| { | |
| "actual_seq_length": 512, | |
| "architectures": [ | |
| "FNetForPreTraining" | |
| ], | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "hidden_act": "gelu_new", | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 1024, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 4096, | |
| "layer_norm_eps": 1e-12, | |
| "max_position_embeddings": 512, | |
| "model_type": "fnet", | |
| "num_hidden_layers": 24, | |
| "pad_token_id": 3, | |
| "torch_dtype": "float32", | |
| "tpu_short_seq_length": 512, | |
| "transformers_version": "4.10.0.dev0", | |
| "type_vocab_size": 4, | |
| "use_latest": false, | |
| "use_fft": true, | |
| "use_tpu_fourier_optimizations": false, | |
| "vocab_size": 32000 | |
| } | |