Publish FinPath AutoScientist model artifact

.gitattributes

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

README.md

@@ -0,0 +1,35 @@
+---
+base_model: openai/gpt-oss-20b
+library_name: peft
+license: apache-2.0
+language:
+- en
+tags:
+- adaption
+- autoscientist
+- personal-finance
+- lora
+---
+
+# FinPath AutoScientist Model
+
+LoRA adapter trained with Adaption AutoScientist for grounded personal-finance reasoning.
+
+## Base Model
+
+`openai/gpt-oss-20b`
+
+## Data
+
+- Source: https://huggingface.co/datasets/doraking/finpath-autoscientist-data
+- Adapted: https://huggingface.co/datasets/doraking/finpath-autoscientist-adapted
+
+## Intended Behavior
+
+Return the direct result, show the calculation, explain tradeoffs, provide practical next steps,
+and distinguish educational information from individualized financial advice.
+
+## Limitations
+
+Validate calculations and applicable rules before real-world use. This model is not a financial
+adviser and does not account for every jurisdiction, product, or personal circumstance.

adapter_config.json

@@ -0,0 +1,36 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "togethercomputer/gpt-oss-20b-bf16",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": [],
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 128,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj",
+    "o_proj",
+    "k_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a48c6a05afda8bd9d87904289f9505888354dced9167561e5badf4cc36079978
+size 127427864

chat_template.jinja

@@ -0,0 +1,397 @@
+{#-
+  In addition to the normal inputs of `messages` and `tools`, this template also accepts the
+  following kwargs:
+  - "builtin_tools": A list, can contain "browser" and/or "python".
+  - "model_identity": A string that optionally describes the model identity.
+  - "reasoning_effort": A string that describes the reasoning effort, defaults to "medium".
+ #}
+
+{#- Tool Definition Rendering ============================================== #}
+{%- macro render_typescript_type(param_spec, required_params, is_nullable=false) -%}
+    {%- if param_spec.type == "array" -%}
+        {%- if param_spec['items'] -%}
+            {%- if param_spec['items']['type'] == "string" -%}
+                {{- "string[]" }}
+            {%- elif param_spec['items']['type'] == "number" -%}
+                {{- "number[]" }}
+            {%- elif param_spec['items']['type'] == "integer" -%}
+                {{- "number[]" }}
+            {%- elif param_spec['items']['type'] == "boolean" -%}
+                {{- "boolean[]" }}
+            {%- else -%}
+                {%- set inner_type = render_typescript_type(param_spec['items'], required_params) -%}
+                {%- if inner_type == "object | object" or inner_type|length > 50 -%}
+                    {{- "any[]" }}
+                {%- else -%}
+                    {{- inner_type + "[]" }}
+                {%- endif -%}
+            {%- endif -%}
+            {%- if param_spec.nullable -%}
+                {{- " | null" }}
+            {%- endif -%}
+        {%- else -%}
+            {{- "any[]" }}
+            {%- if param_spec.nullable -%}
+                {{- " | null" }}
+            {%- endif -%}
+        {%- endif -%}
+    {%- elif param_spec.type is defined and param_spec.type is iterable and param_spec.type is not string and param_spec.type is not mapping and param_spec.type[0] is defined -%}
+        {#- Handle array of types like ["object", "object"] from Union[dict, list] #}
+        {%- if param_spec.type | length > 1 -%}
+            {{- param_spec.type | join(" | ") }}
+        {%- else -%}
+            {{- param_spec.type[0] }}
+        {%- endif -%}
+    {%- elif param_spec.oneOf -%}
+        {#- Handle oneOf schemas - check for complex unions and fallback to any #}
+        {%- set has_object_variants = false -%}
+        {%- for variant in param_spec.oneOf -%}
+            {%- if variant.type == "object" -%}
+                {%- set has_object_variants = true -%}
+            {%- endif -%}
+        {%- endfor -%}
+        {%- if has_object_variants and param_spec.oneOf|length > 1 -%}
+            {{- "any" }}
+        {%- else -%}
+            {%- for variant in param_spec.oneOf -%}
+                {{- render_typescript_type(variant, required_params) -}}
+                {%- if variant.description %}
+                    {{- "// " + variant.description }}
+                {%- endif -%}
+                {%- if variant.default is defined %}
+                    {{ "// default: " + variant.default|tojson }}
+                {%- endif -%}
+                {%- if not loop.last %}
+                    {{- " | " }}
+                {% endif -%}
+            {%- endfor -%}
+        {%- endif -%}
+    {%- elif param_spec.type == "string" -%}
+        {%- if param_spec.enum -%}
+            {{- '"' + param_spec.enum|join('" | "') + '"' -}}
+        {%- else -%}
+            {{- "string" }}
+            {%- if param_spec.nullable %}
+                {{- " | null" }}
+            {%- endif -%}
+        {%- endif -%}
+    {%- elif param_spec.type == "number" -%}
+        {{- "number" }}
+    {%- elif param_spec.type == "integer" -%}
+        {{- "number" }}
+    {%- elif param_spec.type == "boolean" -%}
+        {{- "boolean" }}
+
+    {%- elif param_spec.type == "object" -%}
+        {%- if param_spec.properties -%}
+            {{- "{
+" }}
+            {%- for prop_name, prop_spec in param_spec.properties.items() -%}
+                {{- prop_name -}}
+                {%- if prop_name not in (param_spec.required or []) -%}
+                    {{- "?" }}
+                {%- endif -%}
+                {{- ": " }}
+                {{ render_typescript_type(prop_spec, param_spec.required or []) }}
+                {%- if not loop.last -%}
+                    {{-", " }}
+                {%- endif -%}
+            {%- endfor -%}
+            {{- "}" }}
+        {%- else -%}
+            {{- "object" }}
+        {%- endif -%}
+    {%- else -%}
+        {{- "any" }}
+    {%- endif -%}
+{%- endmacro -%}
+
+{%- macro render_tool_namespace(namespace_name, tools) -%}
+    {{- "## " + namespace_name + "
+
+" }}
+    {{- "namespace " + namespace_name + " {
+
+" }}
+    {%- for tool in tools %}
+        {%- set tool = tool.function %}
+        {{- "// " + tool.description + "
+" }}
+        {{- "type "+ tool.name + " = " }}
+        {%- if tool.parameters and tool.parameters.properties %}
+            {{- "(_: {
+" }}
+            {%- for param_name, param_spec in tool.parameters.properties.items() %}
+                {%- if param_spec.description %}
+                    {{- "// " + param_spec.description + "
+" }}
+                {%- endif %}
+                {{- param_name }}
+                {%- if param_name not in (tool.parameters.required or []) -%}
+                    {{- "?" }}
+                {%- endif -%}
+                {{- ": " }}
+                {{- render_typescript_type(param_spec, tool.parameters.required or []) }}
+                {%- if param_spec.default is defined -%}
+                    {%- if param_spec.enum %}
+                        {{- ", // default: " + param_spec.default }}
+                    {%- elif param_spec.oneOf %}
+                        {{- "// default: " + param_spec.default }}
+                    {%- else %}
+                        {{- ", // default: " + param_spec.default|tojson }}
+                    {%- endif -%}
+                {%- endif -%}
+                {%- if not loop.last %}
+                    {{- ",
+" }}
+                {%- else %}
+                    {{- "
+" }}
+                {%- endif -%}
+            {%- endfor %}
+            {{- "}) => any;
+
+" }}
+        {%- else -%}
+            {{- "() => any;
+
+" }}
+        {%- endif -%}
+    {%- endfor %}
+    {{- "} // namespace " + namespace_name }}
+{%- endmacro -%}
+
+{%- macro render_builtin_tools(browser_tool, python_tool) -%}
+    {%- if browser_tool %}
+        {{- "## browser
+
+" }}
+        {{- "// Tool for browsing.
+" }}
+        {{- "// The `cursor` appears in brackets before each browsing display: `[{cursor}]`.
+" }}
+        {{- "// Cite information from the tool using the following format:
+" }}
+        {{- "// `【{cursor}†L{line_start}(-L{line_end})?】`, for example: `【6†L9-L11】` or `【8†L3】`.
+" }}
+        {{- "// Do not quote more than 10 words directly from the tool output.
+" }}
+        {{- "// sources=web (default: web)
+" }}
+        {{- "namespace browser {
+
+" }}
+        {{- "// Searches for information related to `query` and displays `topn` results.
+" }}
+        {{- "type search = (_: {
+" }}
+        {{- "query: string,
+" }}
+        {{- "topn?: number, // default: 10
+" }}
+        {{- "source?: string,
+" }}
+        {{- "}) => any;
+
+" }}
+        {{- "// Opens the link `id` from the page indicated by `cursor` starting at line number `loc`, showing `num_lines` lines.
+" }}
+        {{- "// Valid link ids are displayed with the formatting: `【{id}†.*】`.
+" }}
+        {{- "// If `cursor` is not provided, the most recent page is implied.
+" }}
+        {{- "// If `id` is a string, it is treated as a fully qualified URL associated with `source`.
+" }}
+        {{- "// If `loc` is not provided, the viewport will be positioned at the beginning of the document or centered on the most relevant passage, if available.
+" }}
+        {{- "// Use this function without `id` to scroll to a new location of an opened page.
+" }}
+        {{- "type open = (_: {
+" }}
+        {{- "id?: number | string, // default: -1
+" }}
+        {{- "cursor?: number, // default: -1
+" }}
+        {{- "loc?: number, // default: -1
+" }}
+        {{- "num_lines?: number, // default: -1
+" }}
+        {{- "view_source?: boolean, // default: false
+" }}
+        {{- "source?: string,
+" }}
+        {{- "}) => any;
+
+" }}
+        {{- "// Finds exact matches of `pattern` in the current page, or the page given by `cursor`.
+" }}
+        {{- "type find = (_: {
+" }}
+        {{- "pattern: string,
+" }}
+        {{- "cursor?: number, // default: -1
+" }}
+        {{- "}) => any;
+
+" }}
+        {{- "} // namespace browser
+
+" }}
+    {%- endif -%}
+
+    {%- if python_tool %}
+        {{- "## python
+
+" }}
+        {{- "Use this tool to execute Python code in your chain of thought. The code will not be shown to the user. This tool should be used for internal reasoning, but not for code that is intended to be visible to the user (e.g. when creating plots, tables, or files).
+
+" }}
+        {{- "When you send a message containing Python code to python, it will be executed in a stateful Jupyter notebook environment. python will respond with the output of the execution or time out after 120.0 seconds. The drive at '/mnt/data' can be used to save and persist user files. Internet access for this session is UNKNOWN. Depends on the cluster.
+
+" }}
+    {%- endif -%}
+{%- endmacro -%}
+
+{#- System Message Construction ============================================ #}
+{%- macro build_system_message() -%}
+    {%- if model_identity is not defined %}
+        {%- set model_identity = "You are ChatGPT, a large language model trained by OpenAI." %}
+    {%- endif %}
+    {{- model_identity + "
+" }}
+    {{- "Knowledge cutoff: 2024-06
+" }}
+    {{- "Current date: " + strftime_now("%Y-%m-%d") + "
+
+" }}
+    {%- if reasoning_effort is not defined %}
+        {%- set reasoning_effort = "medium" %}
+    {%- endif %}
+    {{- "Reasoning: " + reasoning_effort + "
+
+" }}
+    {%- if builtin_tools %}
+        {{- "# Tools
+
+" }}
+        {%- set available_builtin_tools = namespace(browser=false, python=false) %}
+        {%- for tool in builtin_tools %}
+            {%- if tool == "browser" %}
+                {%- set available_builtin_tools.browser = true %}
+            {%- elif tool == "python" %}
+                {%- set available_builtin_tools.python = true %}
+            {%- endif %}
+        {%- endfor %}
+        {{- render_builtin_tools(available_builtin_tools.browser, available_builtin_tools.python) }}
+    {%- endif -%}
+    {{- "# Valid channels: analysis, commentary, final. Channel must be included for every message." }}
+    {%- if tools -%}
+        {{- "
+Calls to these tools must go to the commentary channel: 'functions'." }}
+    {%- endif -%}
+{%- endmacro -%}
+
+{#- Main Template Logic ================================================= #}
+{#- Set defaults #}
+
+{#- Render system message #}
+{{- "<|start|>system<|message|>" }}
+{{- build_system_message() }}
+{{- "<|end|>" }}
+
+{#- Extract developer message #}
+{%- if messages[0].role == "developer" or messages[0].role == "system" %}
+    {%- set developer_message = messages[0].content %}
+    {%- set loop_messages = messages[1:] %}
+{%- else %}
+    {%- set developer_message = "" %}
+    {%- set loop_messages = messages %}
+{%- endif %}
+
+{#- Render developer message #}
+{%- if developer_message or tools %}
+    {{- "<|start|>developer<|message|>" }}
+    {%- if developer_message %}
+        {{- "# Instructions
+
+" }}
+        {{- developer_message }}
+    {%- endif %}
+    {%- if tools -%}
+        {{- "
+
+" }}
+        {{- "# Tools
+
+" }}
+        {{- render_tool_namespace("functions", tools) }}
+    {%- endif -%}
+    {{- "<|end|>" }}
+{%- endif %}
+
+{#- Render messages #}
+{%- set last_tool_call = namespace(name=none) %}
+{%- for message in loop_messages -%}
+    {#- At this point only assistant/user/tool messages should remain #}
+    {%- if message.role == 'assistant' -%}
+        {#- Checks to ensure the messages are being passed in the format we expect #}
+        {%- if "content" in message %}
+            {%- if "<|channel|>analysis<|message|>" in message.content or "<|channel|>final<|message|>" in message.content %}
+                {{- raise_exception("You have passed a message containing <|channel|> tags in the content field. Instead of doing this, you should pass analysis messages (the string between '<|message|>' and '<|end|>') in the 'thinking' field, and final messages (the string between '<|message|>' and '<|end|>') in the 'content' field.") }}
+            {%- endif %}
+        {%- endif %}
+        {%- if "thinking" in message %}
+            {%- if "<|channel|>analysis<|message|>" in message.thinking or "<|channel|>final<|message|>" in message.thinking %}
+                {{- raise_exception("You have passed a message containing <|channel|> tags in the thinking field. Instead of doing this, you should pass analysis messages (the string between '<|message|>' and '<|end|>') in the 'thinking' field, and final messages (the string between '<|message|>' and '<|end|>') in the 'content' field.") }}
+            {%- endif %}
+        {%- endif %}
+        {%- if "tool_calls" in message %}
+            {#- We assume max 1 tool call per message, and so we infer the tool call name #}
+            {#- in "tool" messages from the most recent assistant tool call name #}
+            {%- set tool_call = message.tool_calls[0] %}
+            {%- if tool_call.function %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {%- if message.content and message.thinking %}
+                {{- raise_exception("Cannot pass both content and thinking in an assistant message with tool calls! Put the analysis message in one or the other, but not both.") }}
+            {%- elif message.content %}
+                {{- "<|start|>assistant<|channel|>analysis<|message|>" + message.content + "<|end|>" }}
+            {%- elif message.thinking %}
+                {{- "<|start|>assistant<|channel|>analysis<|message|>" + message.thinking + "<|end|>" }}
+            {%- endif %}
+            {{- "<|start|>assistant to=" }}
+            {{- "functions." + tool_call.name + "<|channel|>commentary " }}
+            {{- (tool_call.content_type if tool_call.content_type is defined else "json") + "<|message|>" }}
+            {{- tool_call.arguments|tojson }}
+            {{- "<|call|>" }}
+            {%- set last_tool_call.name = tool_call.name %}
+        {%- elif loop.last and not add_generation_prompt %}
+            {#- Only render the CoT if the final turn is an assistant turn and add_generation_prompt is false #}
+            {#- This is a situation that should only occur in training, never in inference. #}
+            {%- if "thinking" in message %}
+                {{- "<|start|>assistant<|channel|>analysis<|message|>" + message.thinking + "<|end|>" }}
+            {%- endif %}
+            {#- <|return|> indicates the end of generation, but <|end|> does not #}
+            {#- <|return|> should never be an input to the model, but we include it as the final token #}
+            {#- when training, so the model learns to emit it. #}
+            {{- "<|start|>assistant<|channel|>final<|message|>" + message.content + "<|return|>" }}
+        {%- else %}
+            {#- CoT is dropped during all previous turns, so we never render it for inference #}
+            {{- "<|start|>assistant<|channel|>final<|message|>" + message.content + "<|end|>" }}
+            {%- set last_tool_call.name = none %}
+        {%- endif %}
+    {%- elif message.role == 'tool' -%}
+        {%- if last_tool_call.name is none %}
+            {{- raise_exception("Message has tool role, but there was no previous assistant message with a tool call!") }}
+        {%- endif %}
+        {{- "<|start|>functions." + last_tool_call.name }}
+        {{- " to=assistant<|channel|>commentary<|message|>" + message.content|tojson + "<|end|>" }}
+    {%- elif message.role == 'user' -%}
+        {{- "<|start|>user<|message|>" + message.content + "<|end|>" }}
+    {%- endif -%}
+{%- endfor -%}
+
+{#- Generation prompt #}
+{%- if add_generation_prompt -%}
+<|start|>assistant
+{%- endif -%}

