Add 11 files

.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 +1,99 @@
+---
+base_model: google/gemma-4-31B-it
+library_name: peft
+license: other
+tags:
+  - lora
+  - peft
+  - adapter
+  - adaption
+---
+
+# adaption_verified_math_code_instruct
+
+## Model Training
+
+A LORA adapter for `google/gemma-4-31B-it`. This model was trained with SFT using [Adaption](https://adaptionlabs.ai)'s AutoScientist on the verified_math_code_instruct dataset.
+
+
+![Training metrics](training-metrics.png)
+
+### AutoScientist Config
+
+```json
+{
+  "job_id": "9f46e2b9-ac73-4b97-ad13-cc6ccd1501cf",
+  "training_experiment_id": "3ce14a29-6d5d-48ea-849f-372a6102c360",
+  "original_model_name": "google/gemma-4-31B-it",
+  "trained_model_name": "adaption_verified_math_code_instruct",
+  "training_method": "sft",
+  "training_type": "lora",
+  "data_format": "chat",
+  "hyperparams": {
+    "lora": "true",
+    "lora_r": 8,
+    "n_evals": 5,
+    "n_epochs": 1,
+    "batch_size": "max",
+    "lora_alpha": 8,
+    "lora_dropout": 0,
+    "min_lr_ratio": 0.1,
+    "warmup_ratio": 0.1,
+    "weight_decay": 0,
+    "learning_rate": 0.0001,
+    "max_grad_norm": 2,
+    "base_model_size": "31B",
+    "train_on_inputs": "false",
+    "training_method": "sft",
+    "lr_scheduler_type": "cosine",
+    "scheduler_num_cycles": 0.5,
+    "lora_trainable_modules": "q_proj,v_proj"
+  }
+}
+```
+
+## Training Data
+
+The model was trained on 21,063 rows of adapted data with the following domain distribution: math (56%), code (25%), science (3%), other (1%), technology (1%), history (1%), writing-editing-communication (1%), medical (1%), cooking (1%), games (1%), music (1%), entertainment (1%), language (1%), sports (1%), culture (1%), animal-nature (0%), religion (0%), geography (0%), travel (0%), how-to (0%), governance (0%), academic-education (0%), corporate-business (0%), art (0%), transportation (0%), fitness-sports (0%), personal-growth (0%), career-workplace (0%), personal-finance (0%), product-advice (0%), legal (0%), roleplay (0%), architecture-design (0%), agriculture (0%), marketing (0%), parenting-family (0%), dating (0%), news (0%), data-analysis-visualization (0%), fashion-beauty (0%), market-analysis (0%), social (0%), hr (0%), finance (0%).
+
+## Model Evaluation
+
+The model was evaluated on an in-distribution held-out test set as well as a broader domain-specific test set to measure generalization.
+
+
+![Win rates](win-rates.png)
+
+
+## How to use
+
+```bash
+pip install torch transformers peft
+```
+
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from peft import PeftModel
+
+BASE = "google/gemma-4-31B-it"
+ADAPTER = "<this-repo-id>"
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+dtype = torch.float32 if device == "cpu" else torch.bfloat16
+
+base = AutoModelForCausalLM.from_pretrained(BASE, dtype=dtype).to(device)
+model = PeftModel.from_pretrained(base, ADAPTER)
+# Optional: merge the LoRA weights into the base for faster inference
+model = model.merge_and_unload()
+model.eval()
+
+tokenizer = AutoTokenizer.from_pretrained(BASE)
+messages = [{"role": "user", "content": "Hello!"}]
+text = tokenizer.apply_chat_template(
+    messages, tokenize=False, add_generation_prompt=True)
+inputs = tokenizer(text, return_tensors="pt").to(device)
+
+with torch.inference_mode():
+    out = model.generate(**inputs, max_new_tokens=512)
+print(tokenizer.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
+```

adapter_config.json

@@ -0,0 +1,34 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "google/gemma-4-31B-it",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": [],
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 8,
+  "lora_bias": false,
+  "lora_dropout": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3efb30e9f06457a9dde4acb1c596c5f4f6fa754ab5395e67a3bdbc1db8e06e12
+size 43856576

