6907755647045732,
+      "grad_norm": 0.00982666015625,
+      "learning_rate": 1.898e-05,
+      "loss": 3.4097,
+      "mean_token_accuracy": 0.4377862988039851,
+      "num_tokens": 25267949.0,
+      "step": 950
+    },
+    {
+      "entropy": 2.499985621124506,
+      "epoch": 1.7085790586402583,
+      "grad_norm": 0.007110595703125,
+      "learning_rate": 1.918e-05,
+      "loss": 3.4206,
+      "mean_token_accuracy": 0.4386218637228012,
+      "num_tokens": 25525772.0,
+      "step": 960
+    },
+    {
+      "entropy": 2.518709710240364,
+      "epoch": 1.7263825525759429,
+      "grad_norm": 0.015625,
+      "learning_rate": 1.938e-05,
+      "loss": 3.4744,
+      "mean_token_accuracy": 0.4290819203481078,
+      "num_tokens": 25787727.0,
+      "step": 970
+    },
+    {
+      "entropy": 2.434730440378189,
+      "epoch": 1.744186046511628,
+      "grad_norm": 0.007110595703125,
+      "learning_rate": 1.9580000000000002e-05,
+      "loss": 3.3452,
+      "mean_token_accuracy": 0.4429737152531743,
+      "num_tokens": 26058026.0,
+      "step": 980
+    },
+    {
+      "entropy": 2.405204129964113,
+      "epoch": 1.761989540447313,
+      "grad_norm": 0.007659912109375,
+      "learning_rate": 1.978e-05,
+      "loss": 3.3339,
+      "mean_token_accuracy": 0.44453586284071206,
+      "num_tokens": 26326631.0,
+      "step": 990
+    },
+    {
+      "entropy": 2.473785638809204,
+      "epoch": 1.7797930343829975,
+      "grad_norm": 0.008056640625,
+      "learning_rate": 1.9980000000000002e-05,
+      "loss": 3.3982,
+      "mean_token_accuracy": 0.4379414787515998,
+      "num_tokens": 26591735.0,
+      "step": 1000
+    },
+    {
+      "epoch": 1.7797930343829975,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 26591735.0,
+      "eval_chemistry_runtime": 42.6898,
+      "eval_chemistry_samples_per_second": 11.712,
+      "eval_chemistry_steps_per_second": 2.928,
+      "step": 1000
+    },
+    {
+      "entropy": 2.4923629231750963,
+      "epoch": 1.7975965283186826,
+      "grad_norm": 0.007171630859375,
+      "learning_rate": 1.9980000000000002e-05,
+      "loss": 3.4449,
+      "mean_token_accuracy": 0.4350215895101428,
+      "num_tokens": 26857071.0,
+      "step": 1010
+    },
+    {
+      "entropy": 2.518780706077814,
+      "epoch": 1.8154000222543676,
+      "grad_norm": 0.01068115234375,
+      "learning_rate": 1.995777777777778e-05,
+      "loss": 3.4732,
+      "mean_token_accuracy": 0.42953309677541257,
+      "num_tokens": 27120617.0,
+      "step": 1020
+    },
+    {
+      "entropy": 2.4439337849617004,
+      "epoch": 1.8332035161900522,
+      "grad_norm": 0.00830078125,
+      "learning_rate": 1.9935555555555557e-05,
+      "loss": 3.3826,
+      "mean_token_accuracy": 0.4376870134845376,
+      "num_tokens": 27391174.0,
+      "step": 1030
+    },
+    {
+      "entropy": 2.4611451610922814,
+      "epoch": 1.8510070101257372,
+      "grad_norm": 0.0072021484375,
+      "learning_rate": 1.9913333333333335e-05,
+      "loss": 3.3908,
+      "mean_token_accuracy": 0.4394783997908235,
+      "num_tokens": 27656941.0,
+      "step": 1040
+    },
+    {
+      "entropy": 2.456043781340122,
+      "epoch": 1.868810504061422,
+      "grad_norm": 0.0081787109375,
+      "learning_rate": 1.9891111111111112e-05,
+      "loss": 3.4033,
+      "mean_token_accuracy": 0.4401937620714307,
+      "num_tokens": 27927118.0,
+      "step": 1050
+    },
+    {
+      "entropy": 2.4034786596894264,
+      "epoch": 1.8866139979971068,
+      "grad_norm": 0.009033203125,
+      "learning_rate": 1.986888888888889e-05,
+      "loss": 3.315,
+      "mean_token_accuracy": 0.4454406937584281,
+      "num_tokens": 28192620.0,
+      "step": 1060
+    },
+    {
+      "entropy": 2.451674747467041,
+      "epoch": 1.9044174919327919,
+      "grad_norm": 0.007232666015625,
+      "learning_rate": 1.9846666666666668e-05,
+      "loss": 3.3975,
+      "mean_token_accuracy": 0.4367500660941005,
+      "num_tokens": 28458168.0,
+      "step": 1070
+    },
+    {
+      "entropy": 2.477339595556259,
+      "epoch": 1.9222209858684767,
+      "grad_norm": 0.009033203125,
+      "learning_rate": 1.9824444444444445e-05,
+      "loss": 3.4136,
+      "mean_token_accuracy": 0.4369886856526136,
+      "num_tokens": 28722756.0,
+      "step": 1080
+    },
+    {
+      "entropy": 2.5017261296510696,
+      "epoch": 1.9400244798041615,
+      "grad_norm": 0.010498046875,
+      "learning_rate": 1.9802222222222226e-05,
+      "loss": 3.4096,
+      "mean_token_accuracy": 0.4368307461962104,
+      "num_tokens": 28978375.0,
+      "step": 1090
+    },
+    {
+      "entropy": 2.4673263989388943,
+      "epoch": 1.9578279737398465,
+      "grad_norm": 0.00787353515625,
+      "learning_rate": 1.978e-05,
+      "loss": 3.4071,
+      "mean_token_accuracy": 0.43733639605343344,
+      "num_tokens": 29236865.0,
+      "step": 1100
+    },
+    {
+      "epoch": 1.9578279737398465,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 29236865.0,
+      "eval_chemistry_runtime": 42.6475,
+      "eval_chemistry_samples_per_second": 11.724,
+      "eval_chemistry_steps_per_second": 2.931,
+      "step": 1100
+    },
+    {
+      "entropy": 2.4563313499093056,
+      "epoch": 1.9756314676755313,
+      "grad_norm": 0.008056640625,
+      "learning_rate": 1.975777777777778e-05,
+      "loss": 3.3669,
+      "mean_token_accuracy": 0.4400833772495389,
+      "num_tokens": 29500428.0,
+      "step": 1110
+    },
+    {
+      "entropy": 2.4409970842301845,
+      "epoch": 1.9934349616112161,
+      "grad_norm": 0.009033203125,
+      "learning_rate": 1.9735555555555556e-05,
+      "loss": 3.3781,
+      "mean_token_accuracy": 0.44074930921196936,
+      "num_tokens": 29766739.0,
+      "step": 1120
+    },
+    {
+      "entropy": 2.46847146403405,
+      "epoch": 2.010682096361411,
+      "grad_norm": 0.00885009765625,
+      "learning_rate": 1.9713333333333337e-05,
+      "loss": 3.4114,
+      "mean_token_accuracy": 0.4375636675665455,
+      "num_tokens": 30025427.0,
+      "step": 1130
+    },
+    {
+      "entropy": 2.4604240149259566,
+      "epoch": 2.028485590297096,
+      "grad_norm": 0.0126953125,
+      "learning_rate": 1.969111111111111e-05,
+      "loss": 3.3852,
+      "mean_token_accuracy": 0.43901363387703896,
+      "num_tokens": 30286651.0,
+      "step": 1140
+    },
+    {
