| { | |
| "activation": "gelu", | |
| "alibi": false, | |
| "architectures": [ | |
| "FalconForCausalLM" | |
| ], | |
| "attention_dropout": 0.05, | |
| "bias": true, | |
| "bos_token_id": 11, | |
| "eos_token_id": 11, | |
| "ffn_hidden_size": 2048, | |
| "hidden_dropout": 0.05, | |
| "hidden_size": 512, | |
| "initializer_range": 0.02, | |
| "layer_norm_epsilon": 1e-05, | |
| "max_position_embeddings": 2048, | |
| "model_type": "falcon", | |
| "multi_query": true, | |
| "new_decoder_architecture": false, | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 12, | |
| "num_kv_heads": 16, | |
| "num_ln_in_parallel_attn": null, | |
| "parallel_attn": false, | |
| "rope_scaling": null, | |
| "rope_theta": 10000.0, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.41.2", | |
| "use_cache": true, | |
| "vocab_size": 15001 | |
| } | |