| { | |
| "architectures": [ | |
| "GraniteMoeForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.1, | |
| "attention_multiplier": 0.015625, | |
| "bos_token_id": 0, | |
| "embedding_multiplier": 12.0, | |
| "eos_token_id": 0, | |
| "hidden_act": "silu", | |
| "hidden_size": 1536, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 512, | |
| "logits_scaling": 6.0, | |
| "max_position_embeddings": 131072, | |
| "model_type": "granitemoe", | |
| "num_attention_heads": 24, | |
| "num_experts_per_tok": 8, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 8, | |
| "num_local_experts": 40, | |
| "output_router_logits": false, | |
| "pad_token_id": 0, | |
| "residual_multiplier": 0.22, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 10000000.0, | |
| "router_aux_loss_coef": 0.001, | |
| "tie_word_embeddings": true, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.46.0", | |
| "use_cache": true, | |
| "vocab_size": 49152 | |
| } | |