config.json

@@ -0,0 +1,70 @@
+{
+  "architectures": [
+    "GptOssForCausalLM"
+  ],
+  "attention_bias": true,
+  "attention_dropout": 0.0,
+  "bos_token_id": 199998,
+  "dtype": "bfloat16",
+  "eos_token_id": 200002,
+  "experts_per_token": 4,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 2880,
+  "initial_context_length": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 2880,
+  "layer_types": [
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 131072,
+  "model_type": "gpt_oss",
+  "num_attention_heads": 64,
+  "num_experts_per_tok": 4,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 8,
+  "num_local_experts": 32,
+  "output_router_logits": false,
+  "pad_token_id": 199999,
+  "rms_norm_eps": 1e-05,
+  "rope_parameters": {
+    "beta_fast": 32.0,
+    "beta_slow": 1.0,
+    "factor": 32.0,
+    "original_max_position_embeddings": 4096,
+    "rope_theta": 150000,
+    "rope_type": "yarn",
+    "truncate": false
+  },
+  "router_aux_loss_coef": 0.9,
+  "sliding_window": 128,
+  "swiglu_limit": 7.0,
+  "tie_word_embeddings": false,
+  "transformers_version": "5.10.1",
+  "use_cache": false,
+  "vocab_size": 201088,
+  "torch_dtype": "bfloat16"
+}

dataset-metadata.json

@@ -0,0 +1,11 @@
+{
+  "title": "FinPath AutoScientist Model",
+  "id": "dorakingx/finpath-autoscientist-model",
+  "licenses": [
+    {
+      "name": "Apache 2.0"
+    }
+  ],
+  "subtitle": "AutoScientist LoRA for grounded personal finance reasoning",
+  "description": "LoRA adapter weights, model card, and training metadata."
+}

special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "bos_token": "<|startoftext|>",
+  "eos_token": "<|return|>",
+  "pad_token": "<|endoftext|>"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0614fe83cadab421296e664e1f48f4261fa8fef6e03e63bb75c20f38e37d07d3
+size 27868174

tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|startoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|return|>",
+  "is_local": false,
+  "local_files_only": true,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "tokenizer_class": "TokenizersBackend"
+}