+      "entropy": 2.468199959397316,
+      "epoch": 2.0462890842327806,
+      "grad_norm": 0.007659912109375,
+      "learning_rate": 1.9668888888888892e-05,
+      "loss": 3.4268,
+      "mean_token_accuracy": 0.4350541580468416,
+      "num_tokens": 30552624.0,
+      "step": 1150
+    },
+    {
+      "entropy": 2.4525125212967396,
+      "epoch": 2.0640925781684656,
+      "grad_norm": 0.00823974609375,
+      "learning_rate": 1.9646666666666666e-05,
+      "loss": 3.3892,
+      "mean_token_accuracy": 0.4397247971966863,
+      "num_tokens": 30823746.0,
+      "step": 1160
+    },
+    {
+      "entropy": 2.575217917561531,
+      "epoch": 2.0818960721041506,
+      "grad_norm": 0.007476806640625,
+      "learning_rate": 1.9624444444444447e-05,
+      "loss": 3.5312,
+      "mean_token_accuracy": 0.42372212260961534,
+      "num_tokens": 31076307.0,
+      "step": 1170
+    },
+    {
+      "entropy": 2.5399364054203035,
+      "epoch": 2.099699566039835,
+      "grad_norm": 0.01165771484375,
+      "learning_rate": 1.9602222222222225e-05,
+      "loss": 3.4832,
+      "mean_token_accuracy": 0.4261633366346359,
+      "num_tokens": 31340763.0,
+      "step": 1180
+    },
+    {
+      "entropy": 2.440959357470274,
+      "epoch": 2.1175030599755202,
+      "grad_norm": 0.007232666015625,
+      "learning_rate": 1.9580000000000002e-05,
+      "loss": 3.3932,
+      "mean_token_accuracy": 0.4403660248965025,
+      "num_tokens": 31607200.0,
+      "step": 1190
+    },
+    {
+      "entropy": 2.5287065640091897,
+      "epoch": 2.1353065539112053,
+      "grad_norm": 0.00897216796875,
+      "learning_rate": 1.955777777777778e-05,
+      "loss": 3.4754,
+      "mean_token_accuracy": 0.4273608535528183,
+      "num_tokens": 31874392.0,
+      "step": 1200
+    },
+    {
+      "epoch": 2.1353065539112053,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 31874392.0,
+      "eval_chemistry_runtime": 42.5525,
+      "eval_chemistry_samples_per_second": 11.75,
+      "eval_chemistry_steps_per_second": 2.938,
+      "step": 1200
+    },
+    {
+      "entropy": 2.4604250721633436,
+      "epoch": 2.15311004784689,
+      "grad_norm": 0.00732421875,
+      "learning_rate": 1.9535555555555557e-05,
+      "loss": 3.3662,
+      "mean_token_accuracy": 0.4413184406235814,
+      "num_tokens": 32144016.0,
+      "step": 1210
+    },
+    {
+      "entropy": 2.460251194238663,
+      "epoch": 2.170913541782575,
+      "grad_norm": 0.0098876953125,
+      "learning_rate": 1.9513333333333335e-05,
+      "loss": 3.4265,
+      "mean_token_accuracy": 0.43892472572624686,
+      "num_tokens": 32410902.0,
+      "step": 1220
+    },
+    {
+      "entropy": 2.4349136739969253,
+      "epoch": 2.18871703571826,
+      "grad_norm": 0.007537841796875,
+      "learning_rate": 1.9491111111111113e-05,
+      "loss": 3.386,
+      "mean_token_accuracy": 0.44226096775382756,
+      "num_tokens": 32674338.0,
+      "step": 1230
+    },
+    {
+      "entropy": 2.4614202618598937,
+      "epoch": 2.2065205296539445,
+      "grad_norm": 0.0123291015625,
+      "learning_rate": 1.946888888888889e-05,
+      "loss": 3.3965,
+      "mean_token_accuracy": 0.43613822385668755,
+      "num_tokens": 32930219.0,
+      "step": 1240
+    },
+    {
+      "entropy": 2.494431575387716,
+      "epoch": 2.2243240235896296,
+      "grad_norm": 0.013916015625,
+      "learning_rate": 1.9446666666666668e-05,
+      "loss": 3.4509,
+      "mean_token_accuracy": 0.433765122294426,
+      "num_tokens": 33194318.0,
+      "step": 1250
+    },
+    {
+      "entropy": 2.443953486531973,
+      "epoch": 2.242127517525314,
+      "grad_norm": 0.007659912109375,
+      "learning_rate": 1.9424444444444446e-05,
+      "loss": 3.3948,
+      "mean_token_accuracy": 0.43882110957056286,
+      "num_tokens": 33453292.0,
+      "step": 1260
+    },
+    {
+      "entropy": 2.3844580121338366,
+      "epoch": 2.259931011460999,
+      "grad_norm": 0.0081787109375,
+      "learning_rate": 1.9402222222222223e-05,
+      "loss": 3.3094,
+      "mean_token_accuracy": 0.4473427068442106,
+      "num_tokens": 33728875.0,
+      "step": 1270
+    },
+    {
+      "entropy": 2.5078157439827917,
+      "epoch": 2.277734505396684,
+      "grad_norm": 0.007354736328125,
+      "learning_rate": 1.938e-05,
+      "loss": 3.4404,
+      "mean_token_accuracy": 0.43358458634465935,
+      "num_tokens": 33999305.0,
+      "step": 1280
+    },
+    {
+      "entropy": 2.489903759211302,
+      "epoch": 2.295537999332369,
+      "grad_norm": 0.00750732421875,
+      "learning_rate": 1.935777777777778e-05,
+      "loss": 3.4548,
+      "mean_token_accuracy": 0.4296891471371055,
+      "num_tokens": 34269727.0,
+      "step": 1290
+    },
+    {
+      "entropy": 2.384353207051754,
+      "epoch": 2.313341493268054,
+      "grad_norm": 0.00775146484375,
+      "learning_rate": 1.9335555555555556e-05,
+      "loss": 3.3205,
+      "mean_token_accuracy": 0.44720757827162744,
+      "num_tokens": 34537714.0,
+      "step": 1300
+    },
+    {
+      "epoch": 2.313341493268054,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 34537714.0,
+      "eval_chemistry_runtime": 42.6053,
+      "eval_chemistry_samples_per_second": 11.736,
+      "eval_chemistry_steps_per_second": 2.934,
+      "step": 1300
+    },
+    {
+      "entropy": 2.471423901617527,
+      "epoch": 2.331144987203739,
+      "grad_norm": 0.007781982421875,
+      "learning_rate": 1.9313333333333334e-05,
+      "loss": 3.4072,
+      "mean_token_accuracy": 0.43562198970466853,
+      "num_tokens": 34806647.0,
+      "step": 1310
+    },
+    {
+      "entropy": 2.460309857875109,
+      "epoch": 2.3489484811394234,
+      "grad_norm": 0.01104736328125,
+      "learning_rate": 1.9291111111111115e-05,
+      "loss": 3.4042,
+      "mean_token_accuracy": 0.4393015902489424,
+      "num_tokens": 35076423.0,
+      "step": 1320
+    },
+    {
+      "entropy": 2.50424614623189,
+      "epoch": 2.3667519750751085,
+      "grad_norm": 0.007049560546875,
+      "learning_rate": 1.926888888888889e-05,
+      "loss": 3.4226,
+      "mean_token_accuracy": 0.4346967810764909,
+      "num_tokens": 35343494.0,
+      "step": 1330
+    },
+    {
+      "entropy": 2.48561422675848,
+      "epoch": 2.3845554690107935,
+      "grad_norm": 0.0086669921875,
+      "learning_rate": 1.924666666666667e-05,
+      "loss": 3.4177,
