| { | |
| "_name_or_path": "activation_beacon", | |
| "architectures": [ | |
| "LlamaForCausalLM" | |
| ], | |
| "auto_map": { | |
| "AutoConfig": "configuration_llama.LlamaConfig", | |
| "AutoModelForCausalLM": "modeling_llama.LlamaForCausalLM" | |
| }, | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "beacon_attn": "step-expansion", | |
| "beacon_layers": null, | |
| "beacon_param": ["q", "k", "v", "o"], | |
| "beacon_ratio": [2,4,8,16,32,64,128], | |
| "beacon_ratio_mix": "adapt-1024", | |
| "beacon_stride": [1024], | |
| "beacon_stride_mix": "step-random", | |
| "beacon_window": 1024, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 4096, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 11008, | |
| "max_position_embeddings": 4096, | |
| "model_type": "llama", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 32, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.36.1", | |
| "use_cache": true, | |
| "vocab_size": 32000 | |
| } | |