config.json
| 1 | { |
| 2 | "architectures": [ |
| 3 | "Qwen2ForCausalLM" |
| 4 | ], |
| 5 | "attention_dropout": 0.0, |
| 6 | "dtype": "bfloat16", |
| 7 | "hidden_act": "silu", |
| 8 | "hidden_size": 8, |
| 9 | "initializer_range": 0.02, |
| 10 | "intermediate_size": 32, |
| 11 | "layer_types": [ |
| 12 | "full_attention", |
| 13 | "full_attention" |
| 14 | ], |
| 15 | "max_position_embeddings": 32768, |
| 16 | "max_window_layers": 28, |
| 17 | "model_type": "qwen2", |
| 18 | "num_attention_heads": 4, |
| 19 | "num_hidden_layers": 2, |
| 20 | "num_key_value_heads": 2, |
| 21 | "rms_norm_eps": 1e-06, |
| 22 | "rope_scaling": null, |
| 23 | "rope_theta": 10000.0, |
| 24 | "sliding_window": null, |
| 25 | "tie_word_embeddings": false, |
| 26 | "transformers_version": "4.57.3", |
| 27 | "use_cache": true, |
| 28 | "use_sliding_window": false, |
| 29 | "vocab_size": 151665 |
| 30 | } |
| 31 | |