+      "mean_token_accuracy": 0.4333121033385396,
+      "num_tokens": 35614731.0,
+      "step": 1340
+    },
+    {
+      "entropy": 2.498768079280853,
+      "epoch": 2.402358962946478,
+      "grad_norm": 0.00885009765625,
+      "learning_rate": 1.9224444444444444e-05,
+      "loss": 3.451,
+      "mean_token_accuracy": 0.4327333400025964,
+      "num_tokens": 35877264.0,
+      "step": 1350
+    },
+    {
+      "entropy": 2.420652062445879,
+      "epoch": 2.420162456882163,
+      "grad_norm": 0.007659912109375,
+      "learning_rate": 1.9202222222222225e-05,
+      "loss": 3.343,
+      "mean_token_accuracy": 0.4431915180757642,
+      "num_tokens": 36147587.0,
+      "step": 1360
+    },
+    {
+      "entropy": 2.4841696746647357,
+      "epoch": 2.437965950817848,
+      "grad_norm": 0.007049560546875,
+      "learning_rate": 1.918e-05,
+      "loss": 3.4194,
+      "mean_token_accuracy": 0.43716037590056656,
+      "num_tokens": 36416125.0,
+      "step": 1370
+    },
+    {
+      "entropy": 2.4914293572306634,
+      "epoch": 2.4557694447535328,
+      "grad_norm": 0.0081787109375,
+      "learning_rate": 1.915777777777778e-05,
+      "loss": 3.4345,
+      "mean_token_accuracy": 0.4334576532244682,
+      "num_tokens": 36680383.0,
+      "step": 1380
+    },
+    {
+      "entropy": 2.470908258855343,
+      "epoch": 2.473572938689218,
+      "grad_norm": 0.00921630859375,
+      "learning_rate": 1.9135555555555555e-05,
+      "loss": 3.4071,
+      "mean_token_accuracy": 0.4377790277823806,
+      "num_tokens": 36946349.0,
+      "step": 1390
+    },
+    {
+      "entropy": 2.407320798188448,
+      "epoch": 2.491376432624903,
+      "grad_norm": 0.0089111328125,
+      "learning_rate": 1.9113333333333336e-05,
+      "loss": 3.358,
+      "mean_token_accuracy": 0.44281436298042537,
+      "num_tokens": 37219532.0,
+      "step": 1400
+    },
+    {
+      "epoch": 2.491376432624903,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 37219532.0,
+      "eval_chemistry_runtime": 42.6672,
+      "eval_chemistry_samples_per_second": 11.719,
+      "eval_chemistry_steps_per_second": 2.93,
+      "step": 1400
+    },
+    {
+      "entropy": 2.4535592779517175,
+      "epoch": 2.5091799265605874,
+      "grad_norm": 0.01446533203125,
+      "learning_rate": 1.9091111111111113e-05,
+      "loss": 3.3835,
+      "mean_token_accuracy": 0.4377025572583079,
+      "num_tokens": 37477914.0,
+      "step": 1410
+    },
+    {
+      "entropy": 2.4648735396564008,
+      "epoch": 2.5269834204962724,
+      "grad_norm": 0.00738525390625,
+      "learning_rate": 1.906888888888889e-05,
+      "loss": 3.3722,
+      "mean_token_accuracy": 0.44036708902567623,
+      "num_tokens": 37746690.0,
+      "step": 1420
+    },
+    {
+      "entropy": 2.5168155536055563,
+      "epoch": 2.544786914431957,
+      "grad_norm": 0.007598876953125,
+      "learning_rate": 1.904666666666667e-05,
+      "loss": 3.4422,
+      "mean_token_accuracy": 0.43462312165647743,
+      "num_tokens": 38005792.0,
+      "step": 1430
+    },
+    {
+      "entropy": 2.518996875733137,
+      "epoch": 2.562590408367642,
+      "grad_norm": 0.007598876953125,
+      "learning_rate": 1.9024444444444446e-05,
+      "loss": 3.4453,
+      "mean_token_accuracy": 0.43033833540976046,
+      "num_tokens": 38269021.0,
+      "step": 1440
+    },
+    {
+      "entropy": 2.4837927713990213,
+      "epoch": 2.580393902303327,
+      "grad_norm": 0.00732421875,
+      "learning_rate": 1.9002222222222224e-05,
+      "loss": 3.4046,
+      "mean_token_accuracy": 0.43629784174263475,
+      "num_tokens": 38545224.0,
+      "step": 1450
+    },
+    {
+      "entropy": 2.4778561271727084,
+      "epoch": 2.598197396239012,
+      "grad_norm": 0.007293701171875,
+      "learning_rate": 1.898e-05,
+      "loss": 3.4428,
+      "mean_token_accuracy": 0.4349894070997834,
+      "num_tokens": 38802290.0,
+      "step": 1460
+    },
+    {
+      "entropy": 2.4231295838952063,
+      "epoch": 2.6160008901746967,
+      "grad_norm": 0.009521484375,
+      "learning_rate": 1.895777777777778e-05,
+      "loss": 3.3682,
+      "mean_token_accuracy": 0.4430950226262212,
+      "num_tokens": 39065787.0,
+      "step": 1470
+    },
+    {
+      "entropy": 2.434224319458008,
+      "epoch": 2.6338043841103818,
+      "grad_norm": 0.007781982421875,
+      "learning_rate": 1.8935555555555556e-05,
+      "loss": 3.3787,
+      "mean_token_accuracy": 0.43969413749873637,
+      "num_tokens": 39328517.0,
+      "step": 1480
+    },
+    {
+      "entropy": 2.4157699927687646,
+      "epoch": 2.6516078780460663,
+      "grad_norm": 0.009765625,
+      "learning_rate": 1.8913333333333334e-05,
+      "loss": 3.3348,
+      "mean_token_accuracy": 0.4441758098080754,
+      "num_tokens": 39600641.0,
+      "step": 1490
+    },
+    {
+      "entropy": 2.5363398104906083,
+      "epoch": 2.6694113719817514,
+      "grad_norm": 0.007537841796875,
+      "learning_rate": 1.8891111111111115e-05,
+      "loss": 3.4932,
+      "mean_token_accuracy": 0.42894220296293495,
+      "num_tokens": 39857917.0,
+      "step": 1500
+    },
+    {
+      "epoch": 2.6694113719817514,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 39857917.0,
+      "eval_chemistry_runtime": 42.5983,
+      "eval_chemistry_samples_per_second": 11.738,
+      "eval_chemistry_steps_per_second": 2.934,
+      "step": 1500
+    },
+    {
+      "entropy": 2.475891087204218,
+      "epoch": 2.6872148659174364,
+      "grad_norm": 0.0084228515625,
+      "learning_rate": 1.886888888888889e-05,
+      "loss": 3.4641,
+      "mean_token_accuracy": 0.4327057808637619,
+      "num_tokens": 40119174.0,
+      "step": 1510
+    },
+    {
+      "entropy": 2.538855505734682,
+      "epoch": 2.7050183598531214,
+      "grad_norm": 0.0086669921875,
+      "learning_rate": 1.884666666666667e-05,
+      "loss": 3.4886,
+      "mean_token_accuracy": 0.427852831967175,
+      "num_tokens": 40379054.0,
+      "step": 1520
+    },
+    {
+      "entropy": 2.465169349312782,
+      "epoch": 2.722821853788806,
+      "grad_norm": 0.00823974609375,
+      "learning_rate": 1.8824444444444445e-05,
+      "loss": 3.4114,
+      "mean_token_accuracy": 0.43687186017632484,
+      "num_tokens": 40650970.0,
+      "step": 1530
+    },
+    {
+      "entropy": 2.4815499387681483,
+      "epoch": 2.740625347724491,