trainer_state.json

@@ -0,0 +1,2524 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 70,
+  "global_step": 350,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.014311270125223614,
+      "grad_norm": 3.3410181999206543,
+      "learning_rate": 0.0,
+      "loss": 2.01751708984375,
+      "step": 1
+    },
+    {
+      "epoch": 0.028622540250447227,
+      "grad_norm": 3.347492218017578,
+      "learning_rate": 9.090909090909091e-06,
+      "loss": 1.5863037109375,
+      "step": 2
+    },
+    {
+      "epoch": 0.04293381037567084,
+      "grad_norm": 3.564649820327759,
+      "learning_rate": 1.8181818181818182e-05,
+      "loss": 2.1478271484375,
+      "step": 3
+    },
+    {
+      "epoch": 0.057245080500894455,
+      "grad_norm": 1.6348196268081665,
+      "learning_rate": 2.7272727272727273e-05,
+      "loss": 1.59539794921875,
+      "step": 4
+    },
+    {
+      "epoch": 0.07155635062611806,
+      "grad_norm": 0.5400568246841431,
+      "learning_rate": 3.6363636363636364e-05,
+      "loss": 1.4581298828125,
+      "step": 5
+    },
+    {
+      "epoch": 0.08586762075134168,
+      "grad_norm": 0.33098214864730835,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": 1.194580078125,
+      "step": 6
+    },
+    {
+      "epoch": 0.1001788908765653,
+      "grad_norm": 0.3238024115562439,
+      "learning_rate": 5.4545454545454546e-05,
+      "loss": 1.29010009765625,
+      "step": 7
+    },
+    {
+      "epoch": 0.11449016100178891,
+      "grad_norm": 0.2658863663673401,
+      "learning_rate": 6.363636363636364e-05,
+      "loss": 1.26934814453125,
+      "step": 8
+    },
+    {
+      "epoch": 0.12880143112701253,
+      "grad_norm": 0.22492113709449768,
+      "learning_rate": 7.272727272727273e-05,
+      "loss": 1.147308349609375,
+      "step": 9
+    },
+    {
+      "epoch": 0.14311270125223613,
+      "grad_norm": 0.36609947681427,
+      "learning_rate": 8.181818181818183e-05,
+      "loss": 1.25152587890625,
+      "step": 10
+    },
+    {
+      "epoch": 0.15742397137745975,
+      "grad_norm": 0.34992608428001404,
+      "learning_rate": 9.090909090909092e-05,
+      "loss": 1.27337646484375,
+      "step": 11
+    },
+    {
+      "epoch": 0.17173524150268335,
+      "grad_norm": 0.27957606315612793,
+      "learning_rate": 0.0001,
+      "loss": 1.093017578125,
+      "step": 12
+    },
+    {
+      "epoch": 0.18604651162790697,
+      "grad_norm": 0.2911664843559265,
+      "learning_rate": 9.999806767684109e-05,
+      "loss": 1.251129150390625,
+      "step": 13
+    },
+    {
+      "epoch": 0.2003577817531306,
+      "grad_norm": 0.16858641803264618,
+      "learning_rate": 9.999227087331426e-05,
+      "loss": 1.172821044921875,
+      "step": 14
+    },
+    {
+      "epoch": 0.2146690518783542,
+      "grad_norm": 0.15052799880504608,
+      "learning_rate": 9.998261008725495e-05,
+      "loss": 1.2061767578125,
+      "step": 15
+    },
+    {
+      "epoch": 0.22898032200357782,
+      "grad_norm": 0.12794822454452515,
+      "learning_rate": 9.996908614834141e-05,
+      "loss": 1.0494766235351562,
+      "step": 16
+    },
+    {
+      "epoch": 0.24329159212880144,
+      "grad_norm": 0.1232319176197052,
+      "learning_rate": 9.995170021802342e-05,
+      "loss": 1.1510009765625,
+      "step": 17
+    },
+    {
+      "epoch": 0.25760286225402507,
+      "grad_norm": 0.11607693135738373,
+      "learning_rate": 9.993045378942262e-05,
+      "loss": 1.22509765625,
+      "step": 18
+    },
+    {
+      "epoch": 0.27191413237924866,
+      "grad_norm": 0.10980455577373505,
+      "learning_rate": 9.990534868720414e-05,
+      "loss": 1.2623291015625,
+      "step": 19
+    },
+    {
+      "epoch": 0.28622540250447226,
+      "grad_norm": 0.09627101570367813,
+      "learning_rate": 9.987638706741999e-05,
+      "loss": 1.2161865234375,
+      "step": 20
+    },
+    {
+      "epoch": 0.3005366726296959,
+      "grad_norm": 0.07417790591716766,
+      "learning_rate": 9.984357141732388e-05,
+      "loss": 1.08526611328125,
+      "step": 21
+    },
+    {
+      "epoch": 0.3148479427549195,
+      "grad_norm": 0.06646407395601273,
+      "learning_rate": 9.980690455515766e-05,
+      "loss": 1.020904541015625,
+      "step": 22
+    },
+    {
+      "epoch": 0.3291592128801431,
+      "grad_norm": 0.07123072445392609,
+      "learning_rate": 9.976638962990915e-05,
+      "loss": 1.06396484375,
+      "step": 23
+    },
+    {
+      "epoch": 0.3434704830053667,
+      "grad_norm": 0.07855740189552307,
+      "learning_rate": 9.972203012104187e-05,
+      "loss": 0.968017578125,
+      "step": 24
+    },
+    {
+      "epoch": 0.35778175313059035,
+      "grad_norm": 0.0795491561293602,
+      "learning_rate": 9.967382983819604e-05,
+      "loss": 1.01934814453125,
+      "step": 25
+    },
+    {
+      "epoch": 0.37209302325581395,
+      "grad_norm": 0.07461606711149216,
+      "learning_rate": 9.96217929208616e-05,
+      "loss": 1.0777130126953125,
+      "step": 26
+    },
+    {
+      "epoch": 0.38640429338103754,
+      "grad_norm": 0.07724579423666,
+      "learning_rate": 9.956592383802259e-05,
+      "loss": 1.1514892578125,
+      "step": 27
+    },
+    {
+      "epoch": 0.4007155635062612,
+      "grad_norm": 0.04926382377743721,
+      "learning_rate": 9.950622738777333e-05,
+      "loss": 1.00640869140625,
+      "step": 28
+    },
+    {
+      "epoch": 0.4150268336314848,
+      "grad_norm": 0.04825359955430031,
+      "learning_rate": 9.94427086969064e-05,
+      "loss": 0.926544189453125,
+      "step": 29
+    },
+    {
+      "epoch": 0.4293381037567084,
+      "grad_norm": 0.04287895932793617,
+      "learning_rate": 9.937537322047238e-05,
+      "loss": 0.8801422119140625,
+      "step": 30
+    },
+    {
+      "epoch": 0.44364937388193204,
+      "grad_norm": 0.04662016034126282,
+      "learning_rate": 9.930422674131127e-05,
+      "loss": 1.0087890625,
+      "step": 31
+    },
+    {
+      "epoch": 0.45796064400715564,
+      "grad_norm": 0.06877255439758301,
+      "learning_rate": 9.922927536955593e-05,
+      "loss": 1.153106689453125,
+      "step": 32
+    },
+    {
+      "epoch": 0.47227191413237923,
+      "grad_norm": 0.045458752661943436,
+      "learning_rate": 9.915052554210732e-05,
+      "loss": 1.005767822265625,
+      "step": 33
+    },
+    {
+      "epoch": 0.4865831842576029,
+      "grad_norm": 0.05774962529540062,
+      "learning_rate": 9.906798402208168e-05,
+      "loss": 1.18359375,
+      "step": 34
+    },
+    {
+      "epoch": 0.5008944543828264,
+      "grad_norm": 0.04781227558851242,
+      "learning_rate": 9.89816578982297e-05,
+      "loss": 0.91656494140625,
+      "step": 35
+    },
+    {
+      "epoch": 0.5152057245080501,
+      "grad_norm": 0.05493216589093208,
+      "learning_rate": 9.889155458432776e-05,
+      "loss": 0.9749755859375,
+      "step": 36
+    },
+    {
+      "epoch": 0.5295169946332737,
+      "grad_norm": 0.05125086382031441,
+      "learning_rate": 9.87976818185412e-05,
+      "loss": 1.14154052734375,
+      "step": 37
+    },
+    {
+      "epoch": 0.5438282647584973,
+      "grad_norm": 0.04450112581253052,
+      "learning_rate": 9.870004766275977e-05,
+      "loss": 1.10321044921875,
+      "step": 38
+    },
+    {
+      "epoch": 0.5581395348837209,
+      "grad_norm": 0.044988811016082764,
+      "learning_rate": 9.859866050190522e-05,
+      "loss": 0.89453125,
+      "step": 39
+    },
+    {
+      "epoch": 0.5724508050089445,
+      "grad_norm": 0.04589870572090149,
+      "learning_rate": 9.849352904321135e-05,
+      "loss": 1.0570068359375,
+      "step": 40
+    },
+    {
+      "epoch": 0.5867620751341681,
+      "grad_norm": 0.04538784548640251,
+      "learning_rate": 9.838466231547601e-05,
+      "loss": 0.96771240234375,
+      "step": 41
+    },
+    {
+      "epoch": 0.6010733452593918,
+      "grad_norm": 0.05058218166232109,
+      "learning_rate": 9.827206966828581e-05,
+      "loss": 1.019683837890625,
+      "step": 42
+    },
+    {
+      "epoch": 0.6153846153846154,
+      "grad_norm": 0.04873070865869522,
+      "learning_rate": 9.815576077121321e-05,
+      "loss": 0.9384994506835938,
+      "step": 43
+    },
+    {
+      "epoch": 0.629695885509839,
+      "grad_norm": 0.04834091663360596,
+      "learning_rate": 9.803574561298599e-05,
+      "loss": 1.029266357421875,
+      "step": 44
+    },
+    {
+      "epoch": 0.6440071556350626,
+      "grad_norm": 0.05508863180875778,
+      "learning_rate": 9.79120345006295e-05,
+      "loss": 1.00396728515625,
+      "step": 45
+    },
+    {
+      "epoch": 0.6583184257602862,
+      "grad_norm": 0.04428449273109436,
+      "learning_rate": 9.778463805858136e-05,
+      "loss": 1.0389404296875,
+      "step": 46
+    },
+    {
+      "epoch": 0.6726296958855098,
+      "grad_norm": 0.0409366674721241,
+      "learning_rate": 9.765356722777916e-05,
+      "loss": 0.8250885009765625,
+      "step": 47
+    },
+    {
+      "epoch": 0.6869409660107334,
+      "grad_norm": 0.046333007514476776,
+      "learning_rate": 9.751883326472077e-05,
+      "loss": 0.934814453125,
+      "step": 48
+    },
+    {
+      "epoch": 0.7012522361359571,
+      "grad_norm": 0.04506774619221687,
+      "learning_rate": 9.738044774049758e-05,
+      "loss": 0.97210693359375,
+      "step": 49
+    },
+    {
+      "epoch": 0.7155635062611807,
+      "grad_norm": 0.04564470425248146,
+      "learning_rate": 9.72384225398009e-05,
+      "loss": 0.970306396484375,
+      "step": 50
+    },
+    {
+      "epoch": 0.7298747763864043,
+      "grad_norm": 0.04502520710229874,
+      "learning_rate": 9.709276985990111e-05,
+      "loss": 0.94805908203125,
+      "step": 51
+    },
+    {
+      "epoch": 0.7441860465116279,
+      "grad_norm": 0.038036081939935684,
+      "learning_rate": 9.69435022096003e-05,
+      "loss": 0.768280029296875,
+      "step": 52
+    },
+    {
+      "epoch": 0.7584973166368515,
+      "grad_norm": 0.055236611515283585,
+      "learning_rate": 9.67906324081579e-05,
+      "loss": 1.01470947265625,
+      "step": 53
+    },
+    {
+      "epoch": 0.7728085867620751,
+      "grad_norm": 0.03723809868097305,
+      "learning_rate": 9.663417358418981e-05,
+      "loss": 0.9454345703125,
+      "step": 54
+    },
+    {
+      "epoch": 0.7871198568872988,
+      "grad_norm": 0.041100673377513885,
+      "learning_rate": 9.64741391745409e-05,
+      "loss": 0.86029052734375,
+      "step": 55
+    },
+    {
+      "epoch": 0.8014311270125224,
+      "grad_norm": 0.0422760546207428,
+      "learning_rate": 9.631054292313094e-05,
+      "loss": 0.9407958984375,
+      "step": 56
+    },
+    {
+      "epoch": 0.815742397137746,
+      "grad_norm": 0.037724003195762634,
+      "learning_rate": 9.614339887977443e-05,
+      "loss": 0.7969350218772888,
+      "step": 57
+    },
+    {
+      "epoch": 0.8300536672629696,
+      "grad_norm": 0.04163903743028641,
+      "learning_rate": 9.597272139897384e-05,
+      "loss": 0.86175537109375,
+      "step": 58
+    },
+    {
+      "epoch": 0.8443649373881932,
+      "grad_norm": 0.05268585681915283,
+      "learning_rate": 9.57985251386869e-05,
+      "loss": 0.927215576171875,
+      "step": 59
+    },
+    {
+      "epoch": 0.8586762075134168,
+      "grad_norm": 0.046867866069078445,
+      "learning_rate": 9.56208250590677e-05,
+      "loss": 0.91168212890625,
+      "step": 60
