| { |
| "_name_or_path": "/data/1_pretrained_models/Qwen2-7B-Instruct/", |
| "architectures": [ |
| "Qwen2ForSequenceClassification" |
| ], |
| "attention_dropout": 0.0, |
| "bos_token_id": 151643, |
| "eos_token_id": 151645, |
| "hidden_act": "silu", |
| "hidden_size": 3584, |
| "id2label": { |
| "0": "LABEL_0" |
| }, |
| "initializer_range": 0.02, |
| "intermediate_size": 18944, |
| "label2id": { |
| "LABEL_0": 0 |
| }, |
| "max_position_embeddings": 32768, |
| "max_window_layers": 28, |
| "model_type": "qwen2", |
| "num_attention_heads": 28, |
| "num_hidden_layers": 28, |
| "num_key_value_heads": 4, |
| "pad_token_id": 151646, |
| "rms_norm_eps": 1e-06, |
| "rope_theta": 1000000.0, |
| "sliding_window": 131072, |
| "tie_word_embeddings": false, |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.41.2", |
| "use_cache": false, |
| "use_sliding_window": false, |
| "vocab_size": 151647 |
| } |
|
|
