| { | |
| "activation": "gelu", | |
| "alibi": false, | |
| "architectures": [ | |
| "FalconForSequenceClassification" | |
| ], | |
| "attention_dropout": 0.0, | |
| "bias": false, | |
| "bos_token_id": 1, | |
| "dtype": "float32", | |
| "eos_token_id": 11, | |
| "ffn_hidden_size": 128, | |
| "hidden_dropout": 0.0, | |
| "hidden_size": 32, | |
| "initializer_range": 0.02, | |
| "layer_norm_epsilon": 1e-05, | |
| "max_position_embeddings": 512, | |
| "model_type": "falcon", | |
| "multi_query": true, | |
| "new_decoder_architecture": true, | |
| "num_attention_heads": 2, | |
| "num_hidden_layers": 2, | |
| "num_kv_heads": 2, | |
| "num_ln_in_parallel_attn": 2, | |
| "pad_token_id": 0, | |
| "parallel_attn": true, | |
| "rope_parameters": { | |
| "rope_theta": 10000.0, | |
| "rope_type": "default" | |
| }, | |
| "tie_word_embeddings": false, | |
| "transformers_version": "5.16.0.dev0", | |
| "use_cache": true, | |
| "vocab_size": 65024 | |
| } | |