| { | |
| "architectures": [ | |
| "LlamaForCausalLM" | |
| ], | |
| "model_type": "llama", | |
| "vocab_size": 87, | |
| "hidden_size": 128, | |
| "intermediate_size": 384, | |
| "num_hidden_layers": 4, | |
| "num_attention_heads": 4, | |
| "num_key_value_heads": 4, | |
| "head_dim": 32, | |
| "hidden_act": "silu", | |
| "max_position_embeddings": 512, | |
| "rms_norm_eps": 1e-05, | |
| "rope_theta": 10000.0, | |
| "rope_parameters": { | |
| "rope_type": "default", | |
| "rope_theta": 10000.0 | |
| }, | |
| "tie_word_embeddings": true, | |
| "attention_bias": false, | |
| "mlp_bias": false, | |
| "attention_dropout": 0.0, | |
| "bos_token_id": null, | |
| "eos_token_id": 1, | |
| "pad_token_id": null, | |
| "torch_dtype": "float32" | |
| } | |