| { | |
| "architectures": [ | |
| "LlamaForCausalLMEagle3" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 100000, | |
| "eos_token_id": 100001, | |
| "head_dim": 128, | |
| "hidden_act": "silu", | |
| "hidden_size": 2048, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 10944, | |
| "max_position_embeddings": 163840, | |
| "max_window_layers": 64, | |
| "model_type": "llama", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 1, | |
| "num_key_value_heads": 16, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": { | |
| "beta_fast": 32.0, | |
| "beta_slow": 1.0, | |
| "factor": 40.0, | |
| "mscale": 0.707, | |
| "mscale_all_dim": 0.707, | |
| "original_max_position_embeddings": 4096, | |
| "rope_type": "yarn" | |
| }, | |
| "rope_theta": 10000, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.33.1", | |
| "use_cache": true, | |
| "use_sliding_window": false, | |
| "vocab_size": 102400, | |
| "draft_vocab_size": 32000 | |
| } | |