Zj dSrSSKJr SSKJr SSKJr \"SS9\"SS \555rS /rg ) zFNet model configuration)strict)PreTrainedConfig)auto_docstringzgoogle/fnet-base) checkpointc"\rSrSr%SrSrSr\\S'Sr \\S'Sr \\S 'S r \\S 'S r \ \S 'Sr\\-\S'Sr\\S'Sr\\S'Sr\\S'Sr\\S'Sr\\S'Sr\\S'Sr\S-\S'Sr\S-\S'S r\\\-S-\S!'S"r\\S#'S$rg)% FNetConfiga use_tpu_fourier_optimizations (`bool`, *optional*, defaults to `False`): Determines whether to use TPU optimized FFTs. If `True`, the model will favor axis-wise FFTs transforms. Set to `False` for GPU/CPU hardware, in which case n-dimensional FFTs are used. tpu_short_seq_length (`int`, *optional*, defaults to 512): The sequence length that is expected by the model when using TPUs. This will be used to initialize the DFT matrix only when *use_tpu_fourier_optimizations* is set to `True` and the input sequence is shorter than or equal to 4096 tokens. Example: ```python >>> from transformers import FNetConfig, FNetModel >>> # Initializing a FNet fnet-base style configuration >>> configuration = FNetConfig() >>> # Initializing a model (with random weights) from the fnet-base style configuration >>> model = FNetModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```fneti} vocab_sizei hidden_size num_hidden_layersi intermediate_sizegelu_new hidden_actg?hidden_dropout_probimax_position_embeddingstype_vocab_sizeg{Gz?initializer_rangeg-q=layer_norm_epsFuse_tpu_fourier_optimizationstpu_short_seq_lengthrN pad_token_id bos_token_id eos_token_idTtie_word_embeddings)__name__ __module__ __qualname____firstlineno____doc__ model_typer int__annotations__r rrrstrrfloatrrrrrboolrrrrlistr __static_attributes__r!|/root/GenerationalWealth/GenerationalWealth/venv/lib/python3.13/site-packages/transformers/models/fnet/configuration_fnet.pyr r s 0JJKs!s! J '**#&S&OS#u#!NE!*/!4/ ### L#*  L#* +,L#S /D(, $$r/r N) r&huggingface_hub.dataclassesrconfiguration_utilsrutilsrr __all__r!r/r0r5sG.3#-.*%!*%/*%Z .r/