+      "grad_norm": 0.00726318359375,
+      "learning_rate": 1.8802222222222226e-05,
+      "loss": 3.4391,
+      "mean_token_accuracy": 0.43311921060085296,
+      "num_tokens": 40914339.0,
+      "step": 1540
+    },
+    {
+      "entropy": 2.4799845792353152,
+      "epoch": 2.7584288416601757,
+      "grad_norm": 0.00732421875,
+      "learning_rate": 1.878e-05,
+      "loss": 3.419,
+      "mean_token_accuracy": 0.4352044124156237,
+      "num_tokens": 41181827.0,
+      "step": 1550
+    },
+    {
+      "entropy": 2.432544883340597,
+      "epoch": 2.7762323355958607,
+      "grad_norm": 0.00775146484375,
+      "learning_rate": 1.875777777777778e-05,
+      "loss": 3.3587,
+      "mean_token_accuracy": 0.43972424473613503,
+      "num_tokens": 41448961.0,
+      "step": 1560
+    },
+    {
+      "entropy": 2.4766519904136657,
+      "epoch": 2.7940358295315457,
+      "grad_norm": 0.0140380859375,
+      "learning_rate": 1.873555555555556e-05,
+      "loss": 3.4166,
+      "mean_token_accuracy": 0.43819848690181973,
+      "num_tokens": 41712243.0,
+      "step": 1570
+    },
+    {
+      "entropy": 2.45805629491806,
+      "epoch": 2.8118393234672303,
+      "grad_norm": 0.00811767578125,
+      "learning_rate": 1.8713333333333336e-05,
+      "loss": 3.3882,
+      "mean_token_accuracy": 0.4380352281033993,
+      "num_tokens": 41981755.0,
+      "step": 1580
+    },
+    {
+      "entropy": 2.3916926831007004,
+      "epoch": 2.8296428174029153,
+      "grad_norm": 0.0125732421875,
+      "learning_rate": 1.8691111111111114e-05,
+      "loss": 3.3124,
+      "mean_token_accuracy": 0.4474966680631042,
+      "num_tokens": 42258270.0,
+      "step": 1590
+    },
+    {
+      "entropy": 2.4504558242857457,
+      "epoch": 2.8474463113386,
+      "grad_norm": 0.0087890625,
+      "learning_rate": 1.866888888888889e-05,
+      "loss": 3.3973,
+      "mean_token_accuracy": 0.43714725859463216,
+      "num_tokens": 42529321.0,
+      "step": 1600
+    },
+    {
+      "epoch": 2.8474463113386,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 42529321.0,
+      "eval_chemistry_runtime": 42.5546,
+      "eval_chemistry_samples_per_second": 11.75,
+      "eval_chemistry_steps_per_second": 2.937,
+      "step": 1600
+    },
+    {
+      "entropy": 2.526197586208582,
+      "epoch": 2.865249805274285,
+      "grad_norm": 0.0079345703125,
+      "learning_rate": 1.864666666666667e-05,
+      "loss": 3.4609,
+      "mean_token_accuracy": 0.42937238067388533,
+      "num_tokens": 42793729.0,
+      "step": 1610
+    },
+    {
+      "entropy": 2.4750846356153486,
+      "epoch": 2.88305329920997,
+      "grad_norm": 0.00921630859375,
+      "learning_rate": 1.8624444444444446e-05,
+      "loss": 3.4279,
+      "mean_token_accuracy": 0.432634992338717,
+      "num_tokens": 43067006.0,
+      "step": 1620
+    },
+    {
+      "entropy": 2.4470580220222473,
+      "epoch": 2.900856793145655,
+      "grad_norm": 0.0126953125,
+      "learning_rate": 1.8602222222222224e-05,
+      "loss": 3.3994,
+      "mean_token_accuracy": 0.4387952284887433,
+      "num_tokens": 43333868.0,
+      "step": 1630
+    },
+    {
+      "entropy": 2.4646801233291624,
+      "epoch": 2.9186602870813396,
+      "grad_norm": 0.00909423828125,
+      "learning_rate": 1.858e-05,
+      "loss": 3.397,
+      "mean_token_accuracy": 0.43661086559295653,
+      "num_tokens": 43605151.0,
+      "step": 1640
+    },
+    {
+      "entropy": 2.4074466191232204,
+      "epoch": 2.9364637810170247,
+      "grad_norm": 0.007415771484375,
+      "learning_rate": 1.855777777777778e-05,
+      "loss": 3.3384,
+      "mean_token_accuracy": 0.4428932035341859,
+      "num_tokens": 43863034.0,
+      "step": 1650
+    },
+    {
+      "entropy": 2.5186894349753857,
+      "epoch": 2.9542672749527092,
+      "grad_norm": 0.007354736328125,
+      "learning_rate": 1.8535555555555557e-05,
+      "loss": 3.4249,
+      "mean_token_accuracy": 0.43363551832735536,
+      "num_tokens": 44126796.0,
+      "step": 1660
+    },
+    {
+      "entropy": 2.5221563249826433,
+      "epoch": 2.9720707688883943,
+      "grad_norm": 0.00799560546875,
+      "learning_rate": 1.8513333333333335e-05,
+      "loss": 3.4755,
+      "mean_token_accuracy": 0.4296230224892497,
+      "num_tokens": 44382724.0,
+      "step": 1670
+    },
+    {
+      "entropy": 2.4261739112436773,
+      "epoch": 2.9898742628240793,
+      "grad_norm": 0.00921630859375,
+      "learning_rate": 1.8491111111111112e-05,
+      "loss": 3.347,
+      "mean_token_accuracy": 0.44398615062236785,
+      "num_tokens": 44651471.0,
+      "step": 1680
+    },
+    {
+      "entropy": 2.4371220088774157,
+      "epoch": 3.007121397574274,
+      "grad_norm": 0.008544921875,
+      "learning_rate": 1.846888888888889e-05,
+      "loss": 3.3525,
+      "mean_token_accuracy": 0.44014161959771186,
+      "num_tokens": 44909981.0,
+      "step": 1690
+    },
+    {
+      "entropy": 2.597546722739935,
+      "epoch": 3.0249248915099587,
+      "grad_norm": 0.007354736328125,
+      "learning_rate": 1.8446666666666667e-05,
+      "loss": 3.5249,
+      "mean_token_accuracy": 0.4250665778294206,
+      "num_tokens": 45176547.0,
+      "step": 1700
+    },
+    {
+      "epoch": 3.0249248915099587,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 45176547.0,
+      "eval_chemistry_runtime": 42.5973,
+      "eval_chemistry_samples_per_second": 11.738,
+      "eval_chemistry_steps_per_second": 2.934,
+      "step": 1700
+    },
+    {
+      "entropy": 2.4016756273806097,
+      "epoch": 3.0427283854456437,
+      "grad_norm": 0.008056640625,
+      "learning_rate": 1.842444444444445e-05,
+      "loss": 3.3445,
+      "mean_token_accuracy": 0.44047501031309366,
+      "num_tokens": 45437181.0,
+      "step": 1710
+    },
+    {
+      "entropy": 2.4717557713389398,
+      "epoch": 3.0605318793813288,
+      "grad_norm": 0.00830078125,
+      "learning_rate": 1.8402222222222223e-05,
+      "loss": 3.4273,
+      "mean_token_accuracy": 0.4353035241365433,
+      "num_tokens": 45702431.0,
+      "step": 1720
+    },
+    {
+      "entropy": 2.4562893763184546,
+      "epoch": 3.0783353733170133,
+      "grad_norm": 0.00738525390625,
+      "learning_rate": 1.8380000000000004e-05,
+      "loss": 3.3755,
+      "mean_token_accuracy": 0.43956287633627655,
+      "num_tokens": 45975788.0,
+      "step": 1730
