| { | |
| "_name_or_path": "Deathsquad10/TinyLlama3-EmptyModel", | |
| "architectures": [ | |
| "LlamaForCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 128000, | |
| "dim": 2048, | |
| "eos_token_id": 128001, | |
| "ffn_dim_multiplier": 1.3, | |
| "head_dim": 128, | |
| "hidden_act": "silu", | |
| "hidden_dim": 5632, | |
| "hidden_size": 2048, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 5632, | |
| "max_position_embeddings": 32768, | |
| "model_type": "llama", | |
| "multiple_of": 256, | |
| "n_heads": 32, | |
| "n_kv_heads": 4, | |
| "n_layers": 22, | |
| "norm_eps": 1e-05, | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 22, | |
| "num_key_value_heads": 4, | |
| "pretraining_tp": 1, | |
| "rms_norm_eps": 1e-05, | |
| "rope_scaling": null, | |
| "rope_theta": 500000.0, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.39.3", | |
| "use_cache": true, | |
| "vocab_size": 128256 | |
| } | |