chat_template.jinja

@@ -0,0 +1,347 @@
+{%- macro format_parameters(properties, required) -%}
+    {%- 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 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([])) -}}
+                    }
+                {%- 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'] -%}
+        {{- messages[0]['content'] | trim -}}
+        {%- 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 -%}
+
+            {%- 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 -%}
+
+        {%- if ns.prev_message_type == 'tool_call' and not ns_tr_out.flag -%}
+            {{- '<|tool_response>' -}}
+        {%- elif not (ns_tr_out.flag and not message.get('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' -}}
+        {%- if not enable_thinking | default(false) -%}
+            {{- '<|channel>thought\n<channel|>' -}}
+        {%- endif -%}
+    {%- endif -%}
+{%- endif -%}

config.json

@@ -0,0 +1,178 @@
+{
+  "architectures": [
+    "Gemma4ForConditionalGeneration"
+  ],
+  "audio_config": null,
+  "audio_token_id": 258881,
+  "boa_token_id": 256000,
+  "boi_token_id": 255999,
+  "bos_token_id": 2,
+  "dtype": "bfloat16",
+  "eoa_token_id": 258883,
+  "eoa_token_index": 258883,
+  "eoi_token_id": 258882,
+  "eos_token_id": 1,
+  "image_token_id": 258880,
+  "initializer_range": 0.02,
+  "model_type": "gemma4",
+  "pad_token_id": 0,
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attention_k_eq_v": true,
+    "bos_token_id": 2,
+    "dtype": "bfloat16",
+    "enable_moe_block": false,
+    "eos_token_id": 1,
+    "expert_intermediate_size": null,
+    "final_logit_softcapping": 30.0,
+    "global_head_dim": 512,
+    "head_dim": 256,
+    "hidden_activation": "gelu_pytorch_tanh",
+    "hidden_size": 5376,
+    "hidden_size_per_layer_input": 0,
+    "initializer_range": 0.02,
+    "intermediate_size": 21504,
+    "layer_types": [
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "sliding_attention",
+      "full_attention"
+    ],
+    "max_position_embeddings": 262144,
+    "model_type": "gemma4_text",
+    "moe_intermediate_size": null,
+    "num_attention_heads": 32,
+    "num_experts": null,
+    "num_global_key_value_heads": 4,
+    "num_hidden_layers": 60,
+    "num_key_value_heads": 16,
+    "num_kv_shared_layers": 0,
+    "pad_token_id": 0,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "full_attention": {
+        "partial_rotary_factor": 0.25,
+        "rope_theta": 1000000.0,
+        "rope_type": "proportional"
+      },
+      "sliding_attention": {
+        "rope_theta": 10000.0,
+        "rope_type": "default"
+      }
+    },
+    "sliding_window": 1024,
+    "tie_word_embeddings": true,
+    "top_k_experts": null,
+    "use_bidirectional_attention": "vision",
+    "use_cache": false,
+    "use_double_wide_mlp": false,
+    "vocab_size": 262144,
+    "vocab_size_per_layer_input": 262144,
+    "torch_dtype": "bfloat16"
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.13.0",
+  "use_cache": false,
+  "video_token_id": 258884,
+  "vision_config": {
+    "_name_or_path": "",
+    "architectures": null,
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "chunk_size_feed_forward": 0,
+    "default_output_length": 280,
+    "global_head_dim": 72,
+    "head_dim": 72,
+    "hidden_activation": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "id2label": {
+      "0": "LABEL_0",
+      "1": "LABEL_1"
+    },
+    "initializer_range": 0.02,
+    "intermediate_size": 4304,
+    "is_encoder_decoder": false,
+    "label2id": {
+      "LABEL_0": 0,
+      "LABEL_1": 1
+    },
+    "max_position_embeddings": 131072,
+    "model_type": "gemma4_vision",
+    "num_attention_heads": 16,
+    "num_hidden_layers": 27,
+    "num_key_value_heads": 16,
+    "output_attentions": false,
+    "output_hidden_states": false,
+    "patch_size": 16,
+    "pooling_kernel_size": 3,
+    "position_embedding_size": 10240,
+    "problem_type": null,
+    "return_dict": true,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "rope_theta": 100.0,
+      "rope_type": "default"
+    },
+    "standardize": true,
+    "use_clipped_linears": false
+  },
+  "vision_soft_tokens_per_image": 280,
+  "torch_dtype": "bfloat16"
+}