+    },
+    {
+      "entropy": 2.4700839944183826,
+      "epoch": 3.0961388672526984,
+      "grad_norm": 0.0103759765625,
+      "learning_rate": 1.8357777777777778e-05,
+      "loss": 3.3978,
+      "mean_token_accuracy": 0.4417863454669714,
+      "num_tokens": 46251349.0,
+      "step": 1740
+    },
+    {
+      "entropy": 2.5270811855793,
+      "epoch": 3.1139423611883834,
+      "grad_norm": 0.00775146484375,
+      "learning_rate": 1.833555555555556e-05,
+      "loss": 3.4516,
+      "mean_token_accuracy": 0.42940014600753784,
+      "num_tokens": 46516802.0,
+      "step": 1750
+    },
+    {
+      "entropy": 2.4189466029405593,
+      "epoch": 3.131745855124068,
+      "grad_norm": 0.007537841796875,
+      "learning_rate": 1.8313333333333333e-05,
+      "loss": 3.3473,
+      "mean_token_accuracy": 0.446175323985517,
+      "num_tokens": 46790537.0,
+      "step": 1760
+    },
+    {
+      "entropy": 2.437445816397667,
+      "epoch": 3.149549349059753,
+      "grad_norm": 0.01043701171875,
+      "learning_rate": 1.8291111111111114e-05,
+      "loss": 3.3659,
+      "mean_token_accuracy": 0.44263256788253785,
+      "num_tokens": 47052877.0,
+      "step": 1770
+    },
+    {
+      "entropy": 2.477902936190367,
+      "epoch": 3.167352842995438,
+      "grad_norm": 0.01214599609375,
+      "learning_rate": 1.8268888888888888e-05,
+      "loss": 3.411,
+      "mean_token_accuracy": 0.43642533868551253,
+      "num_tokens": 47315587.0,
+      "step": 1780
+    },
+    {
+      "entropy": 2.4296828620135784,
+      "epoch": 3.1851563369311227,
+      "grad_norm": 0.0081787109375,
+      "learning_rate": 1.824666666666667e-05,
+      "loss": 3.3685,
+      "mean_token_accuracy": 0.4402240352705121,
+      "num_tokens": 47578352.0,
+      "step": 1790
+    },
+    {
+      "entropy": 2.455223270505667,
+      "epoch": 3.2029598308668077,
+      "grad_norm": 0.01470947265625,
+      "learning_rate": 1.8224444444444447e-05,
+      "loss": 3.4172,
+      "mean_token_accuracy": 0.43478179927915334,
+      "num_tokens": 47844623.0,
+      "step": 1800
+    },
+    {
+      "epoch": 3.2029598308668077,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 47844623.0,
+      "eval_chemistry_runtime": 42.5947,
+      "eval_chemistry_samples_per_second": 11.739,
+      "eval_chemistry_steps_per_second": 2.935,
+      "step": 1800
+    },
+    {
+      "entropy": 2.4119900725781918,
+      "epoch": 3.2207633248024923,
+      "grad_norm": 0.0123291015625,
+      "learning_rate": 1.8202222222222225e-05,
+      "loss": 3.3467,
+      "mean_token_accuracy": 0.4440757390111685,
+      "num_tokens": 48112262.0,
+      "step": 1810
+    },
+    {
+      "entropy": 2.5602151431143283,
+      "epoch": 3.2385668187381773,
+      "grad_norm": 0.0103759765625,
+      "learning_rate": 1.8180000000000002e-05,
+      "loss": 3.5068,
+      "mean_token_accuracy": 0.4251785816624761,
+      "num_tokens": 48369097.0,
+      "step": 1820
+    },
+    {
+      "entropy": 2.4663229435682297,
+      "epoch": 3.2563703126738623,
+      "grad_norm": 0.0078125,
+      "learning_rate": 1.815777777777778e-05,
+      "loss": 3.4226,
+      "mean_token_accuracy": 0.4367042934522033,
+      "num_tokens": 48628165.0,
+      "step": 1830
+    },
+    {
+      "entropy": 2.441531176120043,
+      "epoch": 3.274173806609547,
+      "grad_norm": 0.00933837890625,
+      "learning_rate": 1.8135555555555557e-05,
+      "loss": 3.3767,
+      "mean_token_accuracy": 0.43784170281142,
+      "num_tokens": 48898634.0,
+      "step": 1840
+    },
+    {
+      "entropy": 2.437831664085388,
+      "epoch": 3.291977300545232,
+      "grad_norm": 0.0089111328125,
+      "learning_rate": 1.8113333333333335e-05,
+      "loss": 3.3906,
+      "mean_token_accuracy": 0.43811208922415973,
+      "num_tokens": 49167839.0,
+      "step": 1850
+    },
+    {
+      "entropy": 2.4592106595635412,
+      "epoch": 3.309780794480917,
+      "grad_norm": 0.007415771484375,
+      "learning_rate": 1.8091111111111113e-05,
+      "loss": 3.3945,
+      "mean_token_accuracy": 0.43993463516235354,
+      "num_tokens": 49431378.0,
+      "step": 1860
+    },
+    {
+      "entropy": 2.4519283570349217,
+      "epoch": 3.3275842884166016,
+      "grad_norm": 0.00885009765625,
+      "learning_rate": 1.806888888888889e-05,
+      "loss": 3.3876,
+      "mean_token_accuracy": 0.43461998719722034,
+      "num_tokens": 49699309.0,
+      "step": 1870
+    },
+    {
+      "entropy": 2.483532691001892,
+      "epoch": 3.3453877823522866,
+      "grad_norm": 0.00921630859375,
+      "learning_rate": 1.8046666666666668e-05,
+      "loss": 3.4364,
+      "mean_token_accuracy": 0.43375972770154475,
+      "num_tokens": 49972032.0,
+      "step": 1880
+    },
+    {
+      "entropy": 2.470829947292805,
+      "epoch": 3.3631912762879717,
+      "grad_norm": 0.0150146484375,
+      "learning_rate": 1.8024444444444445e-05,
+      "loss": 3.4237,
+      "mean_token_accuracy": 0.4348238715901971,
+      "num_tokens": 50241367.0,
+      "step": 1890
+    },
+    {
+      "entropy": 2.503614215552807,
+      "epoch": 3.3809947702236562,
+      "grad_norm": 0.0074462890625,
+      "learning_rate": 1.8002222222222223e-05,
+      "loss": 3.4753,
+      "mean_token_accuracy": 0.43145324755460024,
+      "num_tokens": 50511167.0,
+      "step": 1900
+    },
+    {
+      "epoch": 3.3809947702236562,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 50511167.0,
+      "eval_chemistry_runtime": 42.5948,
+      "eval_chemistry_samples_per_second": 11.739,
+      "eval_chemistry_steps_per_second": 2.935,
+      "step": 1900
+    },
+    {
+      "entropy": 2.5126790434122084,
+      "epoch": 3.3987982641593413,
+      "grad_norm": 0.0091552734375,
+      "learning_rate": 1.798e-05,
+      "loss": 3.4633,
+      "mean_token_accuracy": 0.43145898953080175,
+      "num_tokens": 50776842.0,
+      "step": 1910
+    },
+    {
+      "entropy": 2.483235966414213,
+      "epoch": 3.4166017580950263,
+      "grad_norm": 0.0087890625,
+      "learning_rate": 1.7957777777777778e-05,
+      "loss": 3.4276,
+      "mean_token_accuracy": 0.4340201187878847,
+      "num_tokens": 51040514.0,
+      "step": 1920
+    },
+    {
+      "entropy": 2.512890145927668,
+      "epoch": 3.434405252030711,
+      "grad_norm": 0.00775146484375,
+      "learning_rate": 1.7935555555555556e-05,
+      "loss": 3.4503,
