Upload folder using huggingface_hub

.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 +1,210 @@
+---
+base_model: unsloth/gemma-4-e2b-it-unsloth-bnb-4bit
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:unsloth/gemma-4-e2b-it-unsloth-bnb-4bit
+- lora
+- sft
+- transformers
+- trl
+- unsloth
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.18.1

adapter_config.json

@@ -0,0 +1,50 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": {
+    "base_model_class": "Gemma4ForConditionalGeneration",
+    "parent_library": "transformers.models.gemma4.modeling_gemma4",
+    "unsloth_fixed": true
+  },
+  "base_model_name_or_path": "unsloth/gemma-4-e2b-it-unsloth-bnb-4bit",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "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",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "up_proj",
+    "v_proj",
+    "q_proj",
+    "k_proj",
+    "o_proj",
+    "gate_proj",
+    "down_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba4bd0236cb0ec67d26d56301ddf84aef566b75ddf4f2503451d5aaff0921d55
+size 496752408

chat_template.jinja

@@ -0,0 +1,351 @@
+{%- macro format_parameters(properties, required, filter_keys=false) -%}
+    {%- set standard_keys = ['description', 'type', 'properties', 'required', 'nullable'] -%}
+    {%- set ns = namespace(found_first=false) -%}
+    {%- for key, value in properties | dictsort -%}
+        {%- set add_comma = false -%}
+        {%- if not filter_keys or key not in standard_keys -%}
+            {%- if ns.found_first %},{% endif -%}
+            {%- set ns.found_first = true -%}
+            {{ key }}:{
+            {%- if value['description'] -%}
+                description:<|"|>{{ value['description'] }}<|"|>
+                {%- set add_comma = true -%}
+            {%- endif -%}
+            {%- if value['type'] | upper == 'STRING' -%}
+                {%- if value['enum'] -%}
+                    {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
+                    enum:{{ format_argument(value['enum']) }}
+                {%- endif -%}
+            {%- elif value['type'] | upper == 'ARRAY' -%}
+                {%- if value['items'] is mapping and value['items'] -%}
+                    {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
+                    items:{
+                    {%- set ns_items = namespace(found_first=false) -%}
+                    {%- for item_key, item_value in value['items'] | dictsort -%}
+                        {%- if item_value is not none -%}
+                            {%- if ns_items.found_first %},{% endif -%}
+                            {%- set ns_items.found_first = true -%}
+                            {%- if item_key == 'properties' -%}
+                                properties:{
+                                {%- if item_value is mapping -%}
+                                    {{- format_parameters(item_value, value['items']['required'] | default([])) -}}
+                                {%- endif -%}
+                                }
+                            {%- elif item_key == 'required' -%}
+                                required:[
+                                {%- for req_item in item_value -%}
+                                    <|"|>{{- req_item -}}<|"|>
+                                    {%- if not loop.last %},{% endif -%}
+                                {%- endfor -%}
+                                ]
+                            {%- elif item_key == 'type' -%}
+                                {%- if item_value is string -%}
+                                    type:{{ format_argument(item_value | upper) }}
+                                {%- else -%}
+                                    type:{{ format_argument(item_value | map('upper') | list) }}
+                                {%- endif -%}
+                            {%- else -%}
+                                {{ item_key }}:{{ format_argument(item_value) }}
+                            {%- endif -%}
+                        {%- endif -%}
+                    {%- endfor -%}
+                    }
+                {%- endif -%}
+            {%- endif -%}
+            {%- if value['nullable'] %}
+                {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
+                nullable:true
+            {%- endif -%}
+            {%- if value['type'] | upper == 'OBJECT' -%}
+                {%- if value['properties'] is defined and value['properties'] is mapping -%}
+                    {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
+                    properties:{
+                    {{- format_parameters(value['properties'], value['required'] | default([])) -}}
+                    }
+                {%- elif value is mapping -%}
+                    {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
+                    properties:{
+                    {{- format_parameters(value, value['required'] | default([]), filter_keys=true) -}}
+                    }
+                {%- endif -%}
+                {%- if value['required'] -%}
+                    {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
+                    required:[
+                    {%- for item in value['required'] | default([]) -%}
+                        <|"|>{{- item -}}<|"|>
+                        {%- if not loop.last %},{% endif -%}
+                    {%- endfor -%}
+                    ]
+                {%- endif -%}
+            {%- endif -%}
+            {%- if add_comma %},{%- else -%} {%- set add_comma = true -%} {% endif -%}
+            type:<|"|>{{ value['type'] | upper }}<|"|>}
+        {%- endif -%}
+    {%- endfor -%}
+{%- endmacro -%}
+{%- macro format_function_declaration(tool_data) -%}
+    declaration:{{- tool_data['function']['name'] -}}{description:<|"|>{{- tool_data['function']['description'] -}}<|"|>
+    {%- set params = tool_data['function']['parameters'] -%}
+    {%- if params -%}
+        ,parameters:{
+        {%- if params['properties'] -%}
+            properties:{ {{- format_parameters(params['properties'], params['required']) -}} },
+        {%- endif -%}
+        {%- if params['required'] -%}
+            required:[
+            {%- for item in params['required'] -%}
+                <|"|>{{- item -}}<|"|>
+                {{- ',' if not loop.last -}}
+            {%- endfor -%}
+            ],
+        {%- endif -%}
+        {%- if params['type'] -%}
+            type:<|"|>{{- params['type'] | upper -}}<|"|>}
+        {%- endif -%}
+    {%- endif -%}
+    {%- if 'response' in tool_data['function'] -%}
+        {%- set response_declaration = tool_data['function']['response'] -%}
+        ,response:{
+        {%- if response_declaration['description'] -%}
+            description:<|"|>{{- response_declaration['description'] -}}<|"|>,
+        {%- endif -%}
+        {%- if response_declaration['type'] | upper == 'OBJECT' -%}
+            type:<|"|>{{- response_declaration['type'] | upper -}}<|"|>}
+        {%- endif -%}
+    {%- endif -%}
+    }
+{%- endmacro -%}
+{%- macro format_argument(argument, escape_keys=True) -%}
+    {%- if argument is string -%}
+        {{- '<|"|>' + argument + '<|"|>' -}}
+    {%- elif argument is boolean -%}
+        {{- 'true' if argument else 'false' -}}
+    {%- elif argument is mapping -%}
+        {{- '{' -}}
+        {%- set ns = namespace(found_first=false) -%}
+        {%- for key, value in argument | dictsort -%}
+            {%- if ns.found_first %},{% endif -%}
+            {%- set ns.found_first = true -%}
+            {%- if escape_keys -%}
+                {{- '<|"|>' + key + '<|"|>' -}}
+            {%- else -%}
+                {{- key -}}
+            {%- endif -%}
+            :{{- format_argument(value, escape_keys=escape_keys) -}}
+        {%- endfor -%}
+        {{- '}' -}}
+    {%- elif argument is sequence -%}
+        {{- '[' -}}
+        {%- for item in argument -%}
+            {{- format_argument(item, escape_keys=escape_keys) -}}
+            {%- if not loop.last %},{% endif -%}
+        {%- endfor -%}
+        {{- ']' -}}
+    {%- else -%}
+        {{- argument -}}
+    {%- endif -%}
+{%- endmacro -%}
+{%- macro strip_thinking(text) -%}
+    {%- set ns = namespace(result='') -%}
+    {%- for part in text.split('<channel|>') -%}
+        {%- if '<|channel>' in part -%}
+            {%- set ns.result = ns.result + part.split('<|channel>')[0] -%}
+        {%- else -%}
+            {%- set ns.result = ns.result + part -%}
+        {%- endif -%}
+    {%- endfor -%}
+    {{- ns.result | trim -}}
+{%- endmacro -%}
+
+{%- macro format_tool_response_block(tool_name, response) -%}
+    {{- '<|tool_response>' -}}
+    {%- if response is mapping -%}
+        {{- 'response:' + tool_name + '{' -}}
+        {%- for key, value in response | dictsort -%}
+            {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
+            {%- if not loop.last %},{% endif -%}
+        {%- endfor -%}
+        {{- '}' -}}
+    {%- else -%}
+        {{- 'response:' + tool_name + '{value:' + format_argument(response, escape_keys=False) + '}' -}}
+    {%- endif -%}
+    {{- '<tool_response|>' -}}
+{%- endmacro -%}
+
+{%- set ns = namespace(prev_message_type=None) -%}
+{%- set loop_messages = messages -%}
+{{- bos_token -}}
+{#- Handle System/Tool Definitions Block -#}
+{%- if (enable_thinking is defined and enable_thinking) or tools or messages[0]['role'] in ['system', 'developer'] -%}
+    {{- '<|turn>system\n' -}}
+    {#- Inject Thinking token at the very top of the FIRST system turn -#}
+    {%- if enable_thinking is defined and enable_thinking -%}
+        {{- '<|think|>\n' -}}
+        {%- set ns.prev_message_type = 'think' -%}
+    {%- endif -%}
+    {%- if messages[0]['role'] in ['system', 'developer'] -%}
+        {%- if messages[0]['content'] is string -%}
+            {{- messages[0]['content'] | trim -}}
+        {%- elif messages[0]['content'] is sequence -%}
+            {%- for item in messages[0]['content'] -%}
+                {{- item['text'] | trim + ' '-}}
+            {%- endfor -%}
+        {%- endif -%}
+        {%- set loop_messages = messages[1:] -%}
+    {%- endif -%}
+    {%- if tools -%}
+        {%- for tool in tools %}
+            {{- '<|tool>' -}}
+            {{- format_function_declaration(tool) | trim -}}
+            {{- '<tool|>' -}}
+        {%- endfor %}
+        {%- set ns.prev_message_type = 'tool' -%}
+    {%- endif -%}
+    {{- '<turn|>\n' -}}
+{%- endif %}
+
+{#- Pre-scan: find last user message index for reasoning guard -#}
+{%- set ns_turn = namespace(last_user_idx=-1) -%}
+{%- for i in range(loop_messages | length) -%}
+    {%- if loop_messages[i]['role'] == 'user' -%}
+        {%- set ns_turn.last_user_idx = i -%}
+    {%- endif -%}
+{%- endfor -%}
+
+{#- Loop through messages -#}
+{%- for message in loop_messages -%}
+    {%- if message['role'] != 'tool' -%}
+    {%- set ns.prev_message_type = None -%}
+    {%- set role = 'model' if message['role'] == 'assistant' else message['role'] -%}
+    {#- Detect continuation: suppress duplicate <|turn>model when previous non-tool message was also assistant -#}
+    {%- set prev_nt = namespace(role=None, found=false) -%}
+    {%- if loop.index0 > 0 -%}
+        {%- for j in range(loop.index0 - 1, -1, -1) -%}
+            {%- if not prev_nt.found -%}
+                {%- if loop_messages[j]['role'] != 'tool' -%}
+                    {%- set prev_nt.role = loop_messages[j]['role'] -%}
+                    {%- set prev_nt.found = true -%}
+                {%- endif -%}
+            {%- endif -%}
+        {%- endfor -%}
+    {%- endif -%}
+    {%- set continue_same_model_turn = (role == 'model' and prev_nt.role == 'assistant') -%}
+    {%- if not continue_same_model_turn -%}
+        {{- '<|turn>' + role + '\n' }}
+    {%- endif -%}
+
+    {#- Render reasoning/reasoning_content as thinking channel -#}
+    {%- set thinking_text = message.get('reasoning') or message.get('reasoning_content') -%}
+    {%- if thinking_text and loop.index0 > ns_turn.last_user_idx and message.get('tool_calls') -%}
+        {{- '<|channel>thought\n' + thinking_text + '\n<channel|>' -}}
+    {%- endif -%}
+
+            {%- if message['tool_calls'] -%}
+                {%- for tool_call in message['tool_calls'] -%}
+                    {%- set function = tool_call['function'] -%}
+                    {{- '<|tool_call>call:' + function['name'] + '{' -}}
+                    {%- if function['arguments'] is mapping -%}
+                        {%- set ns_args = namespace(found_first=false) -%}
+                        {%- for key, value in function['arguments'] | dictsort -%}
+                            {%- if ns_args.found_first %},{% endif -%}
+                            {%- set ns_args.found_first = true -%}
+                            {{- key -}}:{{- format_argument(value, escape_keys=False) -}}
+                        {%- endfor -%}
+                    {%- elif function['arguments'] is string -%}
+                        {{- function['arguments'] -}}
+                    {%- endif -%}
+                    {{- '}<tool_call|>' -}}
+                {%- endfor -%}
+                {%- set ns.prev_message_type = 'tool_call' -%}
+            {%- endif -%}
+
+            {%- set ns_tr_out = namespace(flag=false) -%}
+            {%- if message.get('tool_responses') -%}
+                {#- Legacy: tool_responses embedded on the assistant message (Google/Gemma native) -#}
+                {%- for tool_response in message['tool_responses'] -%}
+                    {{- format_tool_response_block(tool_response['name'] | default('unknown'), tool_response['response']) -}}
+                    {%- set ns_tr_out.flag = true -%}
+                    {%- set ns.prev_message_type = 'tool_response' -%}
+                {%- endfor -%}
+            {%- elif message.get('tool_calls') -%}
+                {#- OpenAI Chat Completions: forward-scan consecutive role:tool messages -#}
+                {%- set ns_tool_scan = namespace(stopped=false) -%}
+                {%- for k in range(loop.index0 + 1, loop_messages | length) -%}
+                    {%- if ns_tool_scan.stopped -%}
+                    {%- elif loop_messages[k]['role'] != 'tool' -%}
+                        {%- set ns_tool_scan.stopped = true -%}
+                    {%- else -%}
+                        {%- set follow = loop_messages[k] -%}
+                        {#- Resolve tool_call_id to function name -#}
+                        {%- set ns_tname = namespace(name=follow.get('name') | default('unknown')) -%}
+                        {%- for tc in message['tool_calls'] -%}
+                            {%- if tc.get('id') == follow.get('tool_call_id') -%}
+                                {%- set ns_tname.name = tc['function']['name'] -%}
+                            {%- endif -%}
+                        {%- endfor -%}
+                        {#- Handle content as string or content-parts array -#}
+                        {%- set tool_body = follow.get('content') -%}
+                        {%- if tool_body is string -%}
+                            {{- format_tool_response_block(ns_tname.name, tool_body) -}}
+                        {%- elif tool_body is sequence and tool_body is not string -%}
+                            {%- set ns_txt = namespace(s='') -%}
+                            {%- for part in tool_body -%}
+                                {%- if part.get('type') == 'text' -%}
+                                    {%- set ns_txt.s = ns_txt.s + (part.get('text') | default('')) -%}
+                                {%- endif -%}
+                            {%- endfor -%}
+                            {{- format_tool_response_block(ns_tname.name, ns_txt.s) -}}
+                        {%- else -%}
+                            {{- format_tool_response_block(ns_tname.name, tool_body) -}}
+                        {%- endif -%}
+                        {%- set ns_tr_out.flag = true -%}
+                        {%- set ns.prev_message_type = 'tool_response' -%}
+                    {%- endif -%}
+                {%- endfor -%}
+            {%- endif -%}
+
+            {%- set captured_content -%}
+            {%- if message['content'] is string -%}
+                {%- if role == 'model' -%}
+                    {{- strip_thinking(message['content']) -}}
+                {%- else -%}
+                    {{- message['content'] | trim -}}
+                {%- endif -%}
+            {%- elif message['content'] is sequence -%}
+                {%- for item in message['content'] -%}
+                    {%- if item['type'] == 'text' -%}
+                        {%- if role == 'model' -%}
+                            {{- strip_thinking(item['text']) -}}
+                        {%- else -%}
+                            {{- item['text'] | trim -}}
+                        {%- endif -%}
+                    {%- elif item['type'] == 'image' -%}
+                        {{- '<|image|>' -}}
+                        {%- set ns.prev_message_type = 'image' -%}
+                    {%- elif item['type'] == 'audio' -%}
+                        {{- '<|audio|>' -}}
+                        {%- set ns.prev_message_type = 'audio' -%}
+                    {%- elif item['type'] == 'video' -%}
+                        {{- '<|video|>' -}}
+                        {%- set ns.prev_message_type = 'video' -%}
+                    {%- endif -%}
+                {%- endfor -%}
+            {%- endif -%}
+            {%- endset -%}
+
+            {{- captured_content -}}
+            {%- set has_content = captured_content | trim | length > 0 -%}
+
+        {%- if ns.prev_message_type == 'tool_call' and not ns_tr_out.flag -%}
+            {{- '<|tool_response>' -}}
+        {%- elif not (ns_tr_out.flag and not has_content) -%}
+            {{- '<turn|>\n' -}}
+        {%- endif -%}
+    {%- endif -%}
+{%- endfor -%}
+
+{%- if add_generation_prompt -%}
+    {%- if ns.prev_message_type != 'tool_response' and ns.prev_message_type != 'tool_call' -%}
+        {{- '<|turn>model\n' -}}
+    {%- endif -%}
+{%- endif -%}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff61320420b47c1e1b4581b4c53c3bd7326f48fca596c78a632bf5c087631369
+size 206452109

