| { |
| "architectures": [ |
| "Qwen2ParScaleForCausalLM" |
| ], |
| "auto_map": { |
| "AutoConfig": "configuration_qwen2_parscale.Qwen2ParScaleConfig", |
| "AutoModelForCausalLM": "modeling_qwen2_parscale.Qwen2ParScaleForCausalLM" |
| }, |
| "attention_dropout": 0.0, |
| "bos_token_id": 151643, |
| "eos_token_id": 151643, |
| "hidden_act": "silu", |
| "hidden_size": 2048, |
| "initializer_range": 0.02, |
| "intermediate_size": 11008, |
| "max_position_embeddings": 2048, |
| "max_window_layers": 36, |
| "model_type": "qwen2_parscale", |
| "num_attention_heads": 16, |
| "num_hidden_layers": 36, |
| "num_key_value_heads": 2, |
| "parscale_attn_smooth": 0.1, |
| "parscale_n": 2, |
| "parscale_n_tokens": 48, |
| "rms_norm_eps": 1e-06, |
| "rope_scaling": null, |
| "rope_theta": 10000.0, |
| "sliding_window": 32768, |
| "tie_word_embeddings": true, |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.46.2", |
| "use_cache": true, |
| "use_sliding_window": false, |
| "vocab_size": 151936 |
| } |