+      "mean_token_accuracy": 0.431868121214211,
+      "num_tokens": 51300696.0,
+      "step": 1930
+    },
+    {
+      "entropy": 2.5034680627286434,
+      "epoch": 3.452208745966396,
+      "grad_norm": 0.00799560546875,
+      "learning_rate": 1.7913333333333337e-05,
+      "loss": 3.4761,
+      "mean_token_accuracy": 0.4282655959948897,
+      "num_tokens": 51558606.0,
+      "step": 1940
+    },
+    {
+      "entropy": 2.49245428070426,
+      "epoch": 3.470012239902081,
+      "grad_norm": 0.00830078125,
+      "learning_rate": 1.789111111111111e-05,
+      "loss": 3.4056,
+      "mean_token_accuracy": 0.434098650328815,
+      "num_tokens": 51827834.0,
+      "step": 1950
+    },
+    {
+      "entropy": 2.443667434155941,
+      "epoch": 3.4878157338377656,
+      "grad_norm": 0.0081787109375,
+      "learning_rate": 1.7868888888888892e-05,
+      "loss": 3.3762,
+      "mean_token_accuracy": 0.4413266645744443,
+      "num_tokens": 52100179.0,
+      "step": 1960
+    },
+    {
+      "entropy": 2.47061547935009,
+      "epoch": 3.5056192277734506,
+      "grad_norm": 0.007080078125,
+      "learning_rate": 1.7846666666666666e-05,
+      "loss": 3.3948,
+      "mean_token_accuracy": 0.43764521069824697,
+      "num_tokens": 52368055.0,
+      "step": 1970
+    },
+    {
+      "entropy": 2.4235463075339796,
+      "epoch": 3.523422721709135,
+      "grad_norm": 0.0072021484375,
+      "learning_rate": 1.7824444444444447e-05,
+      "loss": 3.392,
+      "mean_token_accuracy": 0.4394680192694068,
+      "num_tokens": 52637505.0,
+      "step": 1980
+    },
+    {
+      "entropy": 2.466986934840679,
+      "epoch": 3.54122621564482,
+      "grad_norm": 0.0123291015625,
+      "learning_rate": 1.780222222222222e-05,
+      "loss": 3.3834,
+      "mean_token_accuracy": 0.43876779917627573,
+      "num_tokens": 52906194.0,
+      "step": 1990
+    },
+    {
+      "entropy": 2.3780635252594946,
+      "epoch": 3.5590297095805052,
+      "grad_norm": 0.0089111328125,
+      "learning_rate": 1.7780000000000003e-05,
+      "loss": 3.3102,
+      "mean_token_accuracy": 0.44717768765985966,
+      "num_tokens": 53172376.0,
+      "step": 2000
+    },
+    {
+      "epoch": 3.5590297095805052,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 53172376.0,
+      "eval_chemistry_runtime": 42.5948,
+      "eval_chemistry_samples_per_second": 11.739,
+      "eval_chemistry_steps_per_second": 2.935,
+      "step": 2000
+    },
+    {
+      "entropy": 2.433636147528887,
+      "epoch": 3.5768332035161903,
+      "grad_norm": 0.00811767578125,
+      "learning_rate": 1.7757777777777777e-05,
+      "loss": 3.3646,
+      "mean_token_accuracy": 0.43849537931382654,
+      "num_tokens": 53443036.0,
+      "step": 2010
+    },
+    {
+      "entropy": 2.4704249776899814,
+      "epoch": 3.594636697451875,
+      "grad_norm": 0.007476806640625,
+      "learning_rate": 1.7735555555555558e-05,
+      "loss": 3.3844,
+      "mean_token_accuracy": 0.4379355899989605,
+      "num_tokens": 53706113.0,
+      "step": 2020
+    },
+    {
+      "entropy": 2.4817553743720056,
+      "epoch": 3.61244019138756,
+      "grad_norm": 0.00689697265625,
+      "learning_rate": 1.7713333333333335e-05,
+      "loss": 3.4096,
+      "mean_token_accuracy": 0.4397146377712488,
+      "num_tokens": 53964233.0,
+      "step": 2030
+    },
+    {
+      "entropy": 2.457537139207125,
+      "epoch": 3.6302436853232445,
+      "grad_norm": 0.0101318359375,
+      "learning_rate": 1.7691111111111113e-05,
+      "loss": 3.3613,
+      "mean_token_accuracy": 0.44034327398985623,
+      "num_tokens": 54234732.0,
+      "step": 2040
+    },
+    {
+      "entropy": 2.418001752346754,
+      "epoch": 3.6480471792589295,
+      "grad_norm": 0.0081787109375,
+      "learning_rate": 1.766888888888889e-05,
+      "loss": 3.3832,
+      "mean_token_accuracy": 0.4414777553640306,
+      "num_tokens": 54501459.0,
+      "step": 2050
+    },
+    {
+      "entropy": 2.4330545268952846,
+      "epoch": 3.6658506731946146,
+      "grad_norm": 0.00799560546875,
+      "learning_rate": 1.7646666666666668e-05,
+      "loss": 3.3669,
+      "mean_token_accuracy": 0.4401468733325601,
+      "num_tokens": 54767963.0,
+      "step": 2060
+    },
+    {
+      "entropy": 2.4961201697587967,
+      "epoch": 3.6836541671302996,
+      "grad_norm": 0.007080078125,
+      "learning_rate": 1.7624444444444446e-05,
+      "loss": 3.4403,
+      "mean_token_accuracy": 0.4336260788142681,
+      "num_tokens": 55030717.0,
+      "step": 2070
+    },
+    {
+      "entropy": 2.4974512115120886,
+      "epoch": 3.701457661065984,
+      "grad_norm": 0.0079345703125,
+      "learning_rate": 1.7602222222222223e-05,
+      "loss": 3.4273,
+      "mean_token_accuracy": 0.4334062092006207,
+      "num_tokens": 55305577.0,
+      "step": 2080
+    },
+    {
+      "entropy": 2.4657119020819662,
+      "epoch": 3.719261155001669,
+      "grad_norm": 0.00823974609375,
+      "learning_rate": 1.758e-05,
+      "loss": 3.4296,
+      "mean_token_accuracy": 0.4343094065785408,
+      "num_tokens": 55571633.0,
+      "step": 2090
+    },
+    {
+      "entropy": 2.4537970520555974,
+      "epoch": 3.737064648937354,
+      "grad_norm": 0.0072021484375,
+      "learning_rate": 1.755777777777778e-05,
+      "loss": 3.3807,
+      "mean_token_accuracy": 0.43807193282991647,
+      "num_tokens": 55838629.0,
+      "step": 2100
+    },
+    {
+      "epoch": 3.737064648937354,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 55838629.0,
+      "eval_chemistry_runtime": 42.5535,
+      "eval_chemistry_samples_per_second": 11.75,
+      "eval_chemistry_steps_per_second": 2.937,
+      "step": 2100
+    },
+    {
+      "entropy": 2.4643385350704192,
+      "epoch": 3.754868142873039,
+      "grad_norm": 0.0084228515625,
+      "learning_rate": 1.7535555555555556e-05,
+      "loss": 3.4164,
+      "mean_token_accuracy": 0.43682933039963245,
+      "num_tokens": 56095994.0,
+      "step": 2110
+    },
+    {
+      "entropy": 2.4785969384014606,
+      "epoch": 3.772671636808724,
+      "grad_norm": 0.00848388671875,
+      "learning_rate": 1.7513333333333334e-05,
+      "loss": 3.4194,
+      "mean_token_accuracy": 0.43563003111630677,
+      "num_tokens": 56353653.0,
+      "step": 2120
+    },
+    {
+      "entropy": 2.440801540762186,
+      "epoch": 3.7904751307444084,