processor_config.json

@@ -0,0 +1,75 @@
+{
+  "audio_ms_per_token": 40,
+  "audio_seq_length": 750,
+  "feature_extractor": {
+    "dither": 0.0,
+    "feature_extractor_type": "Gemma4AudioFeatureExtractor",
+    "feature_size": 128,
+    "fft_length": 512,
+    "fft_overdrive": false,
+    "frame_length": 320,
+    "hop_length": 160,
+    "input_scale_factor": 1.0,
+    "max_frequency": 8000.0,
+    "mel_floor": 0.001,
+    "min_frequency": 0.0,
+    "padding_side": "left",
+    "padding_value": 0.0,
+    "per_bin_mean": null,
+    "per_bin_stddev": null,
+    "preemphasis": 0.0,
+    "preemphasis_htk_flavor": true,
+    "return_attention_mask": true,
+    "sampling_rate": 16000
+  },
+  "image_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": false,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.0,
+      0.0,
+      0.0
+    ],
+    "image_processor_type": "Gemma4ImageProcessor",
+    "image_seq_length": 280,
+    "image_std": [
+      1.0,
+      1.0,
+      1.0
+    ],
+    "max_soft_tokens": 280,
+    "patch_size": 16,
+    "pooling_kernel_size": 3,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098
+  },
+  "image_seq_length": 280,
+  "processor_class": "Gemma4Processor",
+  "video_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "image_mean": [
+      0.0,
+      0.0,
+      0.0
+    ],
+    "image_std": [
+      1.0,
+      1.0,
+      1.0
+    ],
+    "max_soft_tokens": 70,
+    "num_frames": 32,
+    "patch_size": 16,
+    "pooling_kernel_size": 3,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "video_processor_type": "Gemma4VideoProcessor"
+  }
+}

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19b15d59b7d2f507771b26ca44b1603b8cdd22ee0992f3217d7e17b763986855
+size 14645

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55d5c1b55b1dc021b0d083fa48f376d4fac1639b18f7a58768d5d035271b9272
+size 1383

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e04623d938a55a7867a6a097661fe33f50bb72d27aeb99e8d2b5b4a017d6cc3
+size 1465

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc8d3a0ce36466ccc1278bf987df5f71db1719b9ca6b4118264f45cb627bfe0f
+size 32169626

tokenizer_config.json

@@ -0,0 +1,289 @@
+{
+  "audio_token": "<|audio|>",
+  "backend": "tokenizers",
+  "boa_token": "<|audio>",
+  "boi_token": "<|image>",
+  "bos_token": "<bos>",
+  "eoa_token": "<audio|>",
+  "eoc_token": "<channel|>",
+  "eoi_token": "<image|>",
+  "eos_token": "<eos>",
+  "eot_token": "<turn|>",
+  "escape_token": "<|\"|>",
+  "etc_token": "<tool_call|>",
+  "etd_token": "<tool|>",
+  "etr_token": "<tool_response|>",
+  "extra_special_tokens": [
+    "<|video|>"
+  ],
+  "image_token": "<|image|>",
+  "is_local": false,
+  "mask_token": "<mask>",
+  "model_max_length": 131072,
+  "model_specific_special_tokens": {
+    "audio_token": "<|audio|>",
+    "boa_token": "<|audio>",
+    "boi_token": "<|image>",
+    "eoa_token": "<audio|>",
+    "eoc_token": "<channel|>",
+    "eoi_token": "<image|>",
+    "eot_token": "<turn|>",
+    "escape_token": "<|\"|>",
+    "etc_token": "<tool_call|>",
+    "etd_token": "<tool|>",
+    "etr_token": "<tool_response|>",
+    "image_token": "<|image|>",
+    "soc_token": "<|channel>",
+    "sot_token": "<|turn>",
+    "stc_token": "<|tool_call>",
+    "std_token": "<|tool>",
+    "str_token": "<|tool_response>",
+    "think_token": "<|think|>"
+  },
+  "pad_token": "<pad>",
+  "padding_side": "right",
+  "processor_class": "Gemma4Processor",
+  "response_schema": {
+    "properties": {
+      "content": {
+        "type": "string"
+      },
+      "role": {
+        "const": "assistant"
+      },
+      "thinking": {
+        "type": "string"
+      },
+      "tool_calls": {
+        "items": {
+          "properties": {
+            "function": {
+              "properties": {
+                "arguments": {
+                  "additionalProperties": {},
+                  "type": "object",
+                  "x-parser": "gemma4-tool-call"
+                },
+                "name": {
+                  "type": "string"
+                }
+              },
+              "type": "object",
+              "x-regex": "call\\:(?P<name>\\w+)(?P<arguments>\\{.*\\})"
+            },
+            "type": {
+              "const": "function"
+            }
+          },
+          "type": "object"
+        },
+        "type": "array",
+        "x-regex-iterator": "<\\|tool_call>(.*?)<tool_call\\|>"
+      }
+    },
+    "type": "object",
+    "x-regex": "(\\<\\|channel\\>thought\\n(?P<thinking>.*?)\\<channel\\|\\>)?(?P<tool_calls>\\<\\|tool_call\\>.*\\<tool_call\\|\\>)?(?P<content>(?:(?!\\<turn\\|\\>)(?!\\<\\|tool_response\\>).)+)?(?:\\<turn\\|\\>|\\<\\|tool_response\\>)?"
+  },
+  "soc_token": "<|channel>",
+  "sot_token": "<|turn>",
+  "stc_token": "<|tool_call>",
+  "std_token": "<|tool>",
+  "str_token": "<|tool_response>",
+  "think_token": "<|think|>",
+  "tokenizer_class": "GemmaTokenizer",
+  "unk_token": "<unk>",
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "1": {
+      "content": "<eos>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "2": {
+      "content": "<bos>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "46": {
+      "content": "<|tool>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "47": {
+      "content": "<tool|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "48": {
+      "content": "<|tool_call>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "49": {
+      "content": "<tool_call|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "50": {
+      "content": "<|tool_response>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "51": {
+      "content": "<tool_response|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "52": {
+      "content": "<|\"|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "98": {
+      "content": "<|think|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "100": {
+      "content": "<|channel>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "101": {
+      "content": "<channel|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "105": {
+      "content": "<|turn>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "106": {
+      "content": "<turn|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "255999": {
+      "content": "<|image>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "256000": {
+      "content": "<|audio>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "258880": {
+      "content": "<|image|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "258881": {
+      "content": "<|audio|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "258882": {
+      "content": "<image|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "258883": {
+      "content": "<audio|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    },
+    "258884": {
+      "content": "<|video|>",
+      "single_word": false,
+      "lstrip": false,
+      "rstrip": false,
+      "normalized": false,
+      "special": true
+    }
+  }
+}

