| { | |
| "transformers_version": "5.8.0", | |
| "architectures": [], | |
| "output_hidden_states": false, | |
| "return_dict": true, | |
| "dtype": null, | |
| "chunk_size_feed_forward": 0, | |
| "is_encoder_decoder": false, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1" | |
| }, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1 | |
| }, | |
| "problem_type": null, | |
| "vocab_size": 1000, | |
| "hidden_size": 256, | |
| "intermediate_size": 1024, | |
| "num_hidden_layers": 4, | |
| "num_attention_heads": 4, | |
| "num_key_value_heads": 4, | |
| "hidden_act": "silu", | |
| "max_position_embeddings": 128, | |
| "initializer_range": 0.02, | |
| "rms_norm_eps": 1e-06, | |
| "use_cache": true, | |
| "tie_word_embeddings": false, | |
| "rope_parameters": { | |
| "rope_theta": 10000.0, | |
| "rope_type": "default" | |
| }, | |
| "use_sliding_window": false, | |
| "sliding_window": null, | |
| "max_window_layers": 28, | |
| "layer_types": [ | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention" | |
| ], | |
| "attention_dropout": 0.0, | |
| "pad_token_id": null, | |
| "bos_token_id": null, | |
| "eos_token_id": null, | |
| "_name_or_path": "", | |
| "model_type": "qwen2", | |
| "output_attentions": false | |
| } |