+    },
+    {
+      "epoch": 0.8729874776386404,
+      "grad_norm": 0.04723203554749489,
+      "learning_rate": 9.543963642118205e-05,
+      "loss": 1.0142822265625,
+      "step": 61
+    },
+    {
+      "epoch": 0.8872987477638641,
+      "grad_norm": 0.042484745383262634,
+      "learning_rate": 9.525497478569663e-05,
+      "loss": 0.91455078125,
+      "step": 62
+    },
+    {
+      "epoch": 0.9016100178890877,
+      "grad_norm": 0.045855678617954254,
+      "learning_rate": 9.506685601154276e-05,
+      "loss": 0.9036865234375,
+      "step": 63
+    },
+    {
+      "epoch": 0.9159212880143113,
+      "grad_norm": 0.049025025218725204,
+      "learning_rate": 9.487529625455443e-05,
+      "loss": 1.02838134765625,
+      "step": 64
+    },
+    {
+      "epoch": 0.9302325581395349,
+      "grad_norm": 0.05009147524833679,
+      "learning_rate": 9.468031196608068e-05,
+      "loss": 0.954498291015625,
+      "step": 65
+    },
+    {
+      "epoch": 0.9445438282647585,
+      "grad_norm": 0.056810859590768814,
+      "learning_rate": 9.448191989157294e-05,
+      "loss": 0.88922119140625,
+      "step": 66
+    },
+    {
+      "epoch": 0.9588550983899821,
+      "grad_norm": 0.05907364562153816,
+      "learning_rate": 9.428013706914678e-05,
+      "loss": 0.931243896484375,
+      "step": 67
+    },
+    {
+      "epoch": 0.9731663685152058,
+      "grad_norm": 0.06500693410634995,
+      "learning_rate": 9.407498082811863e-05,
+      "loss": 0.9434814453125,
+      "step": 68
+    },
+    {
+      "epoch": 0.9874776386404294,
+      "grad_norm": 0.054131485521793365,
+      "learning_rate": 9.386646878751769e-05,
+      "loss": 0.903411865234375,
+      "step": 69
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.06751064956188202,
+      "learning_rate": 9.365461885457259e-05,
+      "loss": 1.0047607421875,
+      "step": 70
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 0.91888427734375,
+      "eval_runtime": 1.8844,
+      "eval_samples_per_second": 2.123,
+      "eval_steps_per_second": 0.531,
+      "step": 70
+    },
+    {
+      "epoch": 1.0143112701252237,
+      "grad_norm": 0.055898845195770264,
+      "learning_rate": 9.343944922317363e-05,
+      "loss": 0.96551513671875,
+      "step": 71
+    },
+    {
+      "epoch": 1.0286225402504472,
+      "grad_norm": 0.052690621465444565,
+      "learning_rate": 9.322097837231026e-05,
+      "loss": 0.7611083984375,
+      "step": 72
+    },
+    {
+      "epoch": 1.042933810375671,
+      "grad_norm": 0.07362278550863266,
+      "learning_rate": 9.299922506448396e-05,
+      "loss": 0.97052001953125,
+      "step": 73
+    },
+    {
+      "epoch": 1.0572450805008944,
+      "grad_norm": 0.06880676746368408,
+      "learning_rate": 9.27742083440971e-05,
+      "loss": 0.90020751953125,
+      "step": 74
+    },
+    {
+      "epoch": 1.071556350626118,
+      "grad_norm": 0.060559019446372986,
+      "learning_rate": 9.254594753581724e-05,
+      "loss": 0.946502685546875,
+      "step": 75
+    },
+    {
+      "epoch": 1.0858676207513416,
+      "grad_norm": 0.04849094897508621,
+      "learning_rate": 9.23144622429175e-05,
+      "loss": 0.82427978515625,
+      "step": 76
+    },
+    {
+      "epoch": 1.1001788908765653,
+      "grad_norm": 0.06903208792209625,
+      "learning_rate": 9.20797723455931e-05,
+      "loss": 0.86859130859375,
+      "step": 77
+    },
+    {
+      "epoch": 1.114490161001789,
+      "grad_norm": 0.07261502742767334,
+      "learning_rate": 9.184189799925404e-05,
+      "loss": 0.86370849609375,
+      "step": 78
+    },
+    {
+      "epoch": 1.1288014311270125,
+      "grad_norm": 0.0514570027589798,
+      "learning_rate": 9.160085963279396e-05,
+      "loss": 0.824951171875,
+      "step": 79
+    },
+    {
+      "epoch": 1.1431127012522362,
+      "grad_norm": 0.07142036408185959,
+      "learning_rate": 9.135667794683593e-05,
+      "loss": 0.918609619140625,
+      "step": 80
+    },
+    {
+      "epoch": 1.1574239713774597,
+      "grad_norm": 0.051894742995500565,
+      "learning_rate": 9.110937391195442e-05,
+      "loss": 0.9404296875,
+      "step": 81
+    },
+    {
+      "epoch": 1.1717352415026834,
+      "grad_norm": 0.05748848617076874,
+      "learning_rate": 9.08589687668745e-05,
+      "loss": 0.82635498046875,
+      "step": 82
+    },
+    {
+      "epoch": 1.1860465116279069,
+      "grad_norm": 0.05919559299945831,
+      "learning_rate": 9.06054840166478e-05,
+      "loss": 0.937835693359375,
+      "step": 83
+    },
+    {
+      "epoch": 1.2003577817531306,
+      "grad_norm": 0.05367245525121689,
+      "learning_rate": 9.034894143080553e-05,
+      "loss": 0.90118408203125,
+      "step": 84
+    },
+    {
+      "epoch": 1.2146690518783543,
+      "grad_norm": 0.04632791131734848,
+      "learning_rate": 9.008936304148907e-05,
+      "loss": 0.933746337890625,
+      "step": 85
+    },
+    {
+      "epoch": 1.2289803220035778,
+      "grad_norm": 0.0533670149743557,
+      "learning_rate": 8.982677114155765e-05,
+      "loss": 0.8226318359375,
+      "step": 86
+    },
+    {
+      "epoch": 1.2432915921288015,
+      "grad_norm": 0.05788963660597801,
+      "learning_rate": 8.95611882826739e-05,
+      "loss": 0.864898681640625,
+      "step": 87
+    },
+    {
+      "epoch": 1.257602862254025,
+      "grad_norm": 0.051566317677497864,
+      "learning_rate": 8.929263727336713e-05,
+      "loss": 0.971893310546875,
+      "step": 88
+    },
+    {
+      "epoch": 1.2719141323792487,
+      "grad_norm": 0.05482623726129532,
+      "learning_rate": 8.902114117707441e-05,
+      "loss": 1.00732421875,
+      "step": 89
+    },
+    {
+      "epoch": 1.2862254025044724,
+      "grad_norm": 0.04872555285692215,
+      "learning_rate": 8.874672331015996e-05,
+      "loss": 0.9832763671875,
+      "step": 90
+    },
+    {
+      "epoch": 1.3005366726296959,
+      "grad_norm": 0.058804403990507126,
+      "learning_rate": 8.846940723991263e-05,
+      "loss": 0.866729736328125,
+      "step": 91
+    },
+    {
+      "epoch": 1.3148479427549196,
+      "grad_norm": 0.04719854146242142,
+      "learning_rate": 8.818921678252199e-05,
+      "loss": 0.811676025390625,
+      "step": 92
+    },
+    {
+      "epoch": 1.329159212880143,
+      "grad_norm": 0.04760557413101196,
+      "learning_rate": 8.790617600103289e-05,
+      "loss": 0.87615966796875,
+      "step": 93
+    },
+    {
+      "epoch": 1.3434704830053668,
+      "grad_norm": 0.03823496401309967,
+      "learning_rate": 8.762030920327898e-05,
+      "loss": 0.804779052734375,
+      "step": 94
+    },
+    {
+      "epoch": 1.3577817531305905,
+      "grad_norm": 0.049484819173812866,
+      "learning_rate": 8.733164093979512e-05,
+      "loss": 0.8387451171875,
+      "step": 95
+    },
+    {
+      "epoch": 1.372093023255814,
+      "grad_norm": 0.04773775488138199,
+      "learning_rate": 8.704019600170883e-05,
+      "loss": 0.894744873046875,
+      "step": 96
+    },
+    {
+      "epoch": 1.3864042933810374,
+      "grad_norm": 0.06226862221956253,
+      "learning_rate": 8.674599941861145e-05,
+      "loss": 0.96490478515625,
+      "step": 97
+    },
+    {
+      "epoch": 1.4007155635062611,
+      "grad_norm": 0.05097726732492447,
+      "learning_rate": 8.64490764564083e-05,
+      "loss": 0.8448486328125,
+      "step": 98
+    },
+    {
+      "epoch": 1.4150268336314848,
+      "grad_norm": 0.05041167885065079,
+      "learning_rate": 8.614945261514903e-05,
+      "loss": 0.78009033203125,
+      "step": 99
+    },
+    {
+      "epoch": 1.4293381037567083,
+      "grad_norm": 0.038732826709747314,
+      "learning_rate": 8.584715362683751e-05,
+      "loss": 0.749969482421875,
+      "step": 100
+    },
+    {
+      "epoch": 1.443649373881932,
+      "grad_norm": 0.04899672418832779,
+      "learning_rate": 8.554220545322197e-05,
+      "loss": 0.861297607421875,
+      "step": 101
+    },
+    {
+      "epoch": 1.4579606440071555,
+      "grad_norm": 0.05308113619685173,
+      "learning_rate": 8.523463428356548e-05,
+      "loss": 0.96209716796875,
+      "step": 102
+    },
+    {
+      "epoch": 1.4722719141323792,
+      "grad_norm": 0.06102287024259567,
+      "learning_rate": 8.492446653239667e-05,
+      "loss": 0.857452392578125,
+      "step": 103
+    },
+    {
+      "epoch": 1.486583184257603,
+      "grad_norm": 0.05445420369505882,
+      "learning_rate": 8.461172883724124e-05,
+      "loss": 0.98309326171875,
+      "step": 104
+    },
+    {
+      "epoch": 1.5008944543828264,
+      "grad_norm": 0.06945395469665527,
+      "learning_rate": 8.429644805633437e-05,
+      "loss": 0.771636962890625,
+      "step": 105
+    },
+    {
+      "epoch": 1.5152057245080501,
+      "grad_norm": 0.07941682636737823,
+      "learning_rate": 8.397865126631402e-05,
+      "loss": 0.82354736328125,
+      "step": 106
+    },
+    {
+      "epoch": 1.5295169946332736,
+      "grad_norm": 0.054743777960538864,
+      "learning_rate": 8.365836575989561e-05,
+      "loss": 0.97491455078125,
+      "step": 107
+    },
+    {
+      "epoch": 1.5438282647584973,
+      "grad_norm": 0.08232340216636658,
+      "learning_rate": 8.333561904352814e-05,
+      "loss": 0.9542999267578125,
+      "step": 108
+    },
+    {
+      "epoch": 1.558139534883721,
+      "grad_norm": 0.05673123523592949,
+      "learning_rate": 8.301043883503177e-05,
+      "loss": 0.77447509765625,
+      "step": 109
+    },
+    {
+      "epoch": 1.5724508050089445,
+      "grad_norm": 0.07514223456382751,
+      "learning_rate": 8.26828530612175e-05,
+      "loss": 0.90985107421875,
+      "step": 110
+    },
+    {
+      "epoch": 1.586762075134168,
+      "grad_norm": 0.07404479384422302,
+      "learning_rate": 8.235288985548881e-05,
+      "loss": 0.84405517578125,
+      "step": 111
+    },
+    {
+      "epoch": 1.6010733452593917,
+      "grad_norm": 0.04848802462220192,
+      "learning_rate": 8.20205775554254e-05,
+      "loss": 0.90277099609375,
+      "step": 112
+    },
+    {
+      "epoch": 1.6153846153846154,
+      "grad_norm": 0.06696207821369171,
+      "learning_rate": 8.168594470034966e-05,
+      "loss": 0.8168106079101562,
+      "step": 113
+    },
+    {
+      "epoch": 1.6296958855098391,
+      "grad_norm": 0.05016935244202614,
+      "learning_rate": 8.13490200288756e-05,
+      "loss": 0.9027099609375,
+      "step": 114
+    },
+    {
+      "epoch": 1.6440071556350626,
+      "grad_norm": 0.08251702785491943,
+      "learning_rate": 8.100983247644082e-05,
+      "loss": 0.8768310546875,
+      "step": 115
+    },
+    {
+      "epoch": 1.658318425760286,
+      "grad_norm": 0.0642937570810318,
+      "learning_rate": 8.06684111728214e-05,
+      "loss": 0.9188232421875,
+      "step": 116
+    },
+    {
+      "epoch": 1.6726296958855098,
+      "grad_norm": 0.043508097529411316,
+      "learning_rate": 8.032478543963038e-05,
+      "loss": 0.73736572265625,
+      "step": 117
+    },
+    {
+      "epoch": 1.6869409660107335,
+      "grad_norm": 0.05819007009267807,
+      "learning_rate": 7.997898478779936e-05,
+      "loss": 0.82122802734375,
+      "step": 118
+    },
+    {
+      "epoch": 1.7012522361359572,
+      "grad_norm": 0.05959341302514076,
+      "learning_rate": 7.96310389150443e-05,
+      "loss": 0.860137939453125,
+      "step": 119
+    },
+    {
+      "epoch": 1.7155635062611807,
+      "grad_norm": 0.07089157402515411,