+      "grad_norm": 0.00799560546875,
+      "learning_rate": 1.749111111111111e-05,
+      "loss": 3.3961,
+      "mean_token_accuracy": 0.4388452073559165,
+      "num_tokens": 56613510.0,
+      "step": 2130
+    },
+    {
+      "entropy": 2.482428055256605,
+      "epoch": 3.8082786246800935,
+      "grad_norm": 0.00762939453125,
+      "learning_rate": 1.746888888888889e-05,
+      "loss": 3.4338,
+      "mean_token_accuracy": 0.43407910857349635,
+      "num_tokens": 56867014.0,
+      "step": 2140
+    },
+    {
+      "entropy": 2.4996217407286165,
+      "epoch": 3.826082118615778,
+      "grad_norm": 0.007232666015625,
+      "learning_rate": 1.7446666666666667e-05,
+      "loss": 3.4325,
+      "mean_token_accuracy": 0.4348539121448994,
+      "num_tokens": 57140038.0,
+      "step": 2150
+    },
+    {
+      "entropy": 2.435756839811802,
+      "epoch": 3.843885612551463,
+      "grad_norm": 0.00811767578125,
+      "learning_rate": 1.7424444444444444e-05,
+      "loss": 3.3889,
+      "mean_token_accuracy": 0.44032932482659815,
+      "num_tokens": 57403931.0,
+      "step": 2160
+    },
+    {
+      "entropy": 2.528154679387808,
+      "epoch": 3.861689106487148,
+      "grad_norm": 0.00714111328125,
+      "learning_rate": 1.7402222222222222e-05,
+      "loss": 3.4758,
+      "mean_token_accuracy": 0.42967555299401283,
+      "num_tokens": 57663118.0,
+      "step": 2170
+    },
+    {
+      "entropy": 2.499790608137846,
+      "epoch": 3.879492600422833,
+      "grad_norm": 0.00836181640625,
+      "learning_rate": 1.7380000000000003e-05,
+      "loss": 3.4298,
+      "mean_token_accuracy": 0.43502972461283207,
+      "num_tokens": 57930933.0,
+      "step": 2180
+    },
+    {
+      "entropy": 2.515374086052179,
+      "epoch": 3.8972960943585178,
+      "grad_norm": 0.00732421875,
+      "learning_rate": 1.735777777777778e-05,
+      "loss": 3.4553,
+      "mean_token_accuracy": 0.4310147315263748,
+      "num_tokens": 58204350.0,
+      "step": 2190
+    },
+    {
+      "entropy": 2.483417534828186,
+      "epoch": 3.915099588294203,
+      "grad_norm": 0.00872802734375,
+      "learning_rate": 1.7335555555555558e-05,
+      "loss": 3.4523,
+      "mean_token_accuracy": 0.4323576783761382,
+      "num_tokens": 58455198.0,
+      "step": 2200
+    },
+    {
+      "epoch": 3.915099588294203,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 58455198.0,
+      "eval_chemistry_runtime": 42.5431,
+      "eval_chemistry_samples_per_second": 11.753,
+      "eval_chemistry_steps_per_second": 2.938,
+      "step": 2200
+    },
+    {
+      "entropy": 2.5100395753979683,
+      "epoch": 3.9329030822298874,
+      "grad_norm": 0.008056640625,
+      "learning_rate": 1.7313333333333336e-05,
+      "loss": 3.4452,
+      "mean_token_accuracy": 0.43508517276495695,
+      "num_tokens": 58728723.0,
+      "step": 2210
+    },
+    {
+      "entropy": 2.4789227142930033,
+      "epoch": 3.9507065761655724,
+      "grad_norm": 0.00872802734375,
+      "learning_rate": 1.7291111111111113e-05,
+      "loss": 3.3731,
+      "mean_token_accuracy": 0.43815778195858,
+      "num_tokens": 58995707.0,
+      "step": 2220
+    },
+    {
+      "entropy": 2.5614672005176544,
+      "epoch": 3.9685100701012574,
+      "grad_norm": 0.00921630859375,
+      "learning_rate": 1.726888888888889e-05,
+      "loss": 3.4765,
+      "mean_token_accuracy": 0.42996694296598437,
+      "num_tokens": 59266644.0,
+      "step": 2230
+    },
+    {
+      "entropy": 2.4420298121869566,
+      "epoch": 3.9863135640369425,
+      "grad_norm": 0.007415771484375,
+      "learning_rate": 1.724666666666667e-05,
+      "loss": 3.385,
+      "mean_token_accuracy": 0.4402593899518251,
+      "num_tokens": 59527189.0,
+      "step": 2240
+    },
+    {
+      "entropy": 2.4246159322800174,
+      "epoch": 4.003560698787137,
+      "grad_norm": 0.008056640625,
+      "learning_rate": 1.7224444444444446e-05,
+      "loss": 3.3519,
+      "mean_token_accuracy": 0.4434736011489745,
+      "num_tokens": 59787590.0,
+      "step": 2250
+    },
+    {
+      "entropy": 2.4212090969085693,
+      "epoch": 4.021364192722822,
+      "grad_norm": 0.0130615234375,
+      "learning_rate": 1.7202222222222224e-05,
+      "loss": 3.312,
+      "mean_token_accuracy": 0.4483447516337037,
+      "num_tokens": 60060387.0,
+      "step": 2260
+    },
+    {
+      "entropy": 2.4608243219554424,
+      "epoch": 4.039167686658507,
+      "grad_norm": 0.00897216796875,
+      "learning_rate": 1.718e-05,
+      "loss": 3.4214,
+      "mean_token_accuracy": 0.43605084978044034,
+      "num_tokens": 60323476.0,
+      "step": 2270
+    },
+    {
+      "entropy": 2.4136872582137583,
+      "epoch": 4.056971180594192,
+      "grad_norm": 0.00714111328125,
+      "learning_rate": 1.715777777777778e-05,
+      "loss": 3.3364,
+      "mean_token_accuracy": 0.444009636156261,
+      "num_tokens": 60596504.0,
+      "step": 2280
+    },
+    {
+      "entropy": 2.4391548201441764,
+      "epoch": 4.074774674529876,
+      "grad_norm": 0.0096435546875,
+      "learning_rate": 1.7135555555555557e-05,
+      "loss": 3.4052,
+      "mean_token_accuracy": 0.438773894123733,
+      "num_tokens": 60865204.0,
+      "step": 2290
+    },
+    {
+      "entropy": 2.469481632858515,
+      "epoch": 4.092578168465561,
+      "grad_norm": 0.00750732421875,
+      "learning_rate": 1.7113333333333334e-05,
+      "loss": 3.4076,
+      "mean_token_accuracy": 0.43690670263022185,
+      "num_tokens": 61136524.0,
+      "step": 2300
+    },
+    {
+      "epoch": 4.092578168465561,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 61136524.0,
+      "eval_chemistry_runtime": 42.6027,
+      "eval_chemistry_samples_per_second": 11.736,
+      "eval_chemistry_steps_per_second": 2.934,
+      "step": 2300
+    },
+    {
+      "entropy": 2.5071383990347385,
+      "epoch": 4.110381662401246,
+      "grad_norm": 0.00836181640625,
+      "learning_rate": 1.7091111111111112e-05,
+      "loss": 3.4724,
+      "mean_token_accuracy": 0.4314867680892348,
+      "num_tokens": 61400311.0,
+      "step": 2310
+    },
+    {
+      "entropy": 2.465171641111374,
+      "epoch": 4.128185156336931,
+      "grad_norm": 0.00811767578125,
+      "learning_rate": 1.706888888888889e-05,
+      "loss": 3.4016,
+      "mean_token_accuracy": 0.4373733725398779,
+      "num_tokens": 61671052.0,