special_tokens_map.json

@@ -0,0 +1,25 @@
+{
+  "audio_token": "<|audio|>",
+  "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|>",
+  "image_token": "<|image|>",
+  "mask_token": "<mask>",
+  "pad_token": "<pad>",
+  "soc_token": "<|channel>",
+  "sot_token": "<|turn>",
+  "stc_token": "<|tool_call>",
+  "std_token": "<|tool>",
+  "str_token": "<|tool_response>",
+  "think_token": "<|think|>",
+  "unk_token": "<unk>"
+}

tokenizer.json

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

tokenizer_config.json

@@ -0,0 +1,96 @@
+{
+  "additional_special_tokens": [
+    "<|video|>"
+  ],
+  "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|>",
+  "image_token": "<|image|>",
+  "is_local": false,
+  "local_files_only": true,
+  "mask_token": "<mask>",
+  "model_max_length": 49152,
+  "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>"
+}

trainer_state.json

@@ -0,0 +1,487 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 11,
+  "global_step": 59,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01694915254237288,
+      "grad_norm": 3.206836223602295,
+      "learning_rate": 0.0,
+      "loss": 3.482421875,
+      "step": 1
+    },
+    {
+      "epoch": 0.03389830508474576,
+      "grad_norm": 2.847381353378296,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 3.3974609375,
+      "step": 2
+    },
+    {
+      "epoch": 0.05084745762711865,
+      "grad_norm": 1.4680771827697754,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 3.1318359375,
+      "step": 3
+    },
+    {
+      "epoch": 0.06779661016949153,
+      "grad_norm": 2.6396775245666504,
+      "learning_rate": 5e-05,
+      "loss": 3.4658203125,
+      "step": 4
+    },
+    {
+      "epoch": 0.0847457627118644,
+      "grad_norm": 23.652711868286133,
+      "learning_rate": 6.666666666666667e-05,
+      "loss": 3.5107421875,
+      "step": 5
+    },
+    {
+      "epoch": 0.1016949152542373,
+      "grad_norm": 48.07037353515625,
+      "learning_rate": 8.333333333333334e-05,
+      "loss": 2.72869873046875,
+      "step": 6
+    },
+    {
+      "epoch": 0.11864406779661017,
+      "grad_norm": 4.403927803039551,
+      "learning_rate": 0.0001,
+      "loss": 3.408203125,
+      "step": 7
+    },
+    {
+      "epoch": 0.13559322033898305,
+      "grad_norm": 9.01473331451416,
+      "learning_rate": 9.992096792939448e-05,
+      "loss": 3.171875,
+      "step": 8
+    },
+    {
+      "epoch": 0.15254237288135594,
+      "grad_norm": 3.0288233757019043,
+      "learning_rate": 9.968414932060826e-05,
+      "loss": 3.2919921875,
+      "step": 9
+    },
+    {
+      "epoch": 0.1694915254237288,
+      "grad_norm": 6.546408176422119,
+      "learning_rate": 9.929037600764186e-05,
+      "loss": 2.862548828125,
+      "step": 10
+    },
+    {
+      "epoch": 0.1864406779661017,
+      "grad_norm": 13.21324634552002,
+      "learning_rate": 9.874103113361849e-05,
+      "loss": 3.15625,
+      "step": 11
+    },
+    {
+      "epoch": 0.2033898305084746,
+      "grad_norm": 4.1443328857421875,
+      "learning_rate": 9.80380442924435e-05,
+      "loss": 2.89453125,
+      "step": 12
+    },
+    {
+      "epoch": 0.22033898305084745,
+      "grad_norm": 2.7522470951080322,
+      "learning_rate": 9.718388475103545e-05,