+      "learning_rate": 7.92809777033148e-05,
+      "loss": 0.859893798828125,
+      "step": 120
+    },
+    {
+      "epoch": 1.7298747763864042,
+      "grad_norm": 0.06324061006307602,
+      "learning_rate": 7.892883121622813e-05,
+      "loss": 0.84295654296875,
+      "step": 121
+    },
+    {
+      "epoch": 1.744186046511628,
+      "grad_norm": 0.046000510454177856,
+      "learning_rate": 7.857462969648702e-05,
+      "loss": 0.693756103515625,
+      "step": 122
+    },
+    {
+      "epoch": 1.7584973166368516,
+      "grad_norm": 0.07442042231559753,
+      "learning_rate": 7.821840356328254e-05,
+      "loss": 0.9080810546875,
+      "step": 123
+    },
+    {
+      "epoch": 1.772808586762075,
+      "grad_norm": 0.0669429674744606,
+      "learning_rate": 7.78601834096817e-05,
+      "loss": 0.85760498046875,
+      "step": 124
+    },
+    {
+      "epoch": 1.7871198568872988,
+      "grad_norm": 0.05742024630308151,
+      "learning_rate": 7.75e-05,
+      "loss": 0.7733154296875,
+      "step": 125
+    },
+    {
+      "epoch": 1.8014311270125223,
+      "grad_norm": 0.0744951069355011,
+      "learning_rate": 7.713788426715938e-05,
+      "loss": 0.857818603515625,
+      "step": 126
+    },
+    {
+      "epoch": 1.815742397137746,
+      "grad_norm": 0.05928736925125122,
+      "learning_rate": 7.677386731003171e-05,
+      "loss": 0.7152498364448547,
+      "step": 127
+    },
+    {
+      "epoch": 1.8300536672629697,
+      "grad_norm": 0.052303068339824677,
+      "learning_rate": 7.640798039076793e-05,
+      "loss": 0.7852783203125,
+      "step": 128
+    },
+    {
+      "epoch": 1.8443649373881932,
+      "grad_norm": 0.05261474475264549,
+      "learning_rate": 7.604025493211327e-05,
+      "loss": 0.84442138671875,
+      "step": 129
+    },
+    {
+      "epoch": 1.8586762075134167,
+      "grad_norm": 0.06918948888778687,
+      "learning_rate": 7.567072251470862e-05,
+      "loss": 0.826507568359375,
+      "step": 130
+    },
+    {
+      "epoch": 1.8729874776386404,
+      "grad_norm": 0.053066812455654144,
+      "learning_rate": 7.529941487437835e-05,
+      "loss": 0.9246826171875,
+      "step": 131
+    },
+    {
+      "epoch": 1.887298747763864,
+      "grad_norm": 0.0635317787528038,
+      "learning_rate": 7.492636389940478e-05,
+      "loss": 0.8355712890625,
+      "step": 132
+    },
+    {
+      "epoch": 1.9016100178890878,
+      "grad_norm": 0.061956532299518585,
+      "learning_rate": 7.455160162778962e-05,
+      "loss": 0.8302001953125,
+      "step": 133
+    },
+    {
+      "epoch": 1.9159212880143113,
+      "grad_norm": 0.06120719015598297,
+      "learning_rate": 7.41751602445025e-05,
+      "loss": 0.9422607421875,
+      "step": 134
+    },
+    {
+      "epoch": 1.9302325581395348,
+      "grad_norm": 0.05771296098828316,
+      "learning_rate": 7.379707207871685e-05,
+      "loss": 0.874969482421875,
+      "step": 135
+    },
+    {
+      "epoch": 1.9445438282647585,
+      "grad_norm": 0.0516636036336422,
+      "learning_rate": 7.341736960103357e-05,
+      "loss": 0.820343017578125,
+      "step": 136
+    },
+    {
+      "epoch": 1.9588550983899822,
+      "grad_norm": 0.05117014795541763,
+      "learning_rate": 7.30360854206922e-05,
+      "loss": 0.859710693359375,
+      "step": 137
+    },
+    {
+      "epoch": 1.9731663685152059,
+      "grad_norm": 0.06877405941486359,
+      "learning_rate": 7.265325228277064e-05,
+      "loss": 0.86651611328125,
+      "step": 138
+    },
+    {
+      "epoch": 1.9874776386404294,
+      "grad_norm": 0.057914137840270996,
+      "learning_rate": 7.22689030653728e-05,
+      "loss": 0.836456298828125,
+      "step": 139
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.0669974833726883,
+      "learning_rate": 7.18830707768051e-05,
+      "loss": 0.92803955078125,
+      "step": 140
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.85125732421875,
+      "eval_runtime": 1.8814,
+      "eval_samples_per_second": 2.126,
+      "eval_steps_per_second": 0.532,
+      "step": 140
+    },
+    {
+      "epoch": 2.0143112701252237,
+      "grad_norm": 0.06470653414726257,
+      "learning_rate": 7.149578855274158e-05,
+      "loss": 0.8944091796875,
+      "step": 141
+    },
+    {
+      "epoch": 2.0286225402504474,
+      "grad_norm": 0.05126171186566353,
+      "learning_rate": 7.110708965337829e-05,
+      "loss": 0.7092361450195312,
+      "step": 142
+    },
+    {
+      "epoch": 2.0429338103756707,
+      "grad_norm": 0.08162374794483185,
+      "learning_rate": 7.071700746057677e-05,
+      "loss": 0.89935302734375,
+      "step": 143
+    },
+    {
+      "epoch": 2.0572450805008944,
+      "grad_norm": 0.07985501736402512,
+      "learning_rate": 7.032557547499726e-05,
+      "loss": 0.83685302734375,
+      "step": 144
+    },
+    {
+      "epoch": 2.071556350626118,
+      "grad_norm": 0.062123995274305344,
+      "learning_rate": 6.993282731322157e-05,
+      "loss": 0.881927490234375,
+      "step": 145
+    },
+    {
+      "epoch": 2.085867620751342,
+      "grad_norm": 0.06731198728084564,
+      "learning_rate": 6.953879670486609e-05,
+      "loss": 0.773712158203125,
+      "step": 146
+    },
+    {
+      "epoch": 2.1001788908765655,
+      "grad_norm": 0.061235297471284866,
+      "learning_rate": 6.914351748968501e-05,
+      "loss": 0.808929443359375,
+      "step": 147
+    },
+    {
+      "epoch": 2.1144901610017888,
+      "grad_norm": 0.07446978241205215,
+      "learning_rate": 6.874702361466423e-05,
+      "loss": 0.800750732421875,
+      "step": 148
+    },
+    {
+      "epoch": 2.1288014311270125,
+      "grad_norm": 0.07796134799718857,
+      "learning_rate": 6.83493491311058e-05,
+      "loss": 0.772918701171875,
+      "step": 149
+    },
+    {
+      "epoch": 2.143112701252236,
+      "grad_norm": 0.08171713352203369,
+      "learning_rate": 6.79505281917037e-05,
+      "loss": 0.86181640625,
+      "step": 150
+    },
+    {
+      "epoch": 2.15742397137746,
+      "grad_norm": 0.062124211341142654,
+      "learning_rate": 6.755059504761073e-05,
+      "loss": 0.881988525390625,
+      "step": 151
+    },
+    {
+      "epoch": 2.171735241502683,
+      "grad_norm": 0.05511016771197319,
+      "learning_rate": 6.714958404549693e-05,
+      "loss": 0.77618408203125,
+      "step": 152
+    },
+    {
+      "epoch": 2.186046511627907,
+      "grad_norm": 0.08269248157739639,
+      "learning_rate": 6.67475296245999e-05,
+      "loss": 0.88116455078125,
+      "step": 153
+    },
+    {
+      "epoch": 2.2003577817531306,
+      "grad_norm": 0.07227372378110886,
+      "learning_rate": 6.634446631376716e-05,
+      "loss": 0.849212646484375,
+      "step": 154
+    },
+    {
+      "epoch": 2.2146690518783543,
+      "grad_norm": 0.06654597073793411,
+      "learning_rate": 6.59404287284907e-05,
+      "loss": 0.88165283203125,
+      "step": 155
+    },
+    {
+      "epoch": 2.228980322003578,
+      "grad_norm": 0.06412966549396515,
+      "learning_rate": 6.553545156793422e-05,
+      "loss": 0.778289794921875,
+      "step": 156
+    },
+    {
+      "epoch": 2.2432915921288012,
+      "grad_norm": 0.06827051192522049,
+      "learning_rate": 6.512956961195312e-05,
+      "loss": 0.81201171875,
+      "step": 157
+    },
+    {
+      "epoch": 2.257602862254025,
+      "grad_norm": 0.0686182752251625,
+      "learning_rate": 6.472281771810751e-05,
+      "loss": 0.920074462890625,
+      "step": 158
+    },
+    {
+      "epoch": 2.2719141323792487,
+      "grad_norm": 0.07015014439821243,
+      "learning_rate": 6.431523081866871e-05,
+      "loss": 0.9527587890625,
+      "step": 159
+    },
+    {
+      "epoch": 2.2862254025044724,
+      "grad_norm": 0.058846067637205124,
+      "learning_rate": 6.39068439176192e-05,
+      "loss": 0.93096923828125,
+      "step": 160
+    },
+    {
+      "epoch": 2.300536672629696,
+      "grad_norm": 0.07005320489406586,
+      "learning_rate": 6.349769208764634e-05,
+      "loss": 0.821746826171875,
+      "step": 161
+    },
+    {
+      "epoch": 2.3148479427549193,
+      "grad_norm": 0.05326821282505989,
+      "learning_rate": 6.308781046713046e-05,
+      "loss": 0.769561767578125,
+      "step": 162
+    },
+    {
+      "epoch": 2.329159212880143,
+      "grad_norm": 0.057699061930179596,
+      "learning_rate": 6.2677234257127e-05,
+      "loss": 0.833038330078125,
+      "step": 163
+    },
+    {
+      "epoch": 2.3434704830053668,
+      "grad_norm": 0.04352062568068504,
+      "learning_rate": 6.226599871834347e-05,
+      "loss": 0.764801025390625,
+      "step": 164
+    },
+    {
+      "epoch": 2.3577817531305905,
+      "grad_norm": 0.05532877892255783,
+      "learning_rate": 6.185413916811122e-05,
+      "loss": 0.79705810546875,
+      "step": 165
+    },
+    {
+      "epoch": 2.3720930232558137,
+      "grad_norm": 0.048729632049798965,
+      "learning_rate": 6.144169097735237e-05,
+      "loss": 0.8542022705078125,
+      "step": 166
+    },
+    {
+      "epoch": 2.3864042933810374,
+      "grad_norm": 0.06301187723875046,
+      "learning_rate": 6.102868956754203e-05,
+      "loss": 0.915985107421875,
+      "step": 167
+    },
+    {
+      "epoch": 2.400715563506261,
+      "grad_norm": 0.06361453235149384,
+      "learning_rate": 6.06151704076664e-05,
+      "loss": 0.806427001953125,
+      "step": 168
+    },
+    {
+      "epoch": 2.415026833631485,
+      "grad_norm": 0.050374533981084824,
+      "learning_rate": 6.020116901117655e-05,
+      "loss": 0.7458038330078125,
+      "step": 169
+    },
+    {
+      "epoch": 2.4293381037567086,
+      "grad_norm": 0.05397063121199608,
+      "learning_rate": 5.978672093293851e-05,
+      "loss": 0.717529296875,
+      "step": 170
+    },
+    {
+      "epoch": 2.4436493738819323,
+      "grad_norm": 0.053685761988162994,
+      "learning_rate": 5.937186176617987e-05,
+      "loss": 0.822509765625,
+      "step": 171
+    },
+    {
+      "epoch": 2.4579606440071555,
+      "grad_norm": 0.07064840197563171,
+      "learning_rate": 5.895662713943285e-05,
+      "loss": 0.916168212890625,
+      "step": 172
+    },
+    {
+      "epoch": 2.4722719141323792,
+      "grad_norm": 0.0546434223651886,
+      "learning_rate": 5.854105271347462e-05,
+      "loss": 0.820068359375,
+      "step": 173
+    },
+    {
+      "epoch": 2.486583184257603,
+      "grad_norm": 0.09119619429111481,
+      "learning_rate": 5.812517417826456e-05,
+      "loss": 0.93585205078125,
+      "step": 174
+    },
+    {
+      "epoch": 2.500894454382826,
+      "grad_norm": 0.08696950227022171,
+      "learning_rate": 5.770902724987942e-05,
+      "loss": 0.737091064453125,
+      "step": 175
+    },
+    {
+      "epoch": 2.51520572450805,
+      "grad_norm": 0.06517396867275238,
+      "learning_rate": 5.7292647667445675e-05,
+      "loss": 0.787078857421875,
+      "step": 176
+    },
+    {
+      "epoch": 2.5295169946332736,
+      "grad_norm": 0.09514431655406952,
+      "learning_rate": 5.687607119007047e-05,
+      "loss": 0.9356689453125,
+      "step": 177
+    },
+    {
+      "epoch": 2.5438282647584973,
+      "grad_norm": 0.05851126089692116,
+      "learning_rate": 5.6459333593770435e-05,
+      "loss": 0.914642333984375,
+      "step": 178
+    },
+    {
+      "epoch": 2.558139534883721,
+      "grad_norm": 0.05984664335846901,
+      "learning_rate": 5.604247066839932e-05,