+      "step": 2320
+    },
+    {
+      "entropy": 2.4659940913319587,
+      "epoch": 4.145988650272616,
+      "grad_norm": 0.00933837890625,
+      "learning_rate": 1.704666666666667e-05,
+      "loss": 3.3943,
+      "mean_token_accuracy": 0.43661461025476456,
+      "num_tokens": 61938772.0,
+      "step": 2330
+    },
+    {
+      "entropy": 2.4147595316171646,
+      "epoch": 4.163792144208301,
+      "grad_norm": 0.0108642578125,
+      "learning_rate": 1.7024444444444445e-05,
+      "loss": 3.3628,
+      "mean_token_accuracy": 0.4426847854629159,
+      "num_tokens": 62212548.0,
+      "step": 2340
+    },
+    {
+      "entropy": 2.5060284301638602,
+      "epoch": 4.181595638143985,
+      "grad_norm": 0.00799560546875,
+      "learning_rate": 1.7002222222222226e-05,
+      "loss": 3.471,
+      "mean_token_accuracy": 0.43175922725349664,
+      "num_tokens": 62469444.0,
+      "step": 2350
+    },
+    {
+      "entropy": 2.463603403419256,
+      "epoch": 4.19939913207967,
+      "grad_norm": 0.00909423828125,
+      "learning_rate": 1.698e-05,
+      "loss": 3.3622,
+      "mean_token_accuracy": 0.4391902921721339,
+      "num_tokens": 62746184.0,
+      "step": 2360
+    },
+    {
+      "entropy": 2.51062641069293,
+      "epoch": 4.2172026260153554,
+      "grad_norm": 0.01251220703125,
+      "learning_rate": 1.695777777777778e-05,
+      "loss": 3.4864,
+      "mean_token_accuracy": 0.4299986753612757,
+      "num_tokens": 63014904.0,
+      "step": 2370
+    },
+    {
+      "entropy": 2.428766914457083,
+      "epoch": 4.2350061199510405,
+      "grad_norm": 0.00927734375,
+      "learning_rate": 1.6935555555555555e-05,
+      "loss": 3.3866,
+      "mean_token_accuracy": 0.4402129024267197,
+      "num_tokens": 63286580.0,
+      "step": 2380
+    },
+    {
+      "entropy": 2.5073558762669563,
+      "epoch": 4.2528096138867255,
+      "grad_norm": 0.00811767578125,
+      "learning_rate": 1.6913333333333336e-05,
+      "loss": 3.4558,
+      "mean_token_accuracy": 0.4320449620485306,
+      "num_tokens": 63544361.0,
+      "step": 2390
+    },
+    {
+      "entropy": 2.427883168309927,
+      "epoch": 4.2706131078224105,
+      "grad_norm": 0.01055908203125,
+      "learning_rate": 1.689111111111111e-05,
+      "loss": 3.3615,
+      "mean_token_accuracy": 0.4403992747887969,
+      "num_tokens": 63813469.0,
+      "step": 2400
+    },
+    {
+      "epoch": 4.2706131078224105,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 63813469.0,
+      "eval_chemistry_runtime": 42.6226,
+      "eval_chemistry_samples_per_second": 11.731,
+      "eval_chemistry_steps_per_second": 2.933,
+      "step": 2400
+    },
+    {
+      "entropy": 2.429547951370478,
+      "epoch": 4.288416601758095,
+      "grad_norm": 0.007659912109375,
+      "learning_rate": 1.686888888888889e-05,
+      "loss": 3.3793,
+      "mean_token_accuracy": 0.44148991294205187,
+      "num_tokens": 64074240.0,
+      "step": 2410
+    },
+    {
+      "entropy": 2.4751460015773774,
+      "epoch": 4.30622009569378,
+      "grad_norm": 0.011962890625,
+      "learning_rate": 1.684666666666667e-05,
+      "loss": 3.3702,
+      "mean_token_accuracy": 0.4387602785602212,
+      "num_tokens": 64338220.0,
+      "step": 2420
+    },
+    {
+      "entropy": 2.445548442751169,
+      "epoch": 4.324023589629465,
+      "grad_norm": 0.0113525390625,
+      "learning_rate": 1.6824444444444447e-05,
+      "loss": 3.3907,
+      "mean_token_accuracy": 0.4380012461915612,
+      "num_tokens": 64612457.0,
+      "step": 2430
+    },
+    {
+      "entropy": 2.426620189845562,
+      "epoch": 4.34182708356515,
+      "grad_norm": 0.010498046875,
+      "learning_rate": 1.6802222222222224e-05,
+      "loss": 3.3617,
+      "mean_token_accuracy": 0.44086786955595014,
+      "num_tokens": 64876908.0,
+      "step": 2440
+    },
+    {
+      "entropy": 2.440635305643082,
+      "epoch": 4.359630577500835,
+      "grad_norm": 0.0089111328125,
+      "learning_rate": 1.6780000000000002e-05,
+      "loss": 3.3926,
+      "mean_token_accuracy": 0.43678140845149754,
+      "num_tokens": 65147596.0,
+      "step": 2450
+    },
+    {
+      "entropy": 2.5261649206280707,
+      "epoch": 4.37743407143652,
+      "grad_norm": 0.00872802734375,
+      "learning_rate": 1.675777777777778e-05,
+      "loss": 3.4782,
+      "mean_token_accuracy": 0.42467477526515723,
+      "num_tokens": 65418480.0,
+      "step": 2460
+    },
+    {
+      "entropy": 2.466297345608473,
+      "epoch": 4.395237565372204,
+      "grad_norm": 0.007171630859375,
+      "learning_rate": 1.6735555555555557e-05,
+      "loss": 3.4087,
+      "mean_token_accuracy": 0.4363992527127266,
+      "num_tokens": 65680301.0,
+      "step": 2470
+    },
+    {
+      "entropy": 2.4904259890317917,
+      "epoch": 4.413041059307889,
+      "grad_norm": 0.00860595703125,
+      "learning_rate": 1.6713333333333335e-05,
+      "loss": 3.4365,
+      "mean_token_accuracy": 0.4327126918360591,
+      "num_tokens": 65945233.0,
+      "step": 2480
+    },
+    {
+      "entropy": 2.4659158870577813,
+      "epoch": 4.430844553243574,
+      "grad_norm": 0.01068115234375,
+      "learning_rate": 1.6691111111111112e-05,
+      "loss": 3.4149,
+      "mean_token_accuracy": 0.4362122168764472,
+      "num_tokens": 66209932.0,
+      "step": 2490
+    },
+    {
+      "entropy": 2.506372135132551,
+      "epoch": 4.448648047179259,
+      "grad_norm": 0.00885009765625,
+      "learning_rate": 1.666888888888889e-05,
+      "loss": 3.4514,
+      "mean_token_accuracy": 0.43281122148036955,
+      "num_tokens": 66463918.0,
+      "step": 2500
+    },
+    {
+      "epoch": 4.448648047179259,
+      "eval_chemistry_entropy": 2.4951189584732054,
+      "eval_chemistry_loss": 3.512981414794922,
+      "eval_chemistry_mean_token_accuracy": 0.4329572098255157,
+      "eval_chemistry_num_tokens": 66463918.0,
+      "eval_chemistry_runtime": 42.5684,
+      "eval_chemistry_samples_per_second": 11.746,
+      "eval_chemistry_steps_per_second": 2.936,
+      "step": 2500
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 10000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 18,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 6.687183122057217e+18,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c203431b23c0b2024be20ba2c3b8d6912d99b5a0839b2ccb6094a319dbeed0e
+size 6481