+      "loss": 2.748046875,
+      "step": 13
+    },
+    {
+      "epoch": 0.23728813559322035,
+      "grad_norm": 172.51380920410156,
+      "learning_rate": 9.618155277593583e-05,
+      "loss": 2.4990234375,
+      "step": 14
+    },
+    {
+      "epoch": 0.2542372881355932,
+      "grad_norm": 645.0208129882812,
+      "learning_rate": 9.503456909476364e-05,
+      "loss": 2.1456298828125,
+      "step": 15
+    },
+    {
+      "epoch": 0.2542372881355932,
+      "eval_loss": 2.36572265625,
+      "eval_runtime": 8.5667,
+      "eval_samples_per_second": 0.584,
+      "eval_steps_per_second": 0.117,
+      "step": 15
+    },
+    {
+      "epoch": 0.2711864406779661,
+      "grad_norm": 2.9692535400390625,
+      "learning_rate": 9.374696252953101e-05,
+      "loss": 2.47412109375,
+      "step": 16
+    },
+    {
+      "epoch": 0.288135593220339,
+      "grad_norm": 4.023967266082764,
+      "learning_rate": 9.23232558452591e-05,
+      "loss": 2.4521484375,
+      "step": 17
+    },
+    {
+      "epoch": 0.3050847457627119,
+      "grad_norm": 2.6005005836486816,
+      "learning_rate": 9.07684498636009e-05,
+      "loss": 2.07373046875,
+      "step": 18
+    },
+    {
+      "epoch": 0.3220338983050847,
+      "grad_norm": 2.6452620029449463,
+      "learning_rate": 8.90880058972729e-05,
+      "loss": 2.11328125,
+      "step": 19
+    },
+    {
+      "epoch": 0.3389830508474576,
+      "grad_norm": 2.769963026046753,
+      "learning_rate": 8.72878265669949e-05,
+      "loss": 1.9404296875,
+      "step": 20
+    },
+    {
+      "epoch": 0.3559322033898305,
+      "grad_norm": 1.5048216581344604,
+      "learning_rate": 8.537423506831948e-05,
+      "loss": 1.5035400390625,
+      "step": 21
+    },
+    {
+      "epoch": 0.3728813559322034,
+      "grad_norm": 2.622932195663452,
+      "learning_rate": 8.335395296117701e-05,
+      "loss": 1.643157958984375,
+      "step": 22
+    },
+    {
+      "epoch": 0.3898305084745763,
+      "grad_norm": 2.5172510147094727,
+      "learning_rate": 8.123407656015125e-05,
+      "loss": 1.89501953125,
+      "step": 23
+    },
+    {
+      "epoch": 0.4067796610169492,
+      "grad_norm": 1.6744011640548706,
+      "learning_rate": 7.902205200841559e-05,
+      "loss": 1.7744140625,
+      "step": 24
+    },
+    {
+      "epoch": 0.423728813559322,
+      "grad_norm": 3.4982826709747314,
+      "learning_rate": 7.672564912288352e-05,
+      "loss": 1.69921875,
+      "step": 25
+    },
+    {
+      "epoch": 0.4406779661016949,
+      "grad_norm": 138.6426544189453,
+      "learning_rate": 7.435293410244342e-05,
+      "loss": 1.6044921875,
+      "step": 26
+    },
+    {
+      "epoch": 0.4406779661016949,
+      "eval_loss": 1.561767578125,
+      "eval_runtime": 8.5591,
+      "eval_samples_per_second": 0.584,
+      "eval_steps_per_second": 0.117,
+      "step": 26
+    },
+    {
+      "epoch": 0.4576271186440678,
+      "grad_norm": 1.965626835823059,
+      "learning_rate": 7.191224119514072e-05,
+      "loss": 1.5244140625,
+      "step": 27
+    },
+    {
+      "epoch": 0.4745762711864407,
+      "grad_norm": 1.6944293975830078,
+      "learning_rate": 6.94121434238277e-05,
+      "loss": 1.533203125,
+      "step": 28
+    },
+    {
