| { | |
| "architectures": [ | |
| "WednesdayModel" | |
| ], | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "hidden_act": "gelu_pytorch_tanh", | |
| "hidden_size": 768, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 3072, | |
| "max_position_embeddings": 32768, | |
| "model_type": "mixtral", | |
| "moe_2layer_gate": true, | |
| "moe_expert_interval": 1, | |
| "moe_feature_no_mul_topk": true, | |
| "moe_gate_norm_std": 1.0, | |
| "moe_use_logits_norm": true, | |
| "moe_use_mixtral_gating": true, | |
| "num_attention_heads": 12, | |
| "num_experts": [ | |
| 8 | |
| ], | |
| "num_hidden_layers": 4, | |
| "num_key_value_heads": 2, | |
| "pad_token_id": 0, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 1000000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.53.1", | |
| "use_cache": true, | |
| "vocab_size": 262400 | |
| } | |