trainer_state.json

@@ -0,0 +1,2506 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.935483870967742,
+  "eval_steps": 30,
+  "global_step": 240,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008064516129032258,
+      "grad_norm": NaN,
+      "learning_rate": 0.0,
+      "loss": 3.4369091987609863,
+      "num_input_tokens_seen": 3376,
+      "step": 1,
+      "train_runtime": 39.4836,
+      "train_tokens_per_second": 85.504
+    },
+    {
+      "epoch": 0.016129032258064516,
+      "grad_norm": 26.560325622558594,
+      "learning_rate": 0.0,
+      "loss": 3.346249580383301,
+      "num_input_tokens_seen": 6750,
+      "step": 2,
+      "train_runtime": 45.351,
+      "train_tokens_per_second": 148.839
+    },
+    {
+      "epoch": 0.024193548387096774,
+      "grad_norm": 26.696680068969727,
+      "learning_rate": 4e-05,
+      "loss": 3.3645849227905273,
+      "num_input_tokens_seen": 10132,
+      "step": 3,
+      "train_runtime": 50.5353,
+      "train_tokens_per_second": 200.493
+    },
+    {
+      "epoch": 0.03225806451612903,
+      "grad_norm": 11.073655128479004,
+      "learning_rate": 8e-05,
+      "loss": 2.3441145420074463,
+      "num_input_tokens_seen": 13496,
+      "step": 4,
+      "train_runtime": 55.9224,
+      "train_tokens_per_second": 241.334
+    },
+    {
+      "epoch": 0.04032258064516129,
+      "grad_norm": 5.371010780334473,
+      "learning_rate": 0.00012,
+      "loss": 1.8813656568527222,
+      "num_input_tokens_seen": 16978,
+      "step": 5,
+      "train_runtime": 61.2789,
+      "train_tokens_per_second": 277.061
+    },
+    {
+      "epoch": 0.04838709677419355,
+      "grad_norm": 2.7816290855407715,
+      "learning_rate": 0.00016,
+      "loss": 1.4953795671463013,
+      "num_input_tokens_seen": 20170,
+      "step": 6,
+      "train_runtime": 66.1452,
+      "train_tokens_per_second": 304.935
+    },
+    {
+      "epoch": 0.056451612903225805,
+      "grad_norm": 2.539138078689575,
+      "learning_rate": 0.0002,
+      "loss": 1.3141508102416992,
+      "num_input_tokens_seen": 23566,
+      "step": 7,
+      "train_runtime": 71.1819,
+      "train_tokens_per_second": 331.067
+    },
+    {
+      "epoch": 0.06451612903225806,
+      "grad_norm": 2.33978009223938,
+      "learning_rate": 0.00019999967645432384,
+      "loss": 1.2497824430465698,
+      "num_input_tokens_seen": 26930,
+      "step": 8,
+      "train_runtime": 75.9867,
+      "train_tokens_per_second": 354.404
+    },
+    {
+      "epoch": 0.07258064516129033,
+      "grad_norm": 1.5424915552139282,
+      "learning_rate": 0.00019999870581938894,
+      "loss": 1.093822956085205,
+      "num_input_tokens_seen": 30108,
+      "step": 9,
+      "train_runtime": 80.6039,
+      "train_tokens_per_second": 373.53
+    },
+    {
+      "epoch": 0.08064516129032258,
+      "grad_norm": 1.7023719549179077,
+      "learning_rate": 0.0001999970881014762,
+      "loss": 1.075514793395996,
+      "num_input_tokens_seen": 33254,
+      "step": 10,
+      "train_runtime": 84.9993,
+      "train_tokens_per_second": 391.227
+    },
+    {
+      "epoch": 0.08870967741935484,
+      "grad_norm": 1.1276626586914062,
+      "learning_rate": 0.00019999482331105377,
+      "loss": 1.0134414434432983,
+      "num_input_tokens_seen": 36458,
+      "step": 11,
+      "train_runtime": 89.4147,
+      "train_tokens_per_second": 407.74
+    },
+    {
+      "epoch": 0.0967741935483871,
+      "grad_norm": 1.089064598083496,
+      "learning_rate": 0.0001999919114627769,
+      "loss": 1.0177619457244873,
+      "num_input_tokens_seen": 39848,
+      "step": 12,
+      "train_runtime": 94.1911,
+      "train_tokens_per_second": 423.055
+    },
+    {
+      "epoch": 0.10483870967741936,
+      "grad_norm": 0.8842246532440186,
+      "learning_rate": 0.00019998835257548786,
+      "loss": 0.9782118797302246,
+      "num_input_tokens_seen": 42964,
+      "step": 13,
+      "train_runtime": 98.5173,
+      "train_tokens_per_second": 436.106
+    },
+    {
+      "epoch": 0.11290322580645161,
+      "grad_norm": 0.7232466340065002,
+      "learning_rate": 0.00019998414667221596,
+      "loss": 0.895933985710144,
+      "num_input_tokens_seen": 46336,
+      "step": 14,
+      "train_runtime": 103.1565,
+      "train_tokens_per_second": 449.182
+    },
+    {
+      "epoch": 0.12096774193548387,
+      "grad_norm": 0.7355626225471497,
+      "learning_rate": 0.00019997929378017725,
+      "loss": 0.9221972227096558,
+      "num_input_tokens_seen": 49524,
+      "step": 15,
+      "train_runtime": 107.7259,
+      "train_tokens_per_second": 459.723
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 0.7629883289337158,
+      "learning_rate": 0.00019997379393077428,
+      "loss": 0.884993314743042,
+      "num_input_tokens_seen": 52878,
+      "step": 16,
+      "train_runtime": 112.3676,
+      "train_tokens_per_second": 470.58
+    },
+    {
+      "epoch": 0.13709677419354838,
+      "grad_norm": 0.7517801523208618,
+      "learning_rate": 0.00019996764715959618,
+      "loss": 0.8044445514678955,
+      "num_input_tokens_seen": 56342,
+      "step": 17,
+      "train_runtime": 117.2539,
+      "train_tokens_per_second": 480.513
+    },
+    {
+      "epoch": 0.14516129032258066,
+      "grad_norm": 0.7528966665267944,
+      "learning_rate": 0.0001999608535064182,
+      "loss": 0.8287011981010437,
+      "num_input_tokens_seen": 59734,
+      "step": 18,
+      "train_runtime": 122.2813,
+      "train_tokens_per_second": 488.497
+    },
+    {
+      "epoch": 0.1532258064516129,
+      "grad_norm": 1.0456326007843018,
+      "learning_rate": 0.0001999534130152014,
+      "loss": 0.8226615190505981,
+      "num_input_tokens_seen": 63022,
+      "step": 19,
+      "train_runtime": 127.0278,
+      "train_tokens_per_second": 496.128
+    },
+    {
+      "epoch": 0.16129032258064516,
+      "grad_norm": 0.9146727323532104,
+      "learning_rate": 0.00019994532573409262,
+      "loss": 0.8422068357467651,
+      "num_input_tokens_seen": 66380,
+      "step": 20,
+      "train_runtime": 131.9068,
+      "train_tokens_per_second": 503.234
+    },
+    {
+      "epoch": 0.1693548387096774,
+      "grad_norm": 0.8606871962547302,
+      "learning_rate": 0.0001999365917154239,
+      "loss": 0.8119312524795532,
+      "num_input_tokens_seen": 69412,
+      "step": 21,
+      "train_runtime": 136.3606,
+      "train_tokens_per_second": 509.033
+    },
+    {
+      "epoch": 0.1774193548387097,
+      "grad_norm": 0.7490668296813965,
+      "learning_rate": 0.00019992721101571236,
+      "loss": 0.8383636474609375,
+      "num_input_tokens_seen": 72682,
+      "step": 22,
+      "train_runtime": 141.0773,
+      "train_tokens_per_second": 515.193
+    },
+    {
+      "epoch": 0.18548387096774194,
+      "grad_norm": 0.7244526743888855,
+      "learning_rate": 0.0001999171836956597,
+      "loss": 0.8476752638816833,
+      "num_input_tokens_seen": 75880,
+      "step": 23,
+      "train_runtime": 145.8525,
+      "train_tokens_per_second": 520.252
+    },
+    {
+      "epoch": 0.1935483870967742,
+      "grad_norm": 0.7046738266944885,
+      "learning_rate": 0.0001999065098201518,
+      "loss": 0.7803716063499451,
+      "num_input_tokens_seen": 79364,
+      "step": 24,
+      "train_runtime": 150.8071,
+      "train_tokens_per_second": 526.262
+    },
+    {
+      "epoch": 0.20161290322580644,
+      "grad_norm": 0.6511566042900085,
+      "learning_rate": 0.00019989518945825844,
+      "loss": 0.7314240336418152,
+      "num_input_tokens_seen": 82600,
+      "step": 25,
+      "train_runtime": 155.6778,
+      "train_tokens_per_second": 530.583
+    },
+    {
+      "epoch": 0.20967741935483872,
+      "grad_norm": 0.7805576324462891,
+      "learning_rate": 0.00019988322268323268,
+      "loss": 0.7415602207183838,
+      "num_input_tokens_seen": 85954,
+      "step": 26,
+      "train_runtime": 160.5146,
+      "train_tokens_per_second": 535.49
+    },
+    {
+      "epoch": 0.21774193548387097,
+      "grad_norm": 0.7170730829238892,
+      "learning_rate": 0.00019987060957251047,
+      "loss": 0.7152543663978577,
+      "num_input_tokens_seen": 89314,
+      "step": 27,
+      "train_runtime": 165.2621,
+      "train_tokens_per_second": 540.438
+    },
+    {
+      "epoch": 0.22580645161290322,
+      "grad_norm": 0.7345016002655029,
+      "learning_rate": 0.00019985735020771017,
+      "loss": 0.7394841909408569,
+      "num_input_tokens_seen": 92550,
+      "step": 28,
+      "train_runtime": 169.9598,
+      "train_tokens_per_second": 544.54
+    },
+    {
+      "epoch": 0.23387096774193547,
+      "grad_norm": 0.6704856157302856,
+      "learning_rate": 0.00019984344467463197,
+      "loss": 0.6681033372879028,
+      "num_input_tokens_seen": 95904,
+      "step": 29,
+      "train_runtime": 174.6804,
+      "train_tokens_per_second": 549.026
+    },
+    {
+      "epoch": 0.24193548387096775,
+      "grad_norm": 0.6933154463768005,
+      "learning_rate": 0.0001998288930632574,
+      "loss": 0.7358725070953369,
+      "num_input_tokens_seen": 99314,
+      "step": 30,
+      "train_runtime": 179.518,
+      "train_tokens_per_second": 553.226
+    },
+    {
+      "epoch": 0.24193548387096775,
+      "eval_loss": 3.1636757850646973,
+      "eval_runtime": 13.9458,
+      "eval_samples_per_second": 3.729,
+      "eval_steps_per_second": 1.864,
+      "num_input_tokens_seen": 99314,
+      "step": 30
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 0.6650590300559998,
+      "learning_rate": 0.00019981369546774865,
+      "loss": 0.7312928438186646,
+      "num_input_tokens_seen": 102632,
+      "step": 31,
+      "train_runtime": 234.0652,
+      "train_tokens_per_second": 438.476
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 0.6843596696853638,
+      "learning_rate": 0.00019979785198644806,
+      "loss": 0.7392816543579102,
+      "num_input_tokens_seen": 106032,
+      "step": 32,
+      "train_runtime": 238.7402,
+      "train_tokens_per_second": 444.131
+    },
+    {
+      "epoch": 0.2661290322580645,
+      "grad_norm": 0.6795448660850525,
+      "learning_rate": 0.00019978136272187747,
+      "loss": 0.6713091135025024,
+      "num_input_tokens_seen": 109374,
+      "step": 33,
+      "train_runtime": 243.5988,
+      "train_tokens_per_second": 448.992
+    },
+    {
+      "epoch": 0.27419354838709675,
+      "grad_norm": 0.6419102549552917,
+      "learning_rate": 0.0001997642277807374,
+      "loss": 0.7344367504119873,
+      "num_input_tokens_seen": 112780,
+      "step": 34,
+      "train_runtime": 248.5335,
+      "train_tokens_per_second": 453.782
+    },
+    {
+      "epoch": 0.28225806451612906,
+      "grad_norm": 0.7528514862060547,
+      "learning_rate": 0.00019974644727390665,
+      "loss": 0.6826475262641907,
+      "num_input_tokens_seen": 116154,
+      "step": 35,
+      "train_runtime": 253.6613,
+      "train_tokens_per_second": 457.91
+    },
+    {
+      "epoch": 0.2903225806451613,
+      "grad_norm": 0.7251984477043152,
+      "learning_rate": 0.00019972802131644127,
+      "loss": 0.702574610710144,
+      "num_input_tokens_seen": 119350,
+      "step": 36,
+      "train_runtime": 258.505,
+      "train_tokens_per_second": 461.693
+    },
+    {
+      "epoch": 0.29838709677419356,
+      "grad_norm": 0.7205802202224731,
+      "learning_rate": 0.00019970895002757413,
+      "loss": 0.7223511934280396,
+      "num_input_tokens_seen": 122566,
+      "step": 37,
+      "train_runtime": 263.411,
+      "train_tokens_per_second": 465.303
+    },
+    {
+      "epoch": 0.3064516129032258,
+      "grad_norm": 0.6758707165718079,
+      "learning_rate": 0.00019968923353071377,
+      "loss": 0.6775951981544495,
+      "num_input_tokens_seen": 125906,
+      "step": 38,
+      "train_runtime": 268.6516,
+      "train_tokens_per_second": 468.659
+    },
+    {
+      "epoch": 0.31451612903225806,
+      "grad_norm": 0.7149205207824707,
+      "learning_rate": 0.00019966887195344403,
+      "loss": 0.6415547132492065,
+      "num_input_tokens_seen": 129178,
+      "step": 39,
+      "train_runtime": 273.5735,
+      "train_tokens_per_second": 472.188
+    },
+    {
+      "epoch": 0.3225806451612903,
+      "grad_norm": 0.8233773708343506,
+      "learning_rate": 0.0001996478654275229,
+      "loss": 0.6871565580368042,
+      "num_input_tokens_seen": 132242,
+      "step": 40,
+      "train_runtime": 278.1257,
+      "train_tokens_per_second": 475.476
+    },
+    {
+      "epoch": 0.33064516129032256,
+      "grad_norm": 0.603967010974884,
+      "learning_rate": 0.00019962621408888177,
+      "loss": 0.6418657302856445,