+      "epoch": 0.4915254237288136,
+      "grad_norm": 3.656485080718994,
+      "learning_rate": 6.686142247310805e-05,
+      "loss": 1.517822265625,
+      "step": 29
+    },
+    {
+      "epoch": 0.5084745762711864,
+      "grad_norm": 0.9953148365020752,
+      "learning_rate": 6.426903784334943e-05,
+      "loss": 1.4697265625,
+      "step": 30
+    },
+    {
+      "epoch": 0.5254237288135594,
+      "grad_norm": 0.8330620527267456,
+      "learning_rate": 6.164409538011167e-05,
+      "loss": 1.4814453125,
+      "step": 31
+    },
+    {
+      "epoch": 0.5423728813559322,
+      "grad_norm": 0.7230986952781677,
+      "learning_rate": 5.899581528953207e-05,
+      "loss": 1.2818603515625,
+      "step": 32
+    },
+    {
+      "epoch": 0.559322033898305,
+      "grad_norm": 1.0042716264724731,
+      "learning_rate": 5.633349975201519e-05,
+      "loss": 1.5009765625,
+      "step": 33
+    },
+    {
+      "epoch": 0.576271186440678,
+      "grad_norm": 7.553008556365967,
+      "learning_rate": 5.366650024798482e-05,
+      "loss": 1.47021484375,
+      "step": 34
+    },
+    {
+      "epoch": 0.5932203389830508,
+      "grad_norm": 0.7747316956520081,
+      "learning_rate": 5.100418471046794e-05,
+      "loss": 1.36279296875,
+      "step": 35
+    },
+    {
+      "epoch": 0.6101694915254238,
+      "grad_norm": 0.7788944840431213,
+      "learning_rate": 4.835590461988836e-05,
+      "loss": 1.51953125,
+      "step": 36
+    },
+    {
+      "epoch": 0.6271186440677966,
+      "grad_norm": 0.8589951992034912,
+      "learning_rate": 4.573096215665058e-05,
+      "loss": 1.33056640625,
+      "step": 37
+    },
+    {
+      "epoch": 0.6271186440677966,
+      "eval_loss": 1.320556640625,
+      "eval_runtime": 8.5885,
+      "eval_samples_per_second": 0.582,
+      "eval_steps_per_second": 0.116,
+      "step": 37
+    },
+    {
+      "epoch": 0.6440677966101694,
+      "grad_norm": 677.2100219726562,
+      "learning_rate": 4.3138577526891964e-05,
+      "loss": 1.2919921875,
+      "step": 38
+    },
+    {
+      "epoch": 0.6610169491525424,
+      "grad_norm": 0.5605570077896118,
+      "learning_rate": 4.058785657617232e-05,
+      "loss": 1.18408203125,
+      "step": 39
+    },
+    {
+      "epoch": 0.6779661016949152,
+      "grad_norm": 389.38153076171875,
+      "learning_rate": 3.8087758804859295e-05,
+      "loss": 1.28271484375,
+      "step": 40
+    },
+    {
+      "epoch": 0.6949152542372882,
+      "grad_norm": 0.6740171909332275,
+      "learning_rate": 3.5647065897556596e-05,
+      "loss": 1.31640625,
+      "step": 41
+    },
+    {
+      "epoch": 0.711864406779661,
+      "grad_norm": 0.5668186545372009,
+      "learning_rate": 3.327435087711649e-05,
+      "loss": 1.2705078125,
+      "step": 42
+    },
+    {
+      "epoch": 0.7288135593220338,
+      "grad_norm": 0.6506308913230896,
+      "learning_rate": 3.097794799158444e-05,
+      "loss": 1.35009765625,
+      "step": 43
+    },
+    {
+      "epoch": 0.7457627118644068,
+      "grad_norm": 1.9056968688964844,
+      "learning_rate": 2.8765923439848764e-05,
+      "loss": 1.0797576904296875,
+      "step": 44
+    },
+    {
+      "epoch": 0.7627118644067796,
+      "grad_norm": 3.448936939239502,
