| { | |
| "architectures": [ | |
| "HyperLlamaForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_hyperllama.HyperLlamaConfig", | |
| "AutoModelForCausalLM": "modeling_hyperllama.HyperLlamaForCausalLM" | |
| }, | |
| "bos_token_id": 0, | |
| "eos_token_id": 0, | |
| "head_dim": 64, | |
| "hidden_act": "gelu", | |
| "hidden_size": 1152, | |
| "initializer_range": 0.041666666666666664, | |
| "intermediate_size": 3072, | |
| "lm_head_normalization_factor": 2, | |
| "max_position_embeddings": 2048, | |
| "mlp_bias": false, | |
| "model_type": "hyperllama", | |
| "num_attention_heads": 18, | |
| "num_hidden_layers": 30, | |
| "num_key_value_heads": 6, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": true, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.51.3", | |
| "use_cache": true, | |
| "vocab_size": 128256 | |
| } | |