+      "num_input_tokens_seen": 135598,
+      "step": 41,
+      "train_runtime": 283.0503,
+      "train_tokens_per_second": 479.06
+    },
+    {
+      "epoch": 0.3387096774193548,
+      "grad_norm": 0.6170377731323242,
+      "learning_rate": 0.00019960391807762463,
+      "loss": 0.6446406245231628,
+      "num_input_tokens_seen": 139036,
+      "step": 42,
+      "train_runtime": 287.8831,
+      "train_tokens_per_second": 482.96
+    },
+    {
+      "epoch": 0.3467741935483871,
+      "grad_norm": 0.6677550077438354,
+      "learning_rate": 0.00019958097753802693,
+      "loss": 0.6513428092002869,
+      "num_input_tokens_seen": 142314,
+      "step": 43,
+      "train_runtime": 292.6153,
+      "train_tokens_per_second": 486.352
+    },
+    {
+      "epoch": 0.3548387096774194,
+      "grad_norm": 0.5713569521903992,
+      "learning_rate": 0.00019955739261853504,
+      "loss": 0.6523911952972412,
+      "num_input_tokens_seen": 145730,
+      "step": 44,
+      "train_runtime": 297.3525,
+      "train_tokens_per_second": 490.092
+    },
+    {
+      "epoch": 0.3629032258064516,
+      "grad_norm": 0.6360762119293213,
+      "learning_rate": 0.00019953316347176488,
+      "loss": 0.6005333662033081,
+      "num_input_tokens_seen": 148950,
+      "step": 45,
+      "train_runtime": 301.7634,
+      "train_tokens_per_second": 493.599
+    },
+    {
+      "epoch": 0.3709677419354839,
+      "grad_norm": 0.5776824355125427,
+      "learning_rate": 0.00019950829025450114,
+      "loss": 0.6482651233673096,
+      "num_input_tokens_seen": 152208,
+      "step": 46,
+      "train_runtime": 306.4442,
+      "train_tokens_per_second": 496.691
+    },
+    {
+      "epoch": 0.3790322580645161,
+      "grad_norm": 0.5635464787483215,
+      "learning_rate": 0.0001994827731276963,
+      "loss": 0.5292215943336487,
+      "num_input_tokens_seen": 155494,
+      "step": 47,
+      "train_runtime": 310.97,
+      "train_tokens_per_second": 500.029
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 0.7464826107025146,
+      "learning_rate": 0.00019945661225646946,
+      "loss": 0.5896004438400269,
+      "num_input_tokens_seen": 158822,
+      "step": 48,
+      "train_runtime": 315.5591,
+      "train_tokens_per_second": 503.304
+    },
+    {
+      "epoch": 0.3951612903225806,
+      "grad_norm": 0.6741185784339905,
+      "learning_rate": 0.0001994298078101054,
+      "loss": 0.610090970993042,
+      "num_input_tokens_seen": 162234,
+      "step": 49,
+      "train_runtime": 320.3691,
+      "train_tokens_per_second": 506.397
+    },
+    {
+      "epoch": 0.4032258064516129,
+      "grad_norm": 0.8294225931167603,
+      "learning_rate": 0.00019940235996205333,
+      "loss": 0.7023945450782776,
+      "num_input_tokens_seen": 165674,
+      "step": 50,
+      "train_runtime": 325.0853,
+      "train_tokens_per_second": 509.632
+    },
+    {
+      "epoch": 0.4112903225806452,
+      "grad_norm": 0.6233593821525574,
+      "learning_rate": 0.0001993742688899259,
+      "loss": 0.5964958667755127,
+      "num_input_tokens_seen": 169012,
+      "step": 51,
+      "train_runtime": 329.9005,
+      "train_tokens_per_second": 512.312
+    },
+    {
+      "epoch": 0.41935483870967744,
+      "grad_norm": 0.6552059054374695,
+      "learning_rate": 0.00019934553477549794,
+      "loss": 0.5970840454101562,
+      "num_input_tokens_seen": 172396,
+      "step": 52,
+      "train_runtime": 334.6436,
+      "train_tokens_per_second": 515.163
+    },
+    {
+      "epoch": 0.4274193548387097,
+      "grad_norm": 0.5657429695129395,
+      "learning_rate": 0.00019931615780470558,
+      "loss": 0.6213586330413818,
+      "num_input_tokens_seen": 175766,
+      "step": 53,
+      "train_runtime": 339.4928,
+      "train_tokens_per_second": 517.731
+    },
+    {
+      "epoch": 0.43548387096774194,
+      "grad_norm": 0.6272680163383484,
+      "learning_rate": 0.00019928613816764458,
+      "loss": 0.5776374340057373,
+      "num_input_tokens_seen": 179034,
+      "step": 54,
+      "train_runtime": 344.3701,
+      "train_tokens_per_second": 519.888
+    },
+    {
+      "epoch": 0.4435483870967742,
+      "grad_norm": 0.5951282382011414,
+      "learning_rate": 0.00019925547605856934,
+      "loss": 0.6662795543670654,
+      "num_input_tokens_seen": 182380,
+      "step": 55,
+      "train_runtime": 349.2001,
+      "train_tokens_per_second": 522.279
+    },
+    {
+      "epoch": 0.45161290322580644,
+      "grad_norm": 0.6823417544364929,
+      "learning_rate": 0.00019922417167589183,
+      "loss": 0.6379531621932983,
+      "num_input_tokens_seen": 185574,
+      "step": 56,
+      "train_runtime": 353.9778,
+      "train_tokens_per_second": 524.253
+    },
+    {
+      "epoch": 0.4596774193548387,
+      "grad_norm": 0.5668640732765198,
+      "learning_rate": 0.00019919222522217996,
+      "loss": 0.6385904550552368,
+      "num_input_tokens_seen": 188970,
+      "step": 57,
+      "train_runtime": 358.9666,
+      "train_tokens_per_second": 526.428
+    },
+    {
+      "epoch": 0.46774193548387094,
+      "grad_norm": 0.5809326171875,
+      "learning_rate": 0.00019915963690415647,
+      "loss": 0.5536263585090637,
+      "num_input_tokens_seen": 192282,
+      "step": 58,
+      "train_runtime": 363.7121,
+      "train_tokens_per_second": 528.665
+    },
+    {
+      "epoch": 0.47580645161290325,
+      "grad_norm": 0.6727699041366577,
+      "learning_rate": 0.00019912640693269752,
+      "loss": 0.6002112627029419,
+      "num_input_tokens_seen": 195644,
+      "step": 59,
+      "train_runtime": 368.6469,
+      "train_tokens_per_second": 530.708
+    },
+    {
+      "epoch": 0.4838709677419355,
+      "grad_norm": 0.6097072958946228,
+      "learning_rate": 0.00019909253552283143,
+      "loss": 0.611109733581543,
+      "num_input_tokens_seen": 198956,
+      "step": 60,
+      "train_runtime": 373.378,
+      "train_tokens_per_second": 532.854
+    },
+    {
+      "epoch": 0.4838709677419355,
+      "eval_loss": 3.0930163860321045,
+      "eval_runtime": 12.3352,
+      "eval_samples_per_second": 4.216,
+      "eval_steps_per_second": 2.108,
+      "num_input_tokens_seen": 198956,
+      "step": 60
+    },
+    {
+      "epoch": 0.49193548387096775,
+      "grad_norm": 0.608394205570221,
+      "learning_rate": 0.00019905802289373715,
+      "loss": 0.5526689887046814,
+      "num_input_tokens_seen": 202314,
+      "step": 61,
+      "train_runtime": 405.9522,
+      "train_tokens_per_second": 498.369
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.6111866235733032,
+      "learning_rate": 0.0001990228692687429,
+      "loss": 0.5586686730384827,
+      "num_input_tokens_seen": 205722,
+      "step": 62,
+      "train_runtime": 410.5864,
+      "train_tokens_per_second": 501.044
+    },
+    {
+      "epoch": 0.5080645161290323,
+      "grad_norm": 0.6161165237426758,
+      "learning_rate": 0.00019898707487532474,
+      "loss": 0.641068160533905,
+      "num_input_tokens_seen": 208882,
+      "step": 63,
+      "train_runtime": 415.1353,
+      "train_tokens_per_second": 503.166
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 0.6087555289268494,
+      "learning_rate": 0.0001989506399451051,
+      "loss": 0.5811461210250854,
+      "num_input_tokens_seen": 212198,
+      "step": 64,
+      "train_runtime": 420.1346,
+      "train_tokens_per_second": 505.071
+    },
+    {
+      "epoch": 0.5241935483870968,
+      "grad_norm": 0.6409975290298462,
+      "learning_rate": 0.0001989135647138513,
+      "loss": 0.6092388033866882,
+      "num_input_tokens_seen": 215586,
+      "step": 65,
+      "train_runtime": 425.1826,
+      "train_tokens_per_second": 507.043
+    },
+    {
+      "epoch": 0.532258064516129,
+      "grad_norm": 0.7127669453620911,
+      "learning_rate": 0.00019887584942147394,
+      "loss": 0.6781883239746094,
+      "num_input_tokens_seen": 218732,
+      "step": 66,
+      "train_runtime": 430.1434,
+      "train_tokens_per_second": 508.509
+    },
+    {
+      "epoch": 0.5403225806451613,
+      "grad_norm": 0.6341464519500732,
+      "learning_rate": 0.0001988374943120254,
+      "loss": 0.6367583274841309,
+      "num_input_tokens_seen": 221938,
+      "step": 67,
+      "train_runtime": 435.098,
+      "train_tokens_per_second": 510.087
+    },
+    {
+      "epoch": 0.5483870967741935,
+      "grad_norm": 0.6676456332206726,
+      "learning_rate": 0.00019879849963369827,
+      "loss": 0.5986863970756531,
+      "num_input_tokens_seen": 225362,
+      "step": 68,
+      "train_runtime": 440.4076,
+      "train_tokens_per_second": 511.712
+    },
+    {
+      "epoch": 0.5564516129032258,
+      "grad_norm": 0.5387278199195862,
+      "learning_rate": 0.00019875886563882375,
+      "loss": 0.5684967637062073,
+      "num_input_tokens_seen": 228656,
+      "step": 69,
+      "train_runtime": 445.6788,
+      "train_tokens_per_second": 513.051
+    },
+    {
+      "epoch": 0.5645161290322581,
+      "grad_norm": 0.5598258972167969,
+      "learning_rate": 0.00019871859258387,
+      "loss": 0.5872923135757446,
+      "num_input_tokens_seen": 232052,
+      "step": 70,
+      "train_runtime": 450.5989,
+      "train_tokens_per_second": 514.986
+    },
+    {
+      "epoch": 0.5725806451612904,
+      "grad_norm": 0.5795265436172485,
+      "learning_rate": 0.00019867768072944045,
+      "loss": 0.5454657673835754,
+      "num_input_tokens_seen": 235430,
+      "step": 71,
+      "train_runtime": 455.462,
+      "train_tokens_per_second": 516.904
+    },
+    {
+      "epoch": 0.5806451612903226,
+      "grad_norm": 0.6275506019592285,
+      "learning_rate": 0.00019863613034027224,
+      "loss": 0.5985739827156067,
+      "num_input_tokens_seen": 238792,
+      "step": 72,
+      "train_runtime": 460.1637,
+      "train_tokens_per_second": 518.928
+    },
+    {
+      "epoch": 0.5887096774193549,
+      "grad_norm": 0.6417560577392578,
+      "learning_rate": 0.0001985939416852343,
+      "loss": 0.6151498556137085,
+      "num_input_tokens_seen": 242136,
+      "step": 73,
+      "train_runtime": 464.8752,
+      "train_tokens_per_second": 520.862
+    },
+    {
+      "epoch": 0.5967741935483871,
+      "grad_norm": 0.6558713912963867,
+      "learning_rate": 0.00019855111503732574,
+      "loss": 0.6127005219459534,
+      "num_input_tokens_seen": 245502,
+      "step": 74,
+      "train_runtime": 469.6002,
+      "train_tokens_per_second": 522.789
+    },
+    {
+      "epoch": 0.6048387096774194,
+      "grad_norm": 0.5617387294769287,
+      "learning_rate": 0.00019850765067367412,
+      "loss": 0.5805050730705261,
+      "num_input_tokens_seen": 248724,
+      "step": 75,
+      "train_runtime": 473.9894,
+      "train_tokens_per_second": 524.746
+    },
+    {
+      "epoch": 0.6129032258064516,
+      "grad_norm": 0.5580056309700012,
+      "learning_rate": 0.00019846354887553358,
+      "loss": 0.5755968689918518,
+      "num_input_tokens_seen": 252068,
+      "step": 76,
+      "train_runtime": 478.6533,
+      "train_tokens_per_second": 526.619
+    },
+    {
+      "epoch": 0.6209677419354839,
+      "grad_norm": 0.5987703204154968,
+      "learning_rate": 0.00019841880992828306,
+      "loss": 0.5941160321235657,
+      "num_input_tokens_seen": 255328,
+      "step": 77,
+      "train_runtime": 483.1368,
+      "train_tokens_per_second": 528.48
+    },
+    {
+      "epoch": 0.6290322580645161,
+      "grad_norm": 0.5982978940010071,
+      "learning_rate": 0.0001983734341214244,
+      "loss": 0.5437498688697815,
+      "num_input_tokens_seen": 258440,
+      "step": 78,
+      "train_runtime": 487.3985,
+      "train_tokens_per_second": 530.244
+    },
+    {
+      "epoch": 0.6370967741935484,
+      "grad_norm": 0.6150534749031067,
+      "learning_rate": 0.00019832742174858052,
+      "loss": 0.4968106150627136,
+      "num_input_tokens_seen": 261670,
+      "step": 79,
+      "train_runtime": 492.0778,
+      "train_tokens_per_second": 531.765
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 0.6039842367172241,
+      "learning_rate": 0.0001982807731074935,
+      "loss": 0.5774258375167847,
+      "num_input_tokens_seen": 265042,
+      "step": 80,
+      "train_runtime": 496.8154,
+      "train_tokens_per_second": 533.482
+    },
+    {
+      "epoch": 0.6532258064516129,
+      "grad_norm": 0.6645222902297974,
+      "learning_rate": 0.00019823348850002268,
+      "loss": 0.5437123775482178,
+      "num_input_tokens_seen": 268396,
+      "step": 81,
+      "train_runtime": 501.4894,
+      "train_tokens_per_second": 535.198
+    },
+    {
+      "epoch": 0.6612903225806451,