+      "learning_rate": 2.6646047038823007e-05,
+      "loss": 1.28173828125,
+      "step": 45
+    },
+    {
+      "epoch": 0.7796610169491526,
+      "grad_norm": 0.5010473728179932,
+      "learning_rate": 2.4625764931680536e-05,
+      "loss": 1.1314711570739746,
+      "step": 46
+    },
+    {
+      "epoch": 0.7966101694915254,
+      "grad_norm": 0.5115064382553101,
+      "learning_rate": 2.271217343300509e-05,
+      "loss": 1.20849609375,
+      "step": 47
+    },
+    {
+      "epoch": 0.8135593220338984,
+      "grad_norm": 0.4829571545124054,
+      "learning_rate": 2.0911994102727095e-05,
+      "loss": 1.28466796875,
+      "step": 48
+    },
+    {
+      "epoch": 0.8135593220338984,
+      "eval_loss": 1.24169921875,
+      "eval_runtime": 8.578,
+      "eval_samples_per_second": 0.583,
+      "eval_steps_per_second": 0.117,
+      "step": 48
+    },
+    {
+      "epoch": 0.8305084745762712,
+      "grad_norm": 0.3891683518886566,
+      "learning_rate": 1.9231550136399102e-05,
+      "loss": 1.206298828125,
+      "step": 49
+    },
+    {
+      "epoch": 0.847457627118644,
+      "grad_norm": 0.39190465211868286,
+      "learning_rate": 1.7676744154740922e-05,
+      "loss": 1.257568359375,
+      "step": 50
+    },
+    {
+      "epoch": 0.864406779661017,
+      "grad_norm": 0.6265923380851746,
+      "learning_rate": 1.6253037470469e-05,
+      "loss": 1.22509765625,
+      "step": 51
+    },
+    {
+      "epoch": 0.8813559322033898,
+      "grad_norm": 1.132917881011963,
+      "learning_rate": 1.4965430905236377e-05,
+      "loss": 1.23193359375,
+      "step": 52
+    },
+    {
+      "epoch": 0.8983050847457628,
+      "grad_norm": 1.4995769262313843,
+      "learning_rate": 1.381844722406417e-05,
+      "loss": 1.137939453125,
+      "step": 53
+    },
+    {
+      "epoch": 0.9152542372881356,
+      "grad_norm": 0.8267571330070496,
+      "learning_rate": 1.281611524896456e-05,
+      "loss": 1.262939453125,
+      "step": 54
+    },
+    {
+      "epoch": 0.9322033898305084,
+      "grad_norm": 0.4488615393638611,
+      "learning_rate": 1.1961955707556493e-05,
+      "loss": 1.243896484375,
+      "step": 55
+    },
+    {
+      "epoch": 0.9491525423728814,
+      "grad_norm": 0.7780895233154297,
+      "learning_rate": 1.1258968866381517e-05,
+      "loss": 1.293701171875,
+      "step": 56
+    },
+    {
+      "epoch": 0.9661016949152542,
+      "grad_norm": 0.5684336423873901,
+      "learning_rate": 1.0709623992358144e-05,
+      "loss": 1.24169921875,
+      "step": 57
+    },
+    {
+      "epoch": 0.9830508474576272,
+      "grad_norm": 1.7781082391738892,
+      "learning_rate": 1.031585067939174e-05,
+      "loss": 1.258056640625,
+      "step": 58
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 40.05562973022461,
+      "learning_rate": 1.007903207060554e-05,
+      "loss": 1.238037109375,
+      "step": 59
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.219970703125,
+      "eval_runtime": 8.5876,
+      "eval_samples_per_second": 0.582,
+      "eval_steps_per_second": 0.116,
+      "step": 59
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 59,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4.0731125942761554e+18,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}