| { |
| "_name_or_path": ".\\BlackSheep-MoE", |
| "architectures": [ |
| "MixtralForCausalLM" |
| ], |
| "attention_dropout": 0.0, |
| "bos_token_id": null, |
| "eos_token_id": 151645, |
| "hidden_act": "silu", |
| "hidden_size": 3584, |
| "initializer_range": 0.02, |
| "intermediate_size": 18944, |
| "max_position_embeddings": 32768, |
| "max_window_layers": 28, |
| "model_type": "mixtral", |
| "num_attention_heads": 28, |
| "num_experts_per_tok": 1, |
| "num_hidden_layers": 20, |
| "num_key_value_heads": 4, |
| "num_local_experts": 2, |
| "output_router_logits": false, |
| "rms_norm_eps": 1e-06, |
| "rope_theta": 1000000.0, |
| "router_aux_loss_coef": 0.001, |
| "router_jitter_noise": 0.0, |
| "sliding_window": null, |
| "tie_word_embeddings": false, |
| "torch_dtype": "float16", |
| "transformers_version": "4.41.2", |
| "use_cache": false, |
| "use_sliding_window": false, |
| "vocab_size": 152064 |
| } |
|
|