+      "loss": 0.74224853515625,
+      "step": 179
+    },
+    {
+      "epoch": 2.5724508050089447,
+      "grad_norm": 0.0807105079293251,
+      "learning_rate": 5.562551821457418e-05,
+      "loss": 0.8707275390625,
+      "step": 180
+    },
+    {
+      "epoch": 2.586762075134168,
+      "grad_norm": 0.06917821615934372,
+      "learning_rate": 5.520851204060096e-05,
+      "loss": 0.810638427734375,
+      "step": 181
+    },
+    {
+      "epoch": 2.6010733452593917,
+      "grad_norm": 0.053058020770549774,
+      "learning_rate": 5.4791487959399054e-05,
+      "loss": 0.868927001953125,
+      "step": 182
+    },
+    {
+      "epoch": 2.6153846153846154,
+      "grad_norm": 0.08303209394216537,
+      "learning_rate": 5.437448178542582e-05,
+      "loss": 0.7835197448730469,
+      "step": 183
+    },
+    {
+      "epoch": 2.629695885509839,
+      "grad_norm": 0.06776992231607437,
+      "learning_rate": 5.3957529331600686e-05,
+      "loss": 0.86773681640625,
+      "step": 184
+    },
+    {
+      "epoch": 2.6440071556350624,
+      "grad_norm": 0.08375553786754608,
+      "learning_rate": 5.3540666406229576e-05,
+      "loss": 0.843017578125,
+      "step": 185
+    },
+    {
+      "epoch": 2.658318425760286,
+      "grad_norm": 0.06997603923082352,
+      "learning_rate": 5.3123928809929534e-05,
+      "loss": 0.88525390625,
+      "step": 186
+    },
+    {
+      "epoch": 2.67262969588551,
+      "grad_norm": 0.05620221421122551,
+      "learning_rate": 5.270735233255435e-05,
+      "loss": 0.7114105224609375,
+      "step": 187
+    },
+    {
+      "epoch": 2.6869409660107335,
+      "grad_norm": 0.05393487587571144,
+      "learning_rate": 5.22909727501206e-05,
+      "loss": 0.7886962890625,
+      "step": 188
+    },
+    {
+      "epoch": 2.701252236135957,
+      "grad_norm": 0.0686308965086937,
+      "learning_rate": 5.187482582173544e-05,
+      "loss": 0.82928466796875,
+      "step": 189
+    },
+    {
+      "epoch": 2.715563506261181,
+      "grad_norm": 0.05664825439453125,
+      "learning_rate": 5.145894728652541e-05,
+      "loss": 0.826416015625,
+      "step": 190
+    },
+    {
+      "epoch": 2.729874776386404,
+      "grad_norm": 0.060908980667591095,
+      "learning_rate": 5.104337286056716e-05,
+      "loss": 0.8122100830078125,
+      "step": 191
+    },
+    {
+      "epoch": 2.744186046511628,
+      "grad_norm": 0.04383637011051178,
+      "learning_rate": 5.0628138233820146e-05,
+      "loss": 0.670745849609375,
+      "step": 192
+    },
+    {
+      "epoch": 2.7584973166368516,
+      "grad_norm": 0.07157369703054428,
+      "learning_rate": 5.02132790670615e-05,
+      "loss": 0.875244140625,
+      "step": 193
+    },
+    {
+      "epoch": 2.772808586762075,
+      "grad_norm": 0.05150938406586647,
+      "learning_rate": 4.9798830988823475e-05,
+      "loss": 0.8294677734375,
+      "step": 194
+    },
+    {
+      "epoch": 2.7871198568872986,
+      "grad_norm": 0.05933945253491402,
+      "learning_rate": 4.9384829592333624e-05,
+      "loss": 0.746856689453125,
+      "step": 195
+    },
+    {
+      "epoch": 2.8014311270125223,
+      "grad_norm": 0.048639580607414246,
+      "learning_rate": 4.8971310432457974e-05,
+      "loss": 0.830596923828125,
+      "step": 196
+    },
+    {
+      "epoch": 2.815742397137746,
+      "grad_norm": 0.0681508481502533,
+      "learning_rate": 4.855830902264764e-05,
+      "loss": 0.6901645064353943,
+      "step": 197
+    },
+    {
+      "epoch": 2.8300536672629697,
+      "grad_norm": 0.048687584698200226,
+      "learning_rate": 4.814586083188878e-05,
+      "loss": 0.75885009765625,
+      "step": 198
+    },
+    {
+      "epoch": 2.8443649373881934,
+      "grad_norm": 0.06612910330295563,
+      "learning_rate": 4.773400128165654e-05,
+      "loss": 0.816986083984375,
+      "step": 199
+    },
+    {
+      "epoch": 2.8586762075134167,
+      "grad_norm": 0.06315340101718903,
+      "learning_rate": 4.7322765742873e-05,
+      "loss": 0.79925537109375,
+      "step": 200
+    },
+    {
+      "epoch": 2.8729874776386404,
+      "grad_norm": 0.07031147181987762,
+      "learning_rate": 4.691218953286954e-05,
+      "loss": 0.89544677734375,
+      "step": 201
+    },
+    {
+      "epoch": 2.887298747763864,
+      "grad_norm": 0.06599035114049911,
+      "learning_rate": 4.650230791235366e-05,
+      "loss": 0.809326171875,
+      "step": 202
+    },
+    {
+      "epoch": 2.901610017889088,
+      "grad_norm": 0.05649695545434952,
+      "learning_rate": 4.609315608238082e-05,
+      "loss": 0.8052978515625,
+      "step": 203
+    },
+    {
+      "epoch": 2.915921288014311,
+      "grad_norm": 0.06291844695806503,
+      "learning_rate": 4.568476918133129e-05,
+      "loss": 0.91290283203125,
+      "step": 204
+    },
+    {
+      "epoch": 2.9302325581395348,
+      "grad_norm": 0.050558559596538544,
+      "learning_rate": 4.5277182281892495e-05,
+      "loss": 0.847320556640625,
+      "step": 205
+    },
+    {
+      "epoch": 2.9445438282647585,
+      "grad_norm": 0.0605635903775692,
+      "learning_rate": 4.48704303880469e-05,
+      "loss": 0.79638671875,
+      "step": 206
+    },
+    {
+      "epoch": 2.958855098389982,
+      "grad_norm": 0.04639475420117378,
+      "learning_rate": 4.446454843206578e-05,
+      "loss": 0.83404541015625,
+      "step": 207
+    },
+    {
+      "epoch": 2.973166368515206,
+      "grad_norm": 0.06873723119497299,
+      "learning_rate": 4.405957127150929e-05,
+      "loss": 0.83966064453125,
+      "step": 208
+    },
+    {
+      "epoch": 2.9874776386404296,
+      "grad_norm": 0.052222803235054016,
+      "learning_rate": 4.3655533686232856e-05,
+      "loss": 0.814117431640625,
+      "step": 209
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.07146501541137695,
+      "learning_rate": 4.3252470375400114e-05,
+      "loss": 0.900634765625,
+      "step": 210
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.82879638671875,
+      "eval_runtime": 1.8766,
+      "eval_samples_per_second": 2.131,
+      "eval_steps_per_second": 0.533,
+      "step": 210
+    },
+    {
+      "epoch": 3.0143112701252237,
+      "grad_norm": 0.054631415754556656,
+      "learning_rate": 4.28504159545031e-05,
+      "loss": 0.868499755859375,
+      "step": 211
+    },
+    {
+      "epoch": 3.0286225402504474,
+      "grad_norm": 0.05336322635412216,
+      "learning_rate": 4.244940495238928e-05,
+      "loss": 0.6902542114257812,
+      "step": 212
+    },
+    {
+      "epoch": 3.0429338103756707,
+      "grad_norm": 0.07756180316209793,
+      "learning_rate": 4.204947180829629e-05,
+      "loss": 0.87384033203125,
+      "step": 213
+    },
+    {
+      "epoch": 3.0572450805008944,
+      "grad_norm": 0.06461610645055771,
+      "learning_rate": 4.165065086889421e-05,
+      "loss": 0.81378173828125,
+      "step": 214
+    },
+    {
+      "epoch": 3.071556350626118,
+      "grad_norm": 0.06902662664651871,
+      "learning_rate": 4.125297638533577e-05,
+      "loss": 0.857696533203125,
+      "step": 215
+    },
+    {
+      "epoch": 3.085867620751342,
+      "grad_norm": 0.05593303591012955,
+      "learning_rate": 4.0856482510314984e-05,
+      "loss": 0.754486083984375,
+      "step": 216
+    },
+    {
+      "epoch": 3.1001788908765655,
+      "grad_norm": 0.07445930689573288,
+      "learning_rate": 4.046120329513393e-05,
+      "loss": 0.786376953125,
+      "step": 217
+    },
+    {
+      "epoch": 3.1144901610017888,
+      "grad_norm": 0.06036040931940079,
+      "learning_rate": 4.0067172686778445e-05,
+      "loss": 0.776580810546875,
+      "step": 218
+    },
+    {
+      "epoch": 3.1288014311270125,
+      "grad_norm": 0.05283353477716446,
+      "learning_rate": 3.9674424525002764e-05,
+      "loss": 0.752655029296875,
+      "step": 219
+    },
+    {
+      "epoch": 3.143112701252236,
+      "grad_norm": 0.05828075110912323,
+      "learning_rate": 3.9282992539423244e-05,
+      "loss": 0.838592529296875,
+      "step": 220
+    },
+    {
+      "epoch": 3.15742397137746,
+      "grad_norm": 0.06032145768404007,
+      "learning_rate": 3.8892910346621714e-05,
+      "loss": 0.859466552734375,
+      "step": 221
+    },
+    {
+      "epoch": 3.171735241502683,
+      "grad_norm": 0.049184400588274,
+      "learning_rate": 3.850421144725844e-05,
+      "loss": 0.7567138671875,
+      "step": 222
+    },
+    {
+      "epoch": 3.186046511627907,
+      "grad_norm": 0.057277943938970566,
+      "learning_rate": 3.811692922319491e-05,
+      "loss": 0.858642578125,
+      "step": 223
+    },
+    {
+      "epoch": 3.2003577817531306,
+      "grad_norm": 0.05168966203927994,
+      "learning_rate": 3.77310969346272e-05,
+      "loss": 0.828826904296875,
+      "step": 224
+    },
+    {
+      "epoch": 3.2146690518783543,
+      "grad_norm": 0.053891345858573914,
+      "learning_rate": 3.734674771722937e-05,
+      "loss": 0.861053466796875,
+      "step": 225
+    },
+    {
+      "epoch": 3.228980322003578,
+      "grad_norm": 0.04935247823596001,
+      "learning_rate": 3.69639145793078e-05,
+      "loss": 0.7604522705078125,
+      "step": 226
+    },
+    {
+      "epoch": 3.2432915921288012,
+      "grad_norm": 0.05808898061513901,
+      "learning_rate": 3.658263039896646e-05,
+      "loss": 0.791290283203125,
+      "step": 227
+    },
+    {
+      "epoch": 3.257602862254025,
+      "grad_norm": 0.05283699184656143,
+      "learning_rate": 3.620292792128315e-05,
+      "loss": 0.900146484375,
+      "step": 228
+    },
+    {
+      "epoch": 3.2719141323792487,
+      "grad_norm": 0.058179955929517746,
+      "learning_rate": 3.582483975549752e-05,
+      "loss": 0.930419921875,
+      "step": 229
+    },
+    {
+      "epoch": 3.2862254025044724,
+      "grad_norm": 0.05453471094369888,
+      "learning_rate": 3.54483983722104e-05,
+      "loss": 0.909332275390625,
+      "step": 230
+    },
+    {
+      "epoch": 3.300536672629696,
+      "grad_norm": 0.050214268267154694,
+      "learning_rate": 3.507363610059522e-05,
+      "loss": 0.803009033203125,
+      "step": 231
+    },
+    {
+      "epoch": 3.3148479427549193,
+      "grad_norm": 0.052127815783023834,
+      "learning_rate": 3.4700585125621664e-05,
+      "loss": 0.75262451171875,
+      "step": 232
+    },
+    {
+      "epoch": 3.329159212880143,
+      "grad_norm": 0.044876426458358765,
+      "learning_rate": 3.4329277485291393e-05,
+      "loss": 0.815032958984375,
+      "step": 233
+    },
+    {
+      "epoch": 3.3434704830053668,
+      "grad_norm": 0.04321800917387009,
+      "learning_rate": 3.395974506788674e-05,
+      "loss": 0.74774169921875,
+      "step": 234
+    },
+    {
+      "epoch": 3.3577817531305905,
+      "grad_norm": 0.044144321233034134,
+      "learning_rate": 3.359201960923209e-05,
+      "loss": 0.780364990234375,
+      "step": 235
+    },
+    {
+      "epoch": 3.3720930232558137,
+      "grad_norm": 0.047196824103593826,
+      "learning_rate": 3.322613268996829e-05,
+      "loss": 0.837310791015625,
+      "step": 236
+    },
+    {
+      "epoch": 3.3864042933810374,
+      "grad_norm": 0.06098597124218941,
+      "learning_rate": 3.286211573284063e-05,
+      "loss": 0.89617919921875,
+      "step": 237
+    },
+    {
+      "epoch": 3.400715563506261,
+      "grad_norm": 0.04971824958920479,
+      "learning_rate": 3.250000000000001e-05,
