| { | |
| "_name_or_path": "budecosystem/boomer-bitnet-634m", | |
| "architectures": [ | |
| "BoomerForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_boomer.BoomerConfig", | |
| "AutoModelForCausalLM": "modelling_boomer.BoomerForCausalLM" | |
| }, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 1024, | |
| "initializer_range": 0.02, | |
| "input_bits": 8, | |
| "intermediate_size": 2560, | |
| "max_position_embeddings": 4096, | |
| "model_type": "boomer", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 48, | |
| "num_key_value_heads": 4, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-06, | |
| "rope_scaling": null, | |
| "rope_theta": 1000000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.39.2", | |
| "use_cache": false, | |
| "vocab_size": 64000, | |
| "weight_bits": 1 | |
| } | |