+      "grad_norm": 0.6312376856803894,
+      "learning_rate": 0.00019818556823214268,
+      "loss": 0.5945921540260315,
+      "num_input_tokens_seen": 271554,
+      "step": 82,
+      "train_runtime": 506.2627,
+      "train_tokens_per_second": 536.389
+    },
+    {
+      "epoch": 0.6693548387096774,
+      "grad_norm": 0.7044872641563416,
+      "learning_rate": 0.00019813701261394136,
+      "loss": 0.6143019199371338,
+      "num_input_tokens_seen": 274730,
+      "step": 83,
+      "train_runtime": 510.8028,
+      "train_tokens_per_second": 537.84
+    },
+    {
+      "epoch": 0.6774193548387096,
+      "grad_norm": 0.5553087592124939,
+      "learning_rate": 0.00019808782195961797,
+      "loss": 0.5542465448379517,
+      "num_input_tokens_seen": 278100,
+      "step": 84,
+      "train_runtime": 515.6576,
+      "train_tokens_per_second": 539.311
+    },
+    {
+      "epoch": 0.6854838709677419,
+      "grad_norm": 0.561292290687561,
+      "learning_rate": 0.00019803799658748094,
+      "loss": 0.5175400376319885,
+      "num_input_tokens_seen": 281468,
+      "step": 85,
+      "train_runtime": 520.6317,
+      "train_tokens_per_second": 540.628
+    },
+    {
+      "epoch": 0.6935483870967742,
+      "grad_norm": 0.5422635078430176,
+      "learning_rate": 0.000197987536819946,
+      "loss": 0.6259896755218506,
+      "num_input_tokens_seen": 284944,
+      "step": 86,
+      "train_runtime": 525.5795,
+      "train_tokens_per_second": 542.152
+    },
+    {
+      "epoch": 0.7016129032258065,
+      "grad_norm": 0.555388867855072,
+      "learning_rate": 0.0001979364429835339,
+      "loss": 0.5073646306991577,
+      "num_input_tokens_seen": 288252,
+      "step": 87,
+      "train_runtime": 530.5048,
+      "train_tokens_per_second": 543.354
+    },
+    {
+      "epoch": 0.7096774193548387,
+      "grad_norm": 0.5314076542854309,
+      "learning_rate": 0.00019788471540886844,
+      "loss": 0.5061646103858948,
+      "num_input_tokens_seen": 291674,
+      "step": 88,
+      "train_runtime": 535.3554,
+      "train_tokens_per_second": 544.823
+    },
+    {
+      "epoch": 0.717741935483871,
+      "grad_norm": 0.531013548374176,
+      "learning_rate": 0.0001978323544306743,
+      "loss": 0.5036947727203369,
+      "num_input_tokens_seen": 295054,
+      "step": 89,
+      "train_runtime": 540.1647,
+      "train_tokens_per_second": 546.23
+    },
+    {
+      "epoch": 0.7258064516129032,
+      "grad_norm": 0.7979210615158081,
+      "learning_rate": 0.00019777936038777483,
+      "loss": 0.613807201385498,
+      "num_input_tokens_seen": 298162,
+      "step": 90,
+      "train_runtime": 544.7969,
+      "train_tokens_per_second": 547.29
+    },
+    {
+      "epoch": 0.7258064516129032,
+      "eval_loss": 2.9071261882781982,
+      "eval_runtime": 12.2185,
+      "eval_samples_per_second": 4.256,
+      "eval_steps_per_second": 2.128,
+      "num_input_tokens_seen": 298162,
+      "step": 90
+    },
+    {
+      "epoch": 0.7338709677419355,
+      "grad_norm": 0.5745816826820374,
+      "learning_rate": 0.0001977257336230899,
+      "loss": 0.48962247371673584,
+      "num_input_tokens_seen": 301500,
+      "step": 91,
+      "train_runtime": 597.652,
+      "train_tokens_per_second": 504.474
+    },
+    {
+      "epoch": 0.7419354838709677,
+      "grad_norm": 0.7026141285896301,
+      "learning_rate": 0.00019767147448363366,
+      "loss": 0.5600932836532593,
+      "num_input_tokens_seen": 304624,
+      "step": 92,
+      "train_runtime": 602.1005,
+      "train_tokens_per_second": 505.936
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.6634042859077454,
+      "learning_rate": 0.00019761658332051235,
+      "loss": 0.5759444236755371,
+      "num_input_tokens_seen": 307956,
+      "step": 93,
+      "train_runtime": 606.9852,
+      "train_tokens_per_second": 507.353
+    },
+    {
+      "epoch": 0.7580645161290323,
+      "grad_norm": 0.5837485790252686,
+      "learning_rate": 0.00019756106048892186,
+      "loss": 0.5830598473548889,
+      "num_input_tokens_seen": 311354,
+      "step": 94,
+      "train_runtime": 611.9247,
+      "train_tokens_per_second": 508.811
+    },
+    {
+      "epoch": 0.7661290322580645,
+      "grad_norm": 0.581575334072113,
+      "learning_rate": 0.00019750490634814572,
+      "loss": 0.5618096590042114,
+      "num_input_tokens_seen": 314592,
+      "step": 95,
+      "train_runtime": 616.8898,
+      "train_tokens_per_second": 509.965
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 0.5666610598564148,
+      "learning_rate": 0.00019744812126155245,
+      "loss": 0.5454350709915161,
+      "num_input_tokens_seen": 317960,
+      "step": 96,
+      "train_runtime": 622.062,
+      "train_tokens_per_second": 511.139
+    },
+    {
+      "epoch": 0.782258064516129,
+      "grad_norm": 0.5417006611824036,
+      "learning_rate": 0.00019739070559659347,
+      "loss": 0.561352014541626,
+      "num_input_tokens_seen": 321274,
+      "step": 97,
+      "train_runtime": 627.2236,
+      "train_tokens_per_second": 512.216
+    },
+    {
+      "epoch": 0.7903225806451613,
+      "grad_norm": 0.5439388751983643,
+      "learning_rate": 0.0001973326597248006,
+      "loss": 0.5779426097869873,
+      "num_input_tokens_seen": 324480,
+      "step": 98,
+      "train_runtime": 632.176,
+      "train_tokens_per_second": 513.275
+    },
+    {
+      "epoch": 0.7983870967741935,
+      "grad_norm": 0.5616183280944824,
+      "learning_rate": 0.0001972739840217836,
+      "loss": 0.49771565198898315,
+      "num_input_tokens_seen": 327678,
+      "step": 99,
+      "train_runtime": 636.8674,
+      "train_tokens_per_second": 514.515
+    },
+    {
+      "epoch": 0.8064516129032258,
+      "grad_norm": 0.8428842425346375,
+      "learning_rate": 0.00019721467886722792,
+      "loss": 0.5603002309799194,
+      "num_input_tokens_seen": 330914,
+      "step": 100,
+      "train_runtime": 641.6893,
+      "train_tokens_per_second": 515.692
+    },
+    {
+      "epoch": 0.8145161290322581,
+      "grad_norm": 0.6361693143844604,
+      "learning_rate": 0.00019715474464489208,
+      "loss": 0.5690774321556091,
+      "num_input_tokens_seen": 334256,
+      "step": 101,
+      "train_runtime": 646.5226,
+      "train_tokens_per_second": 517.006
+    },
+    {
+      "epoch": 0.8225806451612904,
+      "grad_norm": 0.5664668083190918,
+      "learning_rate": 0.0001970941817426052,
+      "loss": 0.5249347686767578,
+      "num_input_tokens_seen": 337722,
+      "step": 102,
+      "train_runtime": 651.2588,
+      "train_tokens_per_second": 518.568
+    },
+    {
+      "epoch": 0.8306451612903226,
+      "grad_norm": 0.564401388168335,
+      "learning_rate": 0.00019703299055226468,
+      "loss": 0.5265264511108398,
+      "num_input_tokens_seen": 341038,
+      "step": 103,
+      "train_runtime": 656.0275,
+      "train_tokens_per_second": 519.853
+    },
+    {
+      "epoch": 0.8387096774193549,
+      "grad_norm": 0.7196506857872009,
+      "learning_rate": 0.00019697117146983334,
+      "loss": 0.5534394979476929,
+      "num_input_tokens_seen": 344066,
+      "step": 104,
+      "train_runtime": 660.2072,
+      "train_tokens_per_second": 521.148
+    },
+    {
+      "epoch": 0.8467741935483871,
+      "grad_norm": 0.579501211643219,
+      "learning_rate": 0.0001969087248953371,
+      "loss": 0.5940977334976196,
+      "num_input_tokens_seen": 347352,
+      "step": 105,
+      "train_runtime": 664.7483,
+      "train_tokens_per_second": 522.532
+    },
+    {
+      "epoch": 0.8548387096774194,
+      "grad_norm": 0.5122610330581665,
+      "learning_rate": 0.00019684565123286244,
+      "loss": 0.5310789942741394,
+      "num_input_tokens_seen": 350654,
+      "step": 106,
+      "train_runtime": 669.4554,
+      "train_tokens_per_second": 523.79
+    },
+    {
+      "epoch": 0.8629032258064516,
+      "grad_norm": 0.5480422973632812,
+      "learning_rate": 0.00019678195089055346,
+      "loss": 0.5730823874473572,
+      "num_input_tokens_seen": 353954,
+      "step": 107,
+      "train_runtime": 673.9271,
+      "train_tokens_per_second": 525.211
+    },
+    {
+      "epoch": 0.8709677419354839,
+      "grad_norm": 0.5518767237663269,
+      "learning_rate": 0.00019671762428060966,
+      "loss": 0.553552508354187,
+      "num_input_tokens_seen": 357348,
+      "step": 108,
+      "train_runtime": 678.7199,
+      "train_tokens_per_second": 526.503
+    },
+    {
+      "epoch": 0.8790322580645161,
+      "grad_norm": 0.5668540596961975,
+      "learning_rate": 0.00019665267181928292,
+      "loss": 0.5294761657714844,
+      "num_input_tokens_seen": 360628,
+      "step": 109,
+      "train_runtime": 683.35,
+      "train_tokens_per_second": 527.735
+    },
+    {
+      "epoch": 0.8870967741935484,
+      "grad_norm": 0.4758228659629822,
+      "learning_rate": 0.00019658709392687506,
+      "loss": 0.5213898420333862,
+      "num_input_tokens_seen": 364030,
+      "step": 110,
+      "train_runtime": 688.0325,
+      "train_tokens_per_second": 529.088
+    },
+    {
+      "epoch": 0.8951612903225806,
+      "grad_norm": 0.604768693447113,
+      "learning_rate": 0.00019652089102773488,
+      "loss": 0.6025131940841675,
+      "num_input_tokens_seen": 367360,
+      "step": 111,
+      "train_runtime": 692.8246,
+      "train_tokens_per_second": 530.235
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 0.5635347366333008,
+      "learning_rate": 0.00019645406355025565,
+      "loss": 0.5061513185501099,
+      "num_input_tokens_seen": 370518,
+      "step": 112,
+      "train_runtime": 697.3954,
+      "train_tokens_per_second": 531.288
+    },
+    {
+      "epoch": 0.9112903225806451,
+      "grad_norm": 0.5575589537620544,
+      "learning_rate": 0.00019638661192687216,
+      "loss": 0.5043578147888184,
+      "num_input_tokens_seen": 373828,
+      "step": 113,
+      "train_runtime": 702.2518,
+      "train_tokens_per_second": 532.328
+    },
+    {
+      "epoch": 0.9193548387096774,
+      "grad_norm": 0.5947410464286804,
+      "learning_rate": 0.00019631853659405807,
+      "loss": 0.5288162231445312,
+      "num_input_tokens_seen": 377226,
+      "step": 114,
+      "train_runtime": 707.3502,
+      "train_tokens_per_second": 533.295
+    },
+    {
+      "epoch": 0.9274193548387096,
+      "grad_norm": 0.5738222599029541,
+      "learning_rate": 0.000196249837992323,
+      "loss": 0.5216498374938965,
+      "num_input_tokens_seen": 380624,
+      "step": 115,
+      "train_runtime": 712.258,
+      "train_tokens_per_second": 534.391
+    },
+    {
+      "epoch": 0.9354838709677419,
+      "grad_norm": 0.5514879822731018,
+      "learning_rate": 0.0001961805165662096,
+      "loss": 0.47155335545539856,
+      "num_input_tokens_seen": 384012,
+      "step": 116,
+      "train_runtime": 717.2598,
+      "train_tokens_per_second": 535.388
+    },
+    {
+      "epoch": 0.9435483870967742,
+      "grad_norm": 0.6073253154754639,
+      "learning_rate": 0.00019611057276429085,
+      "loss": 0.5124592781066895,
+      "num_input_tokens_seen": 387364,
+      "step": 117,
+      "train_runtime": 722.0631,
+      "train_tokens_per_second": 536.468
+    },
+    {
+      "epoch": 0.9516129032258065,
+      "grad_norm": 0.5801448822021484,
+      "learning_rate": 0.00019604000703916705,
+      "loss": 0.5122741460800171,
+      "num_input_tokens_seen": 390740,
+      "step": 118,
+      "train_runtime": 726.8929,
+      "train_tokens_per_second": 537.548
+    },
+    {
+      "epoch": 0.9596774193548387,
+      "grad_norm": 0.5871880650520325,
+      "learning_rate": 0.00019596881984746287,
+      "loss": 0.5125201940536499,
+      "num_input_tokens_seen": 394002,
+      "step": 119,
+      "train_runtime": 731.6556,
+      "train_tokens_per_second": 538.507
+    },
+    {
+      "epoch": 0.967741935483871,
+      "grad_norm": 0.6655816435813904,
+      "learning_rate": 0.00019589701164982452,
+      "loss": 0.524815022945404,
+      "num_input_tokens_seen": 397036,
+      "step": 120,
+      "train_runtime": 735.9727,
+      "train_tokens_per_second": 539.471
+    },
+    {
+      "epoch": 0.967741935483871,
+      "eval_loss": 2.589766025543213,
+      "eval_runtime": 12.1494,
+      "eval_samples_per_second": 4.28,
+      "eval_steps_per_second": 2.14,
+      "num_input_tokens_seen": 397036,
+      "step": 120
+    },
+    {
+      "epoch": 0.9758064516129032,
+      "grad_norm": 0.6723934412002563,
+      "learning_rate": 0.00019582458291091663,
+      "loss": 0.611114501953125,
+      "num_input_tokens_seen": 400216,
+      "step": 121,
+      "train_runtime": 772.7017,