+      "loss": 0.79010009765625,
+      "step": 238
+    },
+    {
+      "epoch": 3.415026833631485,
+      "grad_norm": 0.047822561115026474,
+      "learning_rate": 3.213981659031829e-05,
+      "loss": 0.73162841796875,
+      "step": 239
+    },
+    {
+      "epoch": 3.4293381037567086,
+      "grad_norm": 0.042717307806015015,
+      "learning_rate": 3.1781596436717465e-05,
+      "loss": 0.7030410766601562,
+      "step": 240
+    },
+    {
+      "epoch": 3.4436493738819323,
+      "grad_norm": 0.04302101954817772,
+      "learning_rate": 3.1425370303513e-05,
+      "loss": 0.805511474609375,
+      "step": 241
+    },
+    {
+      "epoch": 3.4579606440071555,
+      "grad_norm": 0.05659914016723633,
+      "learning_rate": 3.107116878377188e-05,
+      "loss": 0.896728515625,
+      "step": 242
+    },
+    {
+      "epoch": 3.4722719141323792,
+      "grad_norm": 0.04469459131360054,
+      "learning_rate": 3.0719022296685204e-05,
+      "loss": 0.8040771484375,
+      "step": 243
+    },
+    {
+      "epoch": 3.486583184257603,
+      "grad_norm": 0.05732662230730057,
+      "learning_rate": 3.0368961084955738e-05,
+      "loss": 0.9154052734375,
+      "step": 244
+    },
+    {
+      "epoch": 3.500894454382826,
+      "grad_norm": 0.044572751969099045,
+      "learning_rate": 3.0021015212200643e-05,
+      "loss": 0.721710205078125,
+      "step": 245
+    },
+    {
+      "epoch": 3.51520572450805,
+      "grad_norm": 0.05070580169558525,
+      "learning_rate": 2.9675214560369634e-05,
+      "loss": 0.7707366943359375,
+      "step": 246
+    },
+    {
+      "epoch": 3.5295169946332736,
+      "grad_norm": 0.05199800059199333,
+      "learning_rate": 2.9331588827178596e-05,
+      "loss": 0.9173583984375,
+      "step": 247
+    },
+    {
+      "epoch": 3.5438282647584973,
+      "grad_norm": 0.05229436233639717,
+      "learning_rate": 2.8990167523559202e-05,
+      "loss": 0.8977813720703125,
+      "step": 248
+    },
+    {
+      "epoch": 3.558139534883721,
+      "grad_norm": 0.04129970073699951,
+      "learning_rate": 2.865097997112442e-05,
+      "loss": 0.72802734375,
+      "step": 249
+    },
+    {
+      "epoch": 3.5724508050089447,
+      "grad_norm": 0.05535365268588066,
+      "learning_rate": 2.8314055299650354e-05,
+      "loss": 0.85382080078125,
+      "step": 250
+    },
+    {
+      "epoch": 3.586762075134168,
+      "grad_norm": 0.04852509871125221,
+      "learning_rate": 2.797942244457461e-05,
+      "loss": 0.7962646484375,
+      "step": 251
+    },
+    {
+      "epoch": 3.6010733452593917,
+      "grad_norm": 0.044942617416381836,
+      "learning_rate": 2.7647110144511213e-05,
+      "loss": 0.854217529296875,
+      "step": 252
+    },
+    {
+      "epoch": 3.6153846153846154,
+      "grad_norm": 0.05859600752592087,
+      "learning_rate": 2.73171469387825e-05,
+      "loss": 0.7682914733886719,
+      "step": 253
+    },
+    {
+      "epoch": 3.629695885509839,
+      "grad_norm": 0.046633996069431305,
+      "learning_rate": 2.6989561164968252e-05,
+      "loss": 0.852691650390625,
+      "step": 254
+    },
+    {
+      "epoch": 3.6440071556350624,
+      "grad_norm": 0.05355151370167732,
+      "learning_rate": 2.6664380956471868e-05,
+      "loss": 0.8275146484375,
+      "step": 255
+    },
+    {
+      "epoch": 3.658318425760286,
+      "grad_norm": 0.054016899317502975,
+      "learning_rate": 2.6341634240104396e-05,
+      "loss": 0.870208740234375,
+      "step": 256
+    },
+    {
+      "epoch": 3.67262969588551,
+      "grad_norm": 0.04277198761701584,
+      "learning_rate": 2.6021348733686003e-05,
+      "loss": 0.69940185546875,
+      "step": 257
+    },
+    {
+      "epoch": 3.6869409660107335,
+      "grad_norm": 0.04847847297787666,
+      "learning_rate": 2.570355194366564e-05,
+      "loss": 0.774078369140625,
+      "step": 258
+    },
+    {
+      "epoch": 3.701252236135957,
+      "grad_norm": 0.04925099015235901,
+      "learning_rate": 2.5388271162758766e-05,
+      "loss": 0.8150634765625,
+      "step": 259
+    },
+    {
+      "epoch": 3.715563506261181,
+      "grad_norm": 0.053078748285770416,
+      "learning_rate": 2.507553346760335e-05,
+      "loss": 0.812469482421875,
+      "step": 260
+    },
+    {
+      "epoch": 3.729874776386404,
+      "grad_norm": 0.04576817527413368,
+      "learning_rate": 2.4765365716434523e-05,
+      "loss": 0.7979583740234375,
+      "step": 261
+    },
+    {
+      "epoch": 3.744186046511628,
+      "grad_norm": 0.0416908785700798,
+      "learning_rate": 2.4457794546778045e-05,
+      "loss": 0.65997314453125,
+      "step": 262
+    },
+    {
+      "epoch": 3.7584973166368516,
+      "grad_norm": 0.06115897744894028,
+      "learning_rate": 2.415284637316251e-05,
+      "loss": 0.86102294921875,
+      "step": 263
+    },
+    {
+      "epoch": 3.772808586762075,
+      "grad_norm": 0.04422996565699577,
+      "learning_rate": 2.3850547384850975e-05,
+      "loss": 0.81732177734375,
+      "step": 264
+    },
+    {
+      "epoch": 3.7871198568872986,
+      "grad_norm": 0.049698926508426666,
+      "learning_rate": 2.35509235435917e-05,
+      "loss": 0.735076904296875,
+      "step": 265
+    },
+    {
+      "epoch": 3.8014311270125223,
+      "grad_norm": 0.04607092961668968,
+      "learning_rate": 2.325400058138855e-05,
+      "loss": 0.81927490234375,
+      "step": 266
+    },
+    {
+      "epoch": 3.815742397137746,
+      "grad_norm": 0.04319271072745323,
+      "learning_rate": 2.2959803998291166e-05,
+      "loss": 0.6791096925735474,
+      "step": 267
+    },
+    {
+      "epoch": 3.8300536672629697,
+      "grad_norm": 0.044164448976516724,
+      "learning_rate": 2.266835906020491e-05,
+      "loss": 0.748077392578125,
+      "step": 268
+    },
+    {
+      "epoch": 3.8443649373881934,
+      "grad_norm": 0.05479202792048454,
+      "learning_rate": 2.237969079672102e-05,
+      "loss": 0.804534912109375,
+      "step": 269
+    },
+    {
+      "epoch": 3.8586762075134167,
+      "grad_norm": 0.053471047431230545,
+      "learning_rate": 2.209382399896712e-05,
+      "loss": 0.786590576171875,
+      "step": 270
+    },
+    {
+      "epoch": 3.8729874776386404,
+      "grad_norm": 0.049129754304885864,
+      "learning_rate": 2.181078321747801e-05,
+      "loss": 0.8824462890625,
+      "step": 271
+    },
+    {
+      "epoch": 3.887298747763864,
+      "grad_norm": 0.04948171600699425,
+      "learning_rate": 2.1530592760087368e-05,
+      "loss": 0.797332763671875,
+      "step": 272
+    },
+    {
+      "epoch": 3.901610017889088,
+      "grad_norm": 0.06070393696427345,
+      "learning_rate": 2.1253276689840047e-05,
+      "loss": 0.794189453125,
+      "step": 273
+    },
+    {
+      "epoch": 3.915921288014311,
+      "grad_norm": 0.05564431473612785,
+      "learning_rate": 2.097885882292559e-05,
+      "loss": 0.8990478515625,
+      "step": 274
+    },
+    {
+      "epoch": 3.9302325581395348,
+      "grad_norm": 0.054521072655916214,
+      "learning_rate": 2.0707362726632883e-05,
+      "loss": 0.835296630859375,
+      "step": 275
+    },
+    {
+      "epoch": 3.9445438282647585,
+      "grad_norm": 0.04621365666389465,
+      "learning_rate": 2.0438811717326113e-05,
+      "loss": 0.785064697265625,
+      "step": 276
+    },
+    {
+      "epoch": 3.958855098389982,
+      "grad_norm": 0.046327609568834305,
+      "learning_rate": 2.017322885844236e-05,
+      "loss": 0.8217315673828125,
+      "step": 277
+    },
+    {
+      "epoch": 3.973166368515206,
+      "grad_norm": 0.06859248876571655,
+      "learning_rate": 1.9910636958510937e-05,
+      "loss": 0.82769775390625,
+      "step": 278
+    },
+    {
+      "epoch": 3.9874776386404296,
+      "grad_norm": 0.05423009395599365,
+      "learning_rate": 1.9651058569194476e-05,
+      "loss": 0.8031005859375,
+      "step": 279
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.06038810685276985,
+      "learning_rate": 1.9394515983352214e-05,
+      "loss": 0.887939453125,
+      "step": 280
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 0.81903076171875,
+      "eval_runtime": 1.8811,
+      "eval_samples_per_second": 2.126,
+      "eval_steps_per_second": 0.532,
+      "step": 280
+    },
+    {
+      "epoch": 4.014311270125224,
+      "grad_norm": 0.07015524059534073,
+      "learning_rate": 1.914103123312551e-05,
+      "loss": 0.857025146484375,
+      "step": 281
+    },
+    {
+      "epoch": 4.028622540250447,
+      "grad_norm": 0.0419398657977581,
+      "learning_rate": 1.8890626088045595e-05,
+      "loss": 0.6813812255859375,
+      "step": 282
+    },
+    {
+      "epoch": 4.042933810375671,
+      "grad_norm": 0.059527792036533356,
+      "learning_rate": 1.8643322053164086e-05,
+      "loss": 0.86236572265625,
+      "step": 283
+    },
+    {
+      "epoch": 4.057245080500895,
+      "grad_norm": 0.051336586475372314,
+      "learning_rate": 1.8399140367206043e-05,
+      "loss": 0.802978515625,
+      "step": 284
+    },
+    {
+      "epoch": 4.071556350626118,
+      "grad_norm": 0.05357631668448448,
+      "learning_rate": 1.815810200074597e-05,
+      "loss": 0.8465576171875,
+      "step": 285
+    },
+    {
+      "epoch": 4.085867620751341,
+      "grad_norm": 0.04732433706521988,
+      "learning_rate": 1.79202276544069e-05,
+      "loss": 0.746490478515625,
+      "step": 286
+    },
+    {
+      "epoch": 4.100178890876565,
+      "grad_norm": 0.04794985428452492,
+      "learning_rate": 1.768553775708252e-05,
+      "loss": 0.77557373046875,
+      "step": 287
+    },
+    {
+      "epoch": 4.114490161001789,
+      "grad_norm": 0.055795907974243164,
+      "learning_rate": 1.7454052464182778e-05,
+      "loss": 0.76605224609375,
+      "step": 288
+    },
+    {
+      "epoch": 4.1288014311270125,
+      "grad_norm": 0.04714180901646614,
+      "learning_rate": 1.7225791655902903e-05,
+      "loss": 0.74322509765625,
+      "step": 289
+    },
+    {
+      "epoch": 4.143112701252236,
+      "grad_norm": 0.04985620453953743,
+      "learning_rate": 1.700077493551604e-05,
+      "loss": 0.82928466796875,
+      "step": 290
+    },
+    {
+      "epoch": 4.15742397137746,
+      "grad_norm": 0.0481402762234211,
+      "learning_rate": 1.6779021627689755e-05,
+      "loss": 0.84930419921875,
+      "step": 291
+    },
+    {
+      "epoch": 4.171735241502684,
+      "grad_norm": 0.04338670149445534,
+      "learning_rate": 1.6560550776826373e-05,
+      "loss": 0.747039794921875,
+      "step": 292
+    },
+    {
+      "epoch": 4.186046511627907,
+      "grad_norm": 0.04750082269310951,
+      "learning_rate": 1.6345381145427418e-05,
+      "loss": 0.848785400390625,
+      "step": 293
+    },
+    {
+      "epoch": 4.200357781753131,
+      "grad_norm": 0.04611317813396454,
+      "learning_rate": 1.6133531212482327e-05,
+      "loss": 0.819671630859375,
+      "step": 294
+    },
+    {
+      "epoch": 4.214669051878354,
+      "grad_norm": 0.04860629141330719,
+      "learning_rate": 1.5925019171881382e-05,
+      "loss": 0.85125732421875,
+      "step": 295
+    },
+    {
+      "epoch": 4.2289803220035775,
+      "grad_norm": 0.0458010658621788,
+      "learning_rate": 1.5719862930853235e-05,
+      "loss": 0.7530288696289062,
+      "step": 296
+    },
+    {
+      "epoch": 4.243291592128801,
+      "grad_norm": 0.04795634746551514,
+      "learning_rate": 1.5518080108427063e-05,
