config.json

{
  "architectures": [
    "Qwen3ForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "chat_template": "{%- set last_message = messages[-1] -%}\n{%- if last_message.role == \"user\" -%}\n{{- '<|problem_start|>\n' + last_message.content + '<|problem_end|>\n' -}}\n{%- elif last_message.role == \"assistant\" -%}\n{%- for message in messages -%}\n{%- if message.role == \"user\" -%}\n{{- '<|problem_start|>\n' + message.content + '<|problem_end|>\n' -}}\n{%- elif message.role == \"assistant\" -%}\n{{- '<|formatted_problem_start|>\n' + message.content + '<|formatted_problem_end|>\n' -}}\n{%- else -%}\n{{- raise('Unknown role: ' + message.role) -}}\n{%- endif -%}\n{%- endfor -%}\n{%- else -%}\n{{- raise('Unsupported role: ' + last_message.role) -}}\n{%- endif -%}\n{%- if add_generation_prompt and last_message.role == \"user\" -%}\n{{- '<|formatted_problem_start|>\n' -}}\n{%- endif -%}",
  "eos_token_id": 151643,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 2048,
  "initializer_range": 0.02,
  "intermediate_size": 6144,
  "max_position_embeddings": 32768,
  "max_window_layers": 28,
  "model_type": "qwen3",
  "num_attention_heads": 16,
  "num_hidden_layers": 28,
  "num_key_value_heads": 8,
  "pad_token_id": 151654,
  "rms_norm_eps": 1e-06,
  "rope_scaling": null,
  "rope_theta": 1000000,
  "sliding_window": null,
  "tie_word_embeddings": true,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.51.3",
  "unsloth_fixed": true,
  "unsloth_version": "2025.5.5",
  "use_cache": true,
  "use_sliding_window": false,
  "vocab_size": 151673
}