+      "train_tokens_per_second": 517.944
+    },
+    {
+      "epoch": 0.9838709677419355,
+      "grad_norm": 0.5538724064826965,
+      "learning_rate": 0.0001957515340994193,
+      "loss": 0.5213549137115479,
+      "num_input_tokens_seen": 403522,
+      "step": 122,
+      "train_runtime": 777.3527,
+      "train_tokens_per_second": 519.098
+    },
+    {
+      "epoch": 0.9919354838709677,
+      "grad_norm": 0.5994356274604797,
+      "learning_rate": 0.000195677865688025,
+      "loss": 0.5735296607017517,
+      "num_input_tokens_seen": 406874,
+      "step": 123,
+      "train_runtime": 782.2491,
+      "train_tokens_per_second": 520.134
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.6571742296218872,
+      "learning_rate": 0.00019560357815343577,
+      "loss": 0.5385057926177979,
+      "num_input_tokens_seen": 410114,
+      "step": 124,
+      "train_runtime": 787.0107,
+      "train_tokens_per_second": 521.103
+    },
+    {
+      "epoch": 1.0080645161290323,
+      "grad_norm": 0.5309604406356812,
+      "learning_rate": 0.00019552867197635974,
+      "loss": 0.3922950029373169,
+      "num_input_tokens_seen": 413264,
+      "step": 125,
+      "train_runtime": 791.8414,
+      "train_tokens_per_second": 521.902
+    },
+    {
+      "epoch": 1.0161290322580645,
+      "grad_norm": 0.4756343960762024,
+      "learning_rate": 0.00019545314764150837,
+      "loss": 0.3772149085998535,
+      "num_input_tokens_seen": 416432,
+      "step": 126,
+      "train_runtime": 796.7161,
+      "train_tokens_per_second": 522.686
+    },
+    {
+      "epoch": 1.0241935483870968,
+      "grad_norm": 0.5248489379882812,
+      "learning_rate": 0.00019537700563759304,
+      "loss": 0.4322376251220703,
+      "num_input_tokens_seen": 419814,
+      "step": 127,
+      "train_runtime": 801.9801,
+      "train_tokens_per_second": 523.472
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 0.5486618280410767,
+      "learning_rate": 0.00019530024645732206,
+      "loss": 0.42461395263671875,
+      "num_input_tokens_seen": 423168,
+      "step": 128,
+      "train_runtime": 807.1128,
+      "train_tokens_per_second": 524.298
+    },
+    {
+      "epoch": 1.0403225806451613,
+      "grad_norm": 0.6822044253349304,
+      "learning_rate": 0.00019522287059739753,
+      "loss": 0.44680896401405334,
+      "num_input_tokens_seen": 426248,
+      "step": 129,
+      "train_runtime": 811.7053,
+      "train_tokens_per_second": 525.127
+    },
+    {
+      "epoch": 1.0483870967741935,
+      "grad_norm": 0.6571690440177917,
+      "learning_rate": 0.00019514487855851184,
+      "loss": 0.45198315382003784,
+      "num_input_tokens_seen": 429726,
+      "step": 130,
+      "train_runtime": 816.7684,
+      "train_tokens_per_second": 526.13
+    },
+    {
+      "epoch": 1.0564516129032258,
+      "grad_norm": 0.6491777896881104,
+      "learning_rate": 0.00019506627084534483,
+      "loss": 0.4377821683883667,
+      "num_input_tokens_seen": 432912,
+      "step": 131,
+      "train_runtime": 821.4589,
+      "train_tokens_per_second": 527.004
+    },
+    {
+      "epoch": 1.064516129032258,
+      "grad_norm": 0.6005299091339111,
+      "learning_rate": 0.00019498704796656018,
+      "loss": 0.45241302251815796,
+      "num_input_tokens_seen": 436246,
+      "step": 132,
+      "train_runtime": 826.1551,
+      "train_tokens_per_second": 528.044
+    },
+    {
+      "epoch": 1.0725806451612903,
+      "grad_norm": 0.5644078850746155,
+      "learning_rate": 0.00019490721043480226,
+      "loss": 0.4442741274833679,
+      "num_input_tokens_seen": 439446,
+      "step": 133,
+      "train_runtime": 830.7581,
+      "train_tokens_per_second": 528.97
+    },
+    {
+      "epoch": 1.0806451612903225,
+      "grad_norm": 0.5728560090065002,
+      "learning_rate": 0.00019482675876669286,
+      "loss": 0.4184889793395996,
+      "num_input_tokens_seen": 442716,
+      "step": 134,
+      "train_runtime": 835.3283,
+      "train_tokens_per_second": 529.99
+    },
+    {
+      "epoch": 1.0887096774193548,
+      "grad_norm": 0.5105488896369934,
+      "learning_rate": 0.00019474569348282774,
+      "loss": 0.4137258529663086,
+      "num_input_tokens_seen": 445948,
+      "step": 135,
+      "train_runtime": 839.8381,
+      "train_tokens_per_second": 530.993
+    },
+    {
+      "epoch": 1.096774193548387,
+      "grad_norm": 0.6140270829200745,
+      "learning_rate": 0.0001946640151077734,
+      "loss": 0.4656698703765869,
+      "num_input_tokens_seen": 449172,
+      "step": 136,
+      "train_runtime": 844.4082,
+      "train_tokens_per_second": 531.937
+    },
+    {
+      "epoch": 1.1048387096774193,
+      "grad_norm": 0.6442403793334961,
+      "learning_rate": 0.00019458172417006347,
+      "loss": 0.519420325756073,
+      "num_input_tokens_seen": 452512,
+      "step": 137,
+      "train_runtime": 848.9974,
+      "train_tokens_per_second": 532.996
+    },
+    {
+      "epoch": 1.1129032258064515,
+      "grad_norm": 0.5361612439155579,
+      "learning_rate": 0.00019449882120219555,
+      "loss": 0.4331057667732239,
+      "num_input_tokens_seen": 455922,
+      "step": 138,
+      "train_runtime": 853.7922,
+      "train_tokens_per_second": 533.996
+    },
+    {
+      "epoch": 1.120967741935484,
+      "grad_norm": 0.56069415807724,
+      "learning_rate": 0.00019441530674062753,
+      "loss": 0.43306776881217957,
+      "num_input_tokens_seen": 459246,
+      "step": 139,
+      "train_runtime": 858.4641,
+      "train_tokens_per_second": 534.962
+    },
+    {
+      "epoch": 1.129032258064516,
+      "grad_norm": 0.5961108803749084,
+      "learning_rate": 0.0001943311813257743,
+      "loss": 0.38593000173568726,
+      "num_input_tokens_seen": 462644,
+      "step": 140,
+      "train_runtime": 863.174,
+      "train_tokens_per_second": 535.98
+    },
+    {
+      "epoch": 1.1370967741935485,
+      "grad_norm": 0.5356931090354919,
+      "learning_rate": 0.00019424644550200415,
+      "loss": 0.45368602871894836,
+      "num_input_tokens_seen": 466004,
+      "step": 141,
+      "train_runtime": 867.9995,
+      "train_tokens_per_second": 536.871
+    },
+    {
+      "epoch": 1.1451612903225807,
+      "grad_norm": 0.6337073445320129,
+      "learning_rate": 0.00019416109981763526,
+      "loss": 0.4836081266403198,
+      "num_input_tokens_seen": 469474,
+      "step": 142,
+      "train_runtime": 872.8737,
+      "train_tokens_per_second": 537.849
+    },
+    {
+      "epoch": 1.153225806451613,
+      "grad_norm": 0.5758848786354065,
+      "learning_rate": 0.00019407514482493214,
+      "loss": 0.39848586916923523,
+      "num_input_tokens_seen": 472830,
+      "step": 143,
+      "train_runtime": 877.6996,
+      "train_tokens_per_second": 538.715
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 0.580061674118042,
+      "learning_rate": 0.00019398858108010217,
+      "loss": 0.4755861461162567,
+      "num_input_tokens_seen": 476146,
+      "step": 144,
+      "train_runtime": 882.5729,
+      "train_tokens_per_second": 539.498
+    },
+    {
+      "epoch": 1.1693548387096775,
+      "grad_norm": 0.6298698782920837,
+      "learning_rate": 0.0001939014091432918,
+      "loss": 0.48262229561805725,
+      "num_input_tokens_seen": 479388,
+      "step": 145,
+      "train_runtime": 887.2589,
+      "train_tokens_per_second": 540.302
+    },
+    {
+      "epoch": 1.1774193548387097,
+      "grad_norm": 0.5250868201255798,
+      "learning_rate": 0.00019381362957858312,
+      "loss": 0.46615517139434814,
+      "num_input_tokens_seen": 482820,
+      "step": 146,
+      "train_runtime": 892.2828,
+      "train_tokens_per_second": 541.106
+    },
+    {
+      "epoch": 1.185483870967742,
+      "grad_norm": 0.5974312424659729,
+      "learning_rate": 0.00019372524295399013,
+      "loss": 0.4052748680114746,
+      "num_input_tokens_seen": 485956,
+      "step": 147,
+      "train_runtime": 896.8248,
+      "train_tokens_per_second": 541.863
+    },
+    {
+      "epoch": 1.1935483870967742,
+      "grad_norm": 0.4935814440250397,
+      "learning_rate": 0.00019363624984145502,
+      "loss": 0.4510989189147949,
+      "num_input_tokens_seen": 489266,
+      "step": 148,
+      "train_runtime": 901.6055,
+      "train_tokens_per_second": 542.661
+    },
+    {
+      "epoch": 1.2016129032258065,
+      "grad_norm": 0.5875913500785828,
+      "learning_rate": 0.00019354665081684446,
+      "loss": 0.39551085233688354,
+      "num_input_tokens_seen": 492712,
+      "step": 149,
+      "train_runtime": 906.6105,
+      "train_tokens_per_second": 543.466
+    },
+    {
+      "epoch": 1.2096774193548387,
+      "grad_norm": 0.5005422234535217,
+      "learning_rate": 0.0001934564464599461,
+      "loss": 0.37073755264282227,
+      "num_input_tokens_seen": 496182,
+      "step": 150,
+      "train_runtime": 911.4473,
+      "train_tokens_per_second": 544.389
+    },
+    {
+      "epoch": 1.2096774193548387,
+      "eval_loss": 2.5809271335601807,
+      "eval_runtime": 12.2545,
+      "eval_samples_per_second": 4.243,
+      "eval_steps_per_second": 2.122,
+      "num_input_tokens_seen": 496182,
+      "step": 150
+    },
+    {
+      "epoch": 1.217741935483871,
+      "grad_norm": 0.6077068448066711,
+      "learning_rate": 0.00019336563735446446,
+      "loss": 0.46918922662734985,
+      "num_input_tokens_seen": 499482,
+      "step": 151,
+      "train_runtime": 946.9561,
+      "train_tokens_per_second": 527.461
+    },
+    {
+      "epoch": 1.2258064516129032,
+      "grad_norm": 0.5957709550857544,
+      "learning_rate": 0.00019327422408801744,
+      "loss": 0.43228402733802795,
+      "num_input_tokens_seen": 502832,
+      "step": 152,
+      "train_runtime": 951.617,
+      "train_tokens_per_second": 528.397
+    },
+    {
+      "epoch": 1.2338709677419355,
+      "grad_norm": 0.5242084264755249,
+      "learning_rate": 0.0001931822072521323,
+      "loss": 0.40411657094955444,
+      "num_input_tokens_seen": 506196,
+      "step": 153,
+      "train_runtime": 956.5068,
+      "train_tokens_per_second": 529.213
+    },
+    {
+      "epoch": 1.2419354838709677,
+      "grad_norm": 0.6395658254623413,
+      "learning_rate": 0.00019308958744224217,
+      "loss": 0.41684025526046753,
+      "num_input_tokens_seen": 509348,
+      "step": 154,
+      "train_runtime": 961.0623,
+      "train_tokens_per_second": 529.984
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.5764552354812622,
+      "learning_rate": 0.00019299636525768173,
+      "loss": 0.42104196548461914,
+      "num_input_tokens_seen": 512728,
+      "step": 155,
+      "train_runtime": 966.1691,
+      "train_tokens_per_second": 530.681
+    },
+    {
+      "epoch": 1.2580645161290323,
+      "grad_norm": 0.6039356589317322,
+      "learning_rate": 0.00019290254130168374,
+      "loss": 0.41675058007240295,
+      "num_input_tokens_seen": 516078,
+      "step": 156,
+      "train_runtime": 971.2647,
+      "train_tokens_per_second": 531.346
+    },
+    {
+      "epoch": 1.2661290322580645,
+      "grad_norm": 0.5306972861289978,
+      "learning_rate": 0.00019280811618137484,
+      "loss": 0.44224196672439575,
+      "num_input_tokens_seen": 519482,
+      "step": 157,
+      "train_runtime": 976.5865,
+      "train_tokens_per_second": 531.936
+    },
+    {
+      "epoch": 1.2741935483870968,
+      "grad_norm": 0.5725175142288208,
+      "learning_rate": 0.00019271309050777183,
+      "loss": 0.4407888650894165,
+      "num_input_tokens_seen": 522758,
+      "step": 158,
+      "train_runtime": 981.664,
+      "train_tokens_per_second": 532.522
+    },
+    {
+      "epoch": 1.282258064516129,
+      "grad_norm": 0.5713056921958923,
+      "learning_rate": 0.00019261746489577765,
+      "loss": 0.4012797474861145,
+      "num_input_tokens_seen": 525784,
+      "step": 159,
+      "train_runtime": 986.2122,
+      "train_tokens_per_second": 533.135
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 0.4923282265663147,
+      "learning_rate": 0.00019252123996417738,
+      "loss": 0.3848939836025238,
+      "num_input_tokens_seen": 529130,
+      "step": 160,
+      "train_runtime": 991.1125,
+      "train_tokens_per_second": 533.875
+    },
+    {
+      "epoch": 1.2983870967741935,
+      "grad_norm": 0.5265042185783386,
+      "learning_rate": 0.00019242441633563417,
+      "loss": 0.4451833665370941,
+      "num_input_tokens_seen": 532308,
+      "step": 161,
+      "train_runtime": 995.6839,
+      "train_tokens_per_second": 534.615
+    },
+    {
+      "epoch": 1.3064516129032258,
+      "grad_norm": 0.499607652425766,
