| { |
| "_name_or_path": "unsloth/Qwen2.5-3B-Instruct", |
| "architectures": [ |
| "Qwen2ForCausalLM" |
| ], |
| "attention_dropout": 0.0, |
| "bos_token_id": 151643, |
| "eos_token_id": 151645, |
| "hidden_act": "silu", |
| "hidden_size": 2048, |
| "initializer_range": 0.02, |
| "intermediate_size": 11008, |
| "max_position_embeddings": 45048, |
| "max_window_layers": 70, |
| "model_type": "qwen2", |
| "num_attention_heads": 16, |
| "num_hidden_layers": 36, |
| "num_key_value_heads": 2, |
| "pad_token_id": 151665, |
| "rms_norm_eps": 1e-06, |
| "rope_scaling": { |
| "factor": 1.374755859375, |
| "rope_type": "linear", |
| "type": "linear" |
| }, |
| "rope_theta": 1000000.0, |
| "sliding_window": null, |
| "tie_word_embeddings": true, |
| "torch_dtype": "float16", |
| "transformers_version": "4.47.1", |
| "unsloth_fixed": true, |
| "unsloth_version": "2025.1.7", |
| "use_cache": true, |
| "use_sliding_window": false, |
| "vocab_size": 151936 |
| } |
|
|
