hf-transformers-bot HF Staff
Update tiny models for FunnelForSequenceClassification
b6e8555 verified | { | |
| "activation_dropout": 0.0, | |
| "architectures": [ | |
| "FunnelForSequenceClassification" | |
| ], | |
| "attention_dropout": 0.1, | |
| "attention_type": "relative_shift", | |
| "block_repeats": [ | |
| 1, | |
| 1, | |
| 1 | |
| ], | |
| "block_sizes": [ | |
| 1, | |
| 1, | |
| 2 | |
| ], | |
| "d_head": 8, | |
| "d_inner": 37, | |
| "d_model": 32, | |
| "dtype": "float32", | |
| "hidden_act": "gelu_new", | |
| "hidden_dropout": 0.1, | |
| "initializer_range": 0.1, | |
| "initializer_std": 0.02, | |
| "layer_norm_eps": 1e-09, | |
| "max_position_embeddings": 512, | |
| "model_type": "funnel", | |
| "n_head": 4, | |
| "num_decoder_layers": 1, | |
| "pad_token_id": null, | |
| "pool_q_only": true, | |
| "pooling_type": "mean", | |
| "separate_cls": true, | |
| "tie_word_embeddings": true, | |
| "transformers_version": "5.16.0.dev0", | |
| "truncate_seq": true, | |
| "type_vocab_size": 3, | |
| "vocab_size": 30522 | |
| } | |