+      "loss": 0.7821044921875,
+      "step": 297
+    },
+    {
+      "epoch": 4.257602862254025,
+      "grad_norm": 0.0468454472720623,
+      "learning_rate": 1.5319688033919318e-05,
+      "loss": 0.89129638671875,
+      "step": 298
+    },
+    {
+      "epoch": 4.271914132379249,
+      "grad_norm": 0.05388948693871498,
+      "learning_rate": 1.5124703745445595e-05,
+      "loss": 0.92071533203125,
+      "step": 299
+    },
+    {
+      "epoch": 4.286225402504472,
+      "grad_norm": 0.04812207818031311,
+      "learning_rate": 1.4933143988457238e-05,
+      "loss": 0.899871826171875,
+      "step": 300
+    },
+    {
+      "epoch": 4.300536672629696,
+      "grad_norm": 0.044448696076869965,
+      "learning_rate": 1.4745025214303371e-05,
+      "loss": 0.795196533203125,
+      "step": 301
+    },
+    {
+      "epoch": 4.31484794275492,
+      "grad_norm": 0.040749210864305496,
+      "learning_rate": 1.4560363578817951e-05,
+      "loss": 0.74505615234375,
+      "step": 302
+    },
+    {
+      "epoch": 4.3291592128801435,
+      "grad_norm": 0.047416653484106064,
+      "learning_rate": 1.4379174940932303e-05,
+      "loss": 0.80706787109375,
+      "step": 303
+    },
+    {
+      "epoch": 4.343470483005366,
+      "grad_norm": 0.04128362610936165,
+      "learning_rate": 1.4201474861313121e-05,
+      "loss": 0.740570068359375,
+      "step": 304
+    },
+    {
+      "epoch": 4.35778175313059,
+      "grad_norm": 0.04533904790878296,
+      "learning_rate": 1.4027278601026172e-05,
+      "loss": 0.7730712890625,
+      "step": 305
+    },
+    {
+      "epoch": 4.372093023255814,
+      "grad_norm": 0.04487519711256027,
+      "learning_rate": 1.3856601120225573e-05,
+      "loss": 0.8300323486328125,
+      "step": 306
+    },
+    {
+      "epoch": 4.386404293381037,
+      "grad_norm": 0.05910687893629074,
+      "learning_rate": 1.3689457076869072e-05,
+      "loss": 0.887298583984375,
+      "step": 307
+    },
+    {
+      "epoch": 4.400715563506261,
+      "grad_norm": 0.04138937592506409,
+      "learning_rate": 1.3525860825459119e-05,
+      "loss": 0.783050537109375,
+      "step": 308
+    },
+    {
+      "epoch": 4.415026833631485,
+      "grad_norm": 0.0495053194463253,
+      "learning_rate": 1.336582641581019e-05,
+      "loss": 0.7254791259765625,
+      "step": 309
+    },
+    {
+      "epoch": 4.4293381037567086,
+      "grad_norm": 0.042171381413936615,
+      "learning_rate": 1.3209367591842106e-05,
+      "loss": 0.6973648071289062,
+      "step": 310
+    },
+    {
+      "epoch": 4.443649373881932,
+      "grad_norm": 0.043345857411623,
+      "learning_rate": 1.3056497790399712e-05,
+      "loss": 0.798492431640625,
+      "step": 311
+    },
+    {
+      "epoch": 4.457960644007156,
+      "grad_norm": 0.05245497077703476,
+      "learning_rate": 1.2907230140098895e-05,
+      "loss": 0.887603759765625,
+      "step": 312
+    },
+    {
+      "epoch": 4.47227191413238,
+      "grad_norm": 0.04273322969675064,
+      "learning_rate": 1.2761577460199106e-05,
+      "loss": 0.797119140625,
+      "step": 313
+    },
+    {
+      "epoch": 4.4865831842576025,
+      "grad_norm": 0.054058343172073364,
+      "learning_rate": 1.261955225950242e-05,
+      "loss": 0.90740966796875,
+      "step": 314
+    },
+    {
+      "epoch": 4.500894454382826,
+      "grad_norm": 0.0415491908788681,
+      "learning_rate": 1.248116673527925e-05,
+      "loss": 0.715087890625,
+      "step": 315
+    },
+    {
+      "epoch": 4.51520572450805,
+      "grad_norm": 0.04854319244623184,
+      "learning_rate": 1.2346432772220848e-05,
+      "loss": 0.76434326171875,
+      "step": 316
+    },
+    {
+      "epoch": 4.529516994633274,
+      "grad_norm": 0.049443408846855164,
+      "learning_rate": 1.2215361941418645e-05,
+      "loss": 0.90960693359375,
+      "step": 317
+    },
+    {
+      "epoch": 4.543828264758497,
+      "grad_norm": 0.04545610770583153,
+      "learning_rate": 1.2087965499370513e-05,
+      "loss": 0.8902740478515625,
+      "step": 318
+    },
+    {
+      "epoch": 4.558139534883721,
+      "grad_norm": 0.04163367301225662,
+      "learning_rate": 1.1964254387014006e-05,
+      "loss": 0.722259521484375,
+      "step": 319
+    },
+    {
+      "epoch": 4.572450805008945,
+      "grad_norm": 0.04975705221295357,
+      "learning_rate": 1.1844239228786792e-05,
+      "loss": 0.84649658203125,
+      "step": 320
+    },
+    {
+      "epoch": 4.5867620751341684,
+      "grad_norm": 0.04515090212225914,
+      "learning_rate": 1.1727930331714196e-05,
+      "loss": 0.790313720703125,
+      "step": 321
+    },
+    {
+      "epoch": 4.601073345259392,
+      "grad_norm": 0.04513639956712723,
+      "learning_rate": 1.1615337684524008e-05,
+      "loss": 0.84814453125,
+      "step": 322
+    },
+    {
+      "epoch": 4.615384615384615,
+      "grad_norm": 0.043038077652454376,
+      "learning_rate": 1.1506470956788658e-05,
+      "loss": 0.7621650695800781,
+      "step": 323
+    },
+    {
+      "epoch": 4.629695885509839,
+      "grad_norm": 0.04268338903784752,
+      "learning_rate": 1.1401339498094774e-05,
+      "loss": 0.84619140625,
+      "step": 324
+    },
+    {
+      "epoch": 4.644007155635062,
+      "grad_norm": 0.05020027235150337,
+      "learning_rate": 1.1299952337240244e-05,
+      "loss": 0.82080078125,
+      "step": 325
+    },
+    {
+      "epoch": 4.658318425760286,
+      "grad_norm": 0.049357157200574875,
+      "learning_rate": 1.1202318181458806e-05,
+      "loss": 0.86383056640625,
+      "step": 326
+    },
+    {
+      "epoch": 4.67262969588551,
+      "grad_norm": 0.03828378766775131,
+      "learning_rate": 1.1108445415672242e-05,
+      "loss": 0.6940536499023438,
+      "step": 327
+    },
+    {
+      "epoch": 4.6869409660107335,
+      "grad_norm": 0.04471000283956528,
+      "learning_rate": 1.1018342101770298e-05,
+      "loss": 0.7677001953125,
+      "step": 328
+    },
+    {
+      "epoch": 4.701252236135957,
+      "grad_norm": 0.043653134256601334,
+      "learning_rate": 1.0932015977918326e-05,
+      "loss": 0.8082275390625,
+      "step": 329
+    },
+    {
+      "epoch": 4.715563506261181,
+      "grad_norm": 0.044682882726192474,
+      "learning_rate": 1.0849474457892681e-05,
+      "loss": 0.806304931640625,
+      "step": 330
+    },
+    {
+      "epoch": 4.729874776386405,
+      "grad_norm": 0.04207319766283035,
+      "learning_rate": 1.0770724630444079e-05,
+      "loss": 0.792144775390625,
+      "step": 331
+    },
+    {
+      "epoch": 4.7441860465116275,
+      "grad_norm": 0.036135170608758926,
+      "learning_rate": 1.0695773258688737e-05,
+      "loss": 0.65496826171875,
+      "step": 332
+    },
+    {
+      "epoch": 4.758497316636851,
+      "grad_norm": 0.051057010889053345,
+      "learning_rate": 1.0624626779527622e-05,
+      "loss": 0.85504150390625,
+      "step": 333
+    },
+    {
+      "epoch": 4.772808586762075,
+      "grad_norm": 0.04251720756292343,
+      "learning_rate": 1.0557291303093595e-05,
+      "loss": 0.811614990234375,
+      "step": 334
+    },
+    {
+      "epoch": 4.787119856887299,
+      "grad_norm": 0.044787079095840454,
+      "learning_rate": 1.0493772612226669e-05,
+      "loss": 0.72979736328125,
+      "step": 335
+    },
+    {
+      "epoch": 4.801431127012522,
+      "grad_norm": 0.04118410870432854,
+      "learning_rate": 1.0434076161977406e-05,
+      "loss": 0.813873291015625,
+      "step": 336
+    },
+    {
+      "epoch": 4.815742397137746,
+      "grad_norm": 0.045466650277376175,
+      "learning_rate": 1.037820707913839e-05,
+      "loss": 0.6741883754730225,
+      "step": 337
+    },
+    {
+      "epoch": 4.83005366726297,
+      "grad_norm": 0.04216128960251808,
+      "learning_rate": 1.0326170161803966e-05,
+      "loss": 0.743499755859375,
+      "step": 338
+    },
+    {
+      "epoch": 4.844364937388193,
+      "grad_norm": 0.04469231888651848,
+      "learning_rate": 1.0277969878958153e-05,
+      "loss": 0.79937744140625,
+      "step": 339
+    },
+    {
+      "epoch": 4.858676207513417,
+      "grad_norm": 0.049886927008628845,
+      "learning_rate": 1.0233610370090848e-05,
+      "loss": 0.781097412109375,
+      "step": 340
+    },
+    {
+      "epoch": 4.87298747763864,
+      "grad_norm": 0.05182114243507385,
+      "learning_rate": 1.0193095444842341e-05,
+      "loss": 0.8765869140625,
+      "step": 341
+    },
+    {
+      "epoch": 4.8872987477638645,
+      "grad_norm": 0.044851623475551605,
+      "learning_rate": 1.0156428582676107e-05,
+      "loss": 0.7928466796875,
+      "step": 342
+    },
+    {
+      "epoch": 4.901610017889087,
+      "grad_norm": 0.05196821317076683,
+      "learning_rate": 1.0123612932580014e-05,
+      "loss": 0.78936767578125,
+      "step": 343
+    },
+    {
+      "epoch": 4.915921288014311,
+      "grad_norm": 0.05170882120728493,
+      "learning_rate": 1.0094651312795863e-05,
+      "loss": 0.89337158203125,
+      "step": 344
+    },
+    {
+      "epoch": 4.930232558139535,
+      "grad_norm": 0.05232628434896469,
+      "learning_rate": 1.0069546210577381e-05,
+      "loss": 0.829742431640625,
+      "step": 345
+    },
+    {
+      "epoch": 4.9445438282647585,
+      "grad_norm": 0.04137434810400009,
+      "learning_rate": 1.0048299781976574e-05,
+      "loss": 0.7803955078125,
+      "step": 346
+    },
+    {
+      "epoch": 4.958855098389982,
+      "grad_norm": 0.04509047046303749,
+      "learning_rate": 1.0030913851658609e-05,
+      "loss": 0.81707763671875,
+      "step": 347
+    },
+    {
+      "epoch": 4.973166368515206,
+      "grad_norm": 0.05586308240890503,
+      "learning_rate": 1.0017389912745069e-05,
+      "loss": 0.82281494140625,
+      "step": 348
+    },
+    {
+      "epoch": 4.98747763864043,
+      "grad_norm": 0.051874782890081406,
+      "learning_rate": 1.0007729126685756e-05,
+      "loss": 0.798797607421875,
+      "step": 349
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.051923204213380814,
+      "learning_rate": 1.0001932323158918e-05,
+      "loss": 0.88311767578125,
+      "step": 350
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 0.81512451171875,
+      "eval_runtime": 1.8807,
+      "eval_samples_per_second": 2.127,
+      "eval_steps_per_second": 0.532,
+      "step": 350
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 350,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 5,
+  "save_steps": 0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.471234892721553e+19,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

training_metadata.json

@@ -0,0 +1,29 @@
+{
+  "model": "openai/gpt-oss-20b",
+  "challenge": "Adaption AutoScientist Challenge Part 2",
+  "project": "finpath-autoscientist",
+  "category": "personal-finance",
+  "training_method": "sft",
+  "training_type": "lora",
+  "data_format": "instruction",
+  "adapted_dataset_rows": 28000,
+  "hyperparams": {
+    "base_model_size": "20B",
+    "n_epochs": 5,
+    "batch_size": "max",
+    "learning_rate": 0.0001,
+    "lora": true,
+    "lora_r": 64,
+    "lora_alpha": 128,
+    "lora_dropout": 0,
+    "lora_trainable_modules": "q_proj,k_proj,v_proj,o_proj",
+    "lr_scheduler_type": "cosine",
+    "min_lr_ratio": 0.1,
+    "scheduler_num_cycles": 0.5,
+    "warmup_ratio": 0.03,
+    "max_grad_norm": 1,
+    "weight_decay": 0.01,
+    "n_evals": 5,
+    "train_on_inputs": false
+  }
+}