+ œ°³j½ãót€Rt^RIt^RIHt^RIHt^RIHt]!RR7]!RR]444tR.t R#) zMAMBA configurationN)Ústrict)ÚPreTrainedConfig)Úauto_docstringzstate-spaces/mamba-2.8b)Ú checkpointcó¾aa€]tRt^toRtRtRtRt^t^ t Rt ^t ^t ^t ^t^tRtRtRtR tRtR tR tR tR tR tRtRtRtRtRtRtV3Rlt ]!R4t"V3Rlt#Rt$Vt%V;t&#)Ú MambaConfiga¾ layer_norm_epsilon (`float`, *optional*, defaults to 1e-05): The epsilon to use in the layer normalization layers. expand (`int`, *optional*, defaults to 2): Expanding factor used to determine the intermediate size. use_bias (`bool`, *optional*, defaults to `False`): Whether or not to use bias in ["in_proj", "out_proj"] of the mixer block use_conv_bias (`bool`, *optional*, defaults to `True`): Whether or not to use bias in the convolution layer of the mixer block. residual_in_fp32 (`bool`, *optional*, defaults to `True`): Whether or not residuals should be in `float32`. If set to `False` residuals will keep the same `dtype` as the rest of the model rescale_prenorm_residual (`bool`, *optional*, defaults to `False`): Whether or not to rescale `out_proj` weights when initializing. use_mambapy (`bool`, *optional*, defaults to `False`): Determines the fallback strategy during training if the CUDA-based official implementation of Mamba is not available. If `True`, the mamba.py implementation is used. If `False`, the naive and slower implementation is used. Consider switching to the naive version if memory is limited. use_associative_scan (`bool`, *optional*, defaults to `True`): Whether to use PyTorch's `torch._higher_order_ops.associative_scan` for the parallel scan instead of the naive sequential implementation. The associative scan is only active during `torch.compile` tracing and requires torch >= 2.9.0. Both paths are tested to produce numerically identical results (see `test_associative_scan_matches_sequential`). Set to `False` to fall back to the sequential loop. Example: ```python >>> from transformers import MambaConfig, MambaModel >>> # Initializing a Mamba configuration >>> configuration = MambaConfig() >>> # Initializing a model (with random weights) from the configuration >>> model = MambaModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```ÚmambaihÄigñh㈵øä>FTÚsilugš™™™™™¹?Úautogð?gü©ñÒMbP?Úrandomg-Cëâ6?c ó<€\VPVP,4VnVPR8Xd(\ P !VP^, 4M VPVn\SV` !R/VBR#)r N©) ÚintÚexpandÚ hidden_sizeÚintermediate_sizeÚtime_step_rankÚmathÚceilÚsuperÚ __post_init__)ÚselfÚkwargsÚ __class__s&,€Ún/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/mamba/configuration_mamba.pyrÚMambaConfig.__post_init__^sfø€Ü!$ T§[¡[°4×3CÑ3CÕ%CÓ!DˆÔà04×0CÑ0CÀvÔ0MŒDIŠId×&Ñ&¨Õ+Ô ,ÐSW×SfÑSfð Ôô ‰ÒÑ' Ô'ócó*€R.VP,#)Úlinear_attention)Únum_hidden_layers)rs&rÚ layer_typesÚMambaConfig.layer_typeses€à"Ð# d×&<Ñ&<Õ<ÐrQsMðñã å.å3Ý#ñÐ4Ô5ØôM=Ð"óM=óó6ðM=ð` ˆ/‚r