+      "learning_rate": 0.00019232699463668542,
+      "loss": 0.37897300720214844,
+      "num_input_tokens_seen": 535662,
+      "step": 162,
+      "train_runtime": 1000.3675,
+      "train_tokens_per_second": 535.465
+    },
+    {
+      "epoch": 1.314516129032258,
+      "grad_norm": 0.6007817983627319,
+      "learning_rate": 0.00019222897549773848,
+      "loss": 0.40716710686683655,
+      "num_input_tokens_seen": 539056,
+      "step": 163,
+      "train_runtime": 1005.1457,
+      "train_tokens_per_second": 536.296
+    },
+    {
+      "epoch": 1.3225806451612903,
+      "grad_norm": 0.5382058620452881,
+      "learning_rate": 0.0001921303595530667,
+      "loss": 0.3810466527938843,
+      "num_input_tokens_seen": 542450,
+      "step": 164,
+      "train_runtime": 1009.817,
+      "train_tokens_per_second": 537.177
+    },
+    {
+      "epoch": 1.3306451612903225,
+      "grad_norm": 0.7238770127296448,
+      "learning_rate": 0.00019203114744080542,
+      "loss": 0.48473918437957764,
+      "num_input_tokens_seen": 545796,
+      "step": 165,
+      "train_runtime": 1014.3995,
+      "train_tokens_per_second": 538.048
+    },
+    {
+      "epoch": 1.3387096774193548,
+      "grad_norm": 0.6477717757225037,
+      "learning_rate": 0.0001919313398029475,
+      "loss": 0.44077277183532715,
+      "num_input_tokens_seen": 548960,
+      "step": 166,
+      "train_runtime": 1018.9076,
+      "train_tokens_per_second": 538.773
+    },
+    {
+      "epoch": 1.346774193548387,
+      "grad_norm": 0.5776754021644592,
+      "learning_rate": 0.00019183093728533966,
+      "loss": 0.428521066904068,
+      "num_input_tokens_seen": 552310,
+      "step": 167,
+      "train_runtime": 1023.4907,
+      "train_tokens_per_second": 539.634
+    },
+    {
+      "epoch": 1.3548387096774195,
+      "grad_norm": 0.589496910572052,
+      "learning_rate": 0.00019172994053767784,
+      "loss": 0.44079262018203735,
+      "num_input_tokens_seen": 555660,
+      "step": 168,
+      "train_runtime": 1028.1528,
+      "train_tokens_per_second": 540.445
+    },
+    {
+      "epoch": 1.3629032258064515,
+      "grad_norm": 0.5510254502296448,
+      "learning_rate": 0.0001916283502135033,
+      "loss": 0.4306449592113495,
+      "num_input_tokens_seen": 558942,
+      "step": 169,
+      "train_runtime": 1032.8701,
+      "train_tokens_per_second": 541.154
+    },
+    {
+      "epoch": 1.370967741935484,
+      "grad_norm": 0.6984896063804626,
+      "learning_rate": 0.00019152616697019822,
+      "loss": 0.4740030765533447,
+      "num_input_tokens_seen": 561914,
+      "step": 170,
+      "train_runtime": 1037.1046,
+      "train_tokens_per_second": 541.81
+    },
+    {
+      "epoch": 1.379032258064516,
+      "grad_norm": 0.5186349153518677,
+      "learning_rate": 0.0001914233914689815,
+      "loss": 0.46226799488067627,
+      "num_input_tokens_seen": 565128,
+      "step": 171,
+      "train_runtime": 1041.8057,
+      "train_tokens_per_second": 542.45
+    },
+    {
+      "epoch": 1.3870967741935485,
+      "grad_norm": 0.6191644668579102,
+      "learning_rate": 0.00019132002437490458,
+      "loss": 0.4157659709453583,
+      "num_input_tokens_seen": 568492,
+      "step": 172,
+      "train_runtime": 1046.6025,
+      "train_tokens_per_second": 543.179
+    },
+    {
+      "epoch": 1.3951612903225805,
+      "grad_norm": 0.5917848944664001,
+      "learning_rate": 0.00019121606635684696,
+      "loss": 0.4816833436489105,
+      "num_input_tokens_seen": 571838,
+      "step": 173,
+      "train_runtime": 1051.4773,
+      "train_tokens_per_second": 543.842
+    },
+    {
+      "epoch": 1.403225806451613,
+      "grad_norm": 0.4845684766769409,
+      "learning_rate": 0.00019111151808751196,
+      "loss": 0.3603062629699707,
+      "num_input_tokens_seen": 575172,
+      "step": 174,
+      "train_runtime": 1056.4402,
+      "train_tokens_per_second": 544.444
+    },
+    {
+      "epoch": 1.4112903225806452,
+      "grad_norm": 0.6082286238670349,
+      "learning_rate": 0.00019100638024342244,
+      "loss": 0.4300457835197449,
+      "num_input_tokens_seen": 578472,
+      "step": 175,
+      "train_runtime": 1061.2665,
+      "train_tokens_per_second": 545.077
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 0.5522608160972595,
+      "learning_rate": 0.00019090065350491626,
+      "loss": 0.41547319293022156,
+      "num_input_tokens_seen": 581854,
+      "step": 176,
+      "train_runtime": 1066.3155,
+      "train_tokens_per_second": 545.668
+    },
+    {
+      "epoch": 1.4274193548387097,
+      "grad_norm": 0.553947925567627,
+      "learning_rate": 0.00019079433855614201,
+      "loss": 0.4243569076061249,
+      "num_input_tokens_seen": 585240,
+      "step": 177,
+      "train_runtime": 1071.1687,
+      "train_tokens_per_second": 546.356
+    },
+    {
+      "epoch": 1.435483870967742,
+      "grad_norm": 0.7811908721923828,
+      "learning_rate": 0.00019068743608505455,
+      "loss": 0.46935024857521057,
+      "num_input_tokens_seen": 588466,
+      "step": 178,
+      "train_runtime": 1075.7542,
+      "train_tokens_per_second": 547.026
+    },
+    {
+      "epoch": 1.4435483870967742,
+      "grad_norm": 0.6735747456550598,
+      "learning_rate": 0.0001905799467834105,
+      "loss": 0.4160788357257843,
+      "num_input_tokens_seen": 591716,
+      "step": 179,
+      "train_runtime": 1080.4651,
+      "train_tokens_per_second": 547.649
+    },
+    {
+      "epoch": 1.4516129032258065,
+      "grad_norm": 0.563190221786499,
+      "learning_rate": 0.00019047187134676387,
+      "loss": 0.4146038293838501,
+      "num_input_tokens_seen": 595124,
+      "step": 180,
+      "train_runtime": 1085.2463,
+      "train_tokens_per_second": 548.377
+    },
+    {
+      "epoch": 1.4516129032258065,
+      "eval_loss": 2.515367269515991,
+      "eval_runtime": 12.2457,
+      "eval_samples_per_second": 4.246,
+      "eval_steps_per_second": 2.123,
+      "num_input_tokens_seen": 595124,
+      "step": 180
+    },
+    {
+      "epoch": 1.4596774193548387,
+      "grad_norm": 0.5683964490890503,
+      "learning_rate": 0.0001903632104744614,
+      "loss": 0.4067952334880829,
+      "num_input_tokens_seen": 598560,
+      "step": 181,
+      "train_runtime": 1121.3958,
+      "train_tokens_per_second": 533.763
+    },
+    {
+      "epoch": 1.467741935483871,
+      "grad_norm": 0.7010737061500549,
+      "learning_rate": 0.00019025396486963827,
+      "loss": 0.47173696756362915,
+      "num_input_tokens_seen": 601832,
+      "step": 182,
+      "train_runtime": 1125.8583,
+      "train_tokens_per_second": 534.554
+    },
+    {
+      "epoch": 1.4758064516129032,
+      "grad_norm": 0.5702352523803711,
+      "learning_rate": 0.0001901441352392133,
+      "loss": 0.45567232370376587,
+      "num_input_tokens_seen": 605124,
+      "step": 183,
+      "train_runtime": 1130.613,
+      "train_tokens_per_second": 535.218
+    },
+    {
+      "epoch": 1.4838709677419355,
+      "grad_norm": 0.5321019291877747,
+      "learning_rate": 0.00019003372229388452,
+      "loss": 0.44429701566696167,
+      "num_input_tokens_seen": 608550,
+      "step": 184,
+      "train_runtime": 1135.4582,
+      "train_tokens_per_second": 535.951
+    },
+    {
+      "epoch": 1.4919354838709677,
+      "grad_norm": 0.5422691106796265,
+      "learning_rate": 0.0001899227267481246,
+      "loss": 0.4688451588153839,
+      "num_input_tokens_seen": 611912,
+      "step": 185,
+      "train_runtime": 1140.4185,
+      "train_tokens_per_second": 536.568
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 0.5946955680847168,
+      "learning_rate": 0.00018981114932017609,
+      "loss": 0.4600903391838074,
+      "num_input_tokens_seen": 615114,
+      "step": 186,
+      "train_runtime": 1145.289,
+      "train_tokens_per_second": 537.082
+    },
+    {
+      "epoch": 1.5080645161290323,
+      "grad_norm": 0.5127443075180054,
+      "learning_rate": 0.00018969899073204686,
+      "loss": 0.3844794034957886,
+      "num_input_tokens_seen": 618532,
+      "step": 187,
+      "train_runtime": 1150.4096,
+      "train_tokens_per_second": 537.662
+    },
+    {
+      "epoch": 1.5161290322580645,
+      "grad_norm": 0.592376172542572,
+      "learning_rate": 0.00018958625170950545,
+      "loss": 0.4546698033809662,
+      "num_input_tokens_seen": 621584,
+      "step": 188,
+      "train_runtime": 1155.2548,
+      "train_tokens_per_second": 538.049
+    },
+    {
+      "epoch": 1.5241935483870968,
+      "grad_norm": 0.5506513714790344,
+      "learning_rate": 0.00018947293298207635,
+      "loss": 0.45888251066207886,
+      "num_input_tokens_seen": 624890,
+      "step": 189,
+      "train_runtime": 1160.301,
+      "train_tokens_per_second": 538.559
+    },
+    {
+      "epoch": 1.532258064516129,
+      "grad_norm": 0.7385762929916382,
+      "learning_rate": 0.00018935903528303523,
+      "loss": 0.47872915863990784,
+      "num_input_tokens_seen": 627898,
+      "step": 190,
+      "train_runtime": 1164.9125,
+      "train_tokens_per_second": 539.009
+    },
+    {
+      "epoch": 1.5403225806451613,
+      "grad_norm": 0.48639681935310364,
+      "learning_rate": 0.0001892445593494042,
+      "loss": 0.3977377414703369,
+      "num_input_tokens_seen": 631260,
+      "step": 191,
+      "train_runtime": 1169.8778,
+      "train_tokens_per_second": 539.595
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 0.5805647969245911,
+      "learning_rate": 0.0001891295059219472,
+      "loss": 0.4197312593460083,
+      "num_input_tokens_seen": 634598,
+      "step": 192,
+      "train_runtime": 1174.6329,
+      "train_tokens_per_second": 540.252
+    },
+    {
+      "epoch": 1.5564516129032258,
+      "grad_norm": 0.55940842628479,
+      "learning_rate": 0.00018901387574516497,
+      "loss": 0.45023104548454285,
+      "num_input_tokens_seen": 637776,
+      "step": 193,
+      "train_runtime": 1179.3065,
+      "train_tokens_per_second": 540.806
+    },
+    {
+      "epoch": 1.564516129032258,
+      "grad_norm": 0.5386122465133667,
+      "learning_rate": 0.00018889766956729044,
+      "loss": 0.4272036850452423,
+      "num_input_tokens_seen": 641100,
+      "step": 194,
+      "train_runtime": 1183.9205,
+      "train_tokens_per_second": 541.506
+    },
+    {
+      "epoch": 1.5725806451612905,
+      "grad_norm": 0.5532417297363281,
+      "learning_rate": 0.00018878088814028364,
+      "loss": 0.40990498661994934,
+      "num_input_tokens_seen": 644472,
+      "step": 195,
+      "train_runtime": 1188.6266,
+      "train_tokens_per_second": 542.199
+    },
+    {
+      "epoch": 1.5806451612903225,
+      "grad_norm": 0.5226889252662659,
+      "learning_rate": 0.00018866353221982718,
+      "loss": 0.3513818383216858,
+      "num_input_tokens_seen": 647738,
+      "step": 196,
+      "train_runtime": 1193.3282,
+      "train_tokens_per_second": 542.8
+    },
+    {
+      "epoch": 1.588709677419355,
+      "grad_norm": 0.6106642484664917,
+      "learning_rate": 0.000188545602565321,
+      "loss": 0.4655504822731018,
+      "num_input_tokens_seen": 651030,
+      "step": 197,
+      "train_runtime": 1197.831,
+      "train_tokens_per_second": 543.507
+    },
+    {
+      "epoch": 1.596774193548387,
+      "grad_norm": 0.6960707902908325,
+      "learning_rate": 0.00018842709993987776,
+      "loss": 0.46613919734954834,
+      "num_input_tokens_seen": 654362,
+      "step": 198,
+      "train_runtime": 1202.5116,
+      "train_tokens_per_second": 544.163
+    },
+    {
+      "epoch": 1.6048387096774195,
+      "grad_norm": 0.6360778212547302,
+      "learning_rate": 0.00018830802511031762,
+      "loss": 0.40627503395080566,
+      "num_input_tokens_seen": 657668,
+      "step": 199,
+      "train_runtime": 1207.1242,
+      "train_tokens_per_second": 544.822
+    },
+    {
+      "epoch": 1.6129032258064515,
+      "grad_norm": 0.5545072555541992,
+      "learning_rate": 0.0001881883788471636,
+      "loss": 0.4130961298942566,
+      "num_input_tokens_seen": 661006,
+      "step": 200,
+      "train_runtime": 1211.7449,
+      "train_tokens_per_second": 545.499
+    },
+    {
+      "epoch": 1.620967741935484,
+      "grad_norm": 0.5094169974327087,
+      "learning_rate": 0.00018806816192463625,
+      "loss": 0.40930330753326416,
+      "num_input_tokens_seen": 664466,
+      "step": 201,
+      "train_runtime": 1216.6247,
+      "train_tokens_per_second": 546.155
+    },
+    {
+      "epoch": 1.629032258064516,
+      "grad_norm": 0.6566628813743591,
+      "learning_rate": 0.0001879473751206489,
+      "loss": 0.4623265266418457,
+      "num_input_tokens_seen": 667634,
+      "step": 202,
+      "train_runtime": 1221.1278,
+      "train_tokens_per_second": 546.736
+    },
+    {
+      "epoch": 1.6370967741935485,
+      "grad_norm": 0.542569100856781,
+      "learning_rate": 0.00018782601921680256,
+      "loss": 0.44509240984916687,
+      "num_input_tokens_seen": 671080,
+      "step": 203,
+      "train_runtime": 1225.9964,
+      "train_tokens_per_second": 547.375
+    },
+    {
+      "epoch": 1.6451612903225805,
+      "grad_norm": 0.628744900226593,
+      "learning_rate": 0.00018770409499838073,
+      "loss": 0.4878479540348053,
+      "num_input_tokens_seen": 674508,
+      "step": 204,
+      "train_runtime": 1230.9685,
+      "train_tokens_per_second": 547.949
+    },
+    {
+      "epoch": 1.653225806451613,
+      "grad_norm": 0.558655858039856,
+      "learning_rate": 0.0001875816032543445,
+      "loss": 0.41668811440467834,
+      "num_input_tokens_seen": 677700,
+      "step": 205,
+      "train_runtime": 1235.5957,
+      "train_tokens_per_second": 548.48
+    },
+    {
+      "epoch": 1.661290322580645,
+      "grad_norm": 0.5680912137031555,
+      "learning_rate": 0.00018745854477732733,
+      "loss": 0.42818352580070496,
+      "num_input_tokens_seen": 680978,
+      "step": 206,
+      "train_runtime": 1240.4198,
+      "train_tokens_per_second": 548.99
+    },
+    {
+      "epoch": 1.6693548387096775,
+      "grad_norm": 0.5377702713012695,
+      "learning_rate": 0.00018733492036363005,
+      "loss": 0.4158581495285034,
+      "num_input_tokens_seen": 684328,
+      "step": 207,
+      "train_runtime": 1245.2876,
+      "train_tokens_per_second": 549.534
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 0.5735633969306946,
+      "learning_rate": 0.0001872107308132155,
+      "loss": 0.4311006963253021,
+      "num_input_tokens_seen": 687676,
+      "step": 208,
+      "train_runtime": 1250.0579,
+      "train_tokens_per_second": 550.115
+    },
+    {
+      "epoch": 1.685483870967742,
+      "grad_norm": 0.5267180800437927,
+      "learning_rate": 0.00018708597692970353,
+      "loss": 0.38378968834877014,
+      "num_input_tokens_seen": 691000,
+      "step": 209,
+      "train_runtime": 1254.9146,
+      "train_tokens_per_second": 550.635
+    },
+    {
+      "epoch": 1.6935483870967742,
+      "grad_norm": 0.693577766418457,
+      "learning_rate": 0.00018696065952036571,
+      "loss": 0.46765580773353577,
+      "num_input_tokens_seen": 694156,
+      "step": 210,
+      "train_runtime": 1259.4417,
+      "train_tokens_per_second": 551.162
+    },
+    {
+      "epoch": 1.6935483870967742,
+      "eval_loss": 2.4968807697296143,
+      "eval_runtime": 12.2486,
+      "eval_samples_per_second": 4.245,
+      "eval_steps_per_second": 2.123,
+      "num_input_tokens_seen": 694156,
+      "step": 210
+    },
+    {
+      "epoch": 1.7016129032258065,
+      "grad_norm": 0.5160212516784668,
+      "learning_rate": 0.00018683477939612021,
+      "loss": 0.383802592754364,
+      "num_input_tokens_seen": 697510,
+      "step": 211,
+      "train_runtime": 1296.5517,
+      "train_tokens_per_second": 537.973
+    },
+    {
+      "epoch": 1.7096774193548387,
+      "grad_norm": 0.5070053935050964,
+      "learning_rate": 0.0001867083373715264,
+      "loss": 0.39444947242736816,
+      "num_input_tokens_seen": 700884,
+      "step": 212,
+      "train_runtime": 1301.2787,
+      "train_tokens_per_second": 538.612
+    },
+    {
+      "epoch": 1.717741935483871,
+      "grad_norm": 0.6223205327987671,
+      "learning_rate": 0.00018658133426477965,
+      "loss": 0.4722726345062256,
+      "num_input_tokens_seen": 704288,
+      "step": 213,
+      "train_runtime": 1306.1565,
+      "train_tokens_per_second": 539.206
+    },
+    {
+      "epoch": 1.7258064516129032,
+      "grad_norm": 0.6386722326278687,
+      "learning_rate": 0.00018645377089770616,
+      "loss": 0.40639519691467285,
+      "num_input_tokens_seen": 707404,
+      "step": 214,
+      "train_runtime": 1310.6997,
+      "train_tokens_per_second": 539.715
+    },
+    {
+      "epoch": 1.7338709677419355,
+      "grad_norm": 0.5457852482795715,
+      "learning_rate": 0.00018632564809575742,
+      "loss": 0.40314507484436035,
+      "num_input_tokens_seen": 710738,
+      "step": 215,
+      "train_runtime": 1315.6988,
+      "train_tokens_per_second": 540.198
+    },
+    {
+      "epoch": 1.7419354838709677,
+      "grad_norm": 0.6622642278671265,
+      "learning_rate": 0.00018619696668800492,
+      "loss": 0.4788181185722351,
+      "num_input_tokens_seen": 714052,
+      "step": 216,
+      "train_runtime": 1320.6297,
+      "train_tokens_per_second": 540.691
+    },
+    {
+      "epoch": 1.75,
+      "grad_norm": 0.5140097737312317,
+      "learning_rate": 0.00018606772750713504,
+      "loss": 0.4071129262447357,
+      "num_input_tokens_seen": 717406,
+      "step": 217,
+      "train_runtime": 1325.7428,
+      "train_tokens_per_second": 541.135
+    },
+    {
+      "epoch": 1.7580645161290323,
+      "grad_norm": 0.5269467830657959,
+      "learning_rate": 0.00018593793138944328,
+      "loss": 0.4001181721687317,
+      "num_input_tokens_seen": 720724,
+      "step": 218,
+      "train_runtime": 1330.8634,
+      "train_tokens_per_second": 541.546
+    },
+    {
+      "epoch": 1.7661290322580645,
+      "grad_norm": 0.6335601806640625,
+      "learning_rate": 0.0001858075791748291,
+      "loss": 0.5110459923744202,
+      "num_input_tokens_seen": 724156,
+      "step": 219,
+      "train_runtime": 1335.9177,
+      "train_tokens_per_second": 542.066
+    },
+    {
+      "epoch": 1.7741935483870968,
+      "grad_norm": 0.5240700840950012,
+      "learning_rate": 0.0001856766717067904,
+      "loss": 0.42234110832214355,
+      "num_input_tokens_seen": 727386,
+      "step": 220,
+      "train_runtime": 1340.9375,
+      "train_tokens_per_second": 542.446
+    },
+    {
+      "epoch": 1.782258064516129,
+      "grad_norm": 0.4940207898616791,
+      "learning_rate": 0.00018554520983241814,
+      "loss": 0.40678343176841736,
+      "num_input_tokens_seen": 730576,
+      "step": 221,
+      "train_runtime": 1345.6557,
+      "train_tokens_per_second": 542.914
+    },
+    {
+      "epoch": 1.7903225806451613,
+      "grad_norm": 0.4842962324619293,
+      "learning_rate": 0.00018541319440239066,
+      "loss": 0.37585100531578064,
+      "num_input_tokens_seen": 734042,
+      "step": 222,
+      "train_runtime": 1350.5324,
+      "train_tokens_per_second": 543.52
+    },
+    {
+      "epoch": 1.7983870967741935,
+      "grad_norm": 0.5687828660011292,
+      "learning_rate": 0.00018528062627096845,
+      "loss": 0.4230225086212158,
+      "num_input_tokens_seen": 737388,
+      "step": 223,
+      "train_runtime": 1355.36,
+      "train_tokens_per_second": 544.053
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 0.5137950778007507,
+      "learning_rate": 0.0001851475062959884,
+      "loss": 0.4101966619491577,
+      "num_input_tokens_seen": 740714,
+      "step": 224,
+      "train_runtime": 1359.9896,
+      "train_tokens_per_second": 544.647
+    },
+    {
+      "epoch": 1.814516129032258,
+      "grad_norm": 0.5618765950202942,
+      "learning_rate": 0.00018501383533885837,
+      "loss": 0.3829328417778015,
+      "num_input_tokens_seen": 743952,
+      "step": 225,
+      "train_runtime": 1364.615,
+      "train_tokens_per_second": 545.174
+    },
+    {
+      "epoch": 1.8225806451612905,
+      "grad_norm": 0.6131100058555603,
+      "learning_rate": 0.00018487961426455157,
+      "loss": 0.4687027335166931,
+      "num_input_tokens_seen": 747324,
+      "step": 226,
+      "train_runtime": 1369.3307,
+      "train_tokens_per_second": 545.759
+    },
+    {
+      "epoch": 1.8306451612903225,
+      "grad_norm": 0.6085920929908752,
+      "learning_rate": 0.0001847448439416009,
+      "loss": 0.4704275131225586,
+      "num_input_tokens_seen": 750660,
+      "step": 227,
+      "train_runtime": 1373.9767,
+      "train_tokens_per_second": 546.341
+    },
+    {
+      "epoch": 1.838709677419355,
+      "grad_norm": 0.682384192943573,
+      "learning_rate": 0.00018460952524209355,
+      "loss": 0.4709826111793518,
+      "num_input_tokens_seen": 753890,
+      "step": 228,
+      "train_runtime": 1378.611,
+      "train_tokens_per_second": 546.848
+    },
+    {
+      "epoch": 1.846774193548387,
+      "grad_norm": 0.6233918070793152,
+      "learning_rate": 0.0001844736590416651,
+      "loss": 0.41492459177970886,
+      "num_input_tokens_seen": 757196,
+      "step": 229,
+      "train_runtime": 1383.1991,
+      "train_tokens_per_second": 547.424
+    },
+    {
+      "epoch": 1.8548387096774195,
+      "grad_norm": 0.6080527305603027,
+      "learning_rate": 0.00018433724621949392,
+      "loss": 0.41505876183509827,
+      "num_input_tokens_seen": 760606,
+      "step": 230,
+      "train_runtime": 1387.9745,
+      "train_tokens_per_second": 547.997
+    },
+    {
+      "epoch": 1.8629032258064515,
+      "grad_norm": 0.5890636444091797,
+      "learning_rate": 0.00018420028765829568,
+      "loss": 0.4203844964504242,
+      "num_input_tokens_seen": 763910,
+      "step": 231,
+      "train_runtime": 1392.7824,
+      "train_tokens_per_second": 548.478
+    },
+    {
+      "epoch": 1.870967741935484,
+      "grad_norm": 0.5636559724807739,
+      "learning_rate": 0.00018406278424431736,
+      "loss": 0.44402995705604553,
+      "num_input_tokens_seen": 767166,
+      "step": 232,
+      "train_runtime": 1397.3628,
+      "train_tokens_per_second": 549.01
+    },
+    {
+      "epoch": 1.879032258064516,
+      "grad_norm": 0.5407444834709167,
+      "learning_rate": 0.00018392473686733163,
+      "loss": 0.43737179040908813,
+      "num_input_tokens_seen": 770592,
+      "step": 233,
+      "train_runtime": 1402.2872,
+      "train_tokens_per_second": 549.525
+    },
+    {
+      "epoch": 1.8870967741935485,
+      "grad_norm": 0.6623079776763916,
+      "learning_rate": 0.00018378614642063115,
+      "loss": 0.4496549367904663,
+      "num_input_tokens_seen": 773594,
+      "step": 234,
+      "train_runtime": 1406.7685,
+      "train_tokens_per_second": 549.909
+    },
+    {
+      "epoch": 1.8951612903225805,
+      "grad_norm": 0.676179826259613,
+      "learning_rate": 0.00018364701380102266,
+      "loss": 0.4767211973667145,
+      "num_input_tokens_seen": 776592,
+      "step": 235,
+      "train_runtime": 1411.1241,
+      "train_tokens_per_second": 550.336
+    },
+    {
+      "epoch": 1.903225806451613,
+      "grad_norm": 0.5864387154579163,
+      "learning_rate": 0.0001835073399088214,
+      "loss": 0.4903409779071808,
+      "num_input_tokens_seen": 779942,
+      "step": 236,
+      "train_runtime": 1416.1385,
+      "train_tokens_per_second": 550.753
+    },
+    {
+      "epoch": 1.911290322580645,
+      "grad_norm": 0.4824385643005371,
+      "learning_rate": 0.00018336712564784503,
+      "loss": 0.4071444869041443,
+      "num_input_tokens_seen": 783252,
+      "step": 237,
+      "train_runtime": 1420.8937,
+      "train_tokens_per_second": 551.239
+    },
+    {
+      "epoch": 1.9193548387096775,
+      "grad_norm": 0.5865503549575806,
+      "learning_rate": 0.00018322637192540785,
+      "loss": 0.4477483034133911,
+      "num_input_tokens_seen": 786556,
+      "step": 238,
+      "train_runtime": 1425.6472,
+      "train_tokens_per_second": 551.719
+    },
+    {
+      "epoch": 1.9274193548387095,
+      "grad_norm": 0.6280811429023743,
+      "learning_rate": 0.00018308507965231508,
+      "loss": 0.4521775245666504,
+      "num_input_tokens_seen": 789882,
+      "step": 239,
+      "train_runtime": 1430.4668,
+      "train_tokens_per_second": 552.185
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 0.5393728613853455,
+      "learning_rate": 0.00018294324974285677,
+      "loss": 0.3891618549823761,
+      "num_input_tokens_seen": 793254,
+      "step": 240,
+      "train_runtime": 1435.2181,
+      "train_tokens_per_second": 552.706
+    },
+    {
+      "epoch": 1.935483870967742,
+      "eval_loss": 2.3643229007720947,
+      "eval_runtime": 12.1283,
+      "eval_samples_per_second": 4.287,
+      "eval_steps_per_second": 2.144,
+      "num_input_tokens_seen": 793254,
+      "step": 240
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 1240,
+  "num_input_tokens_seen": 793254,
+  "num_train_epochs": 10,
+  "save_steps": 30,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.1879165274109056e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbb176183e719cb4390ae273728327f28a311740b24c3049448e069def331d2d
+size 5713