+ j-Rt^RIHt^RIHt^RIHt^RIHtH t ] P!] 4t ]!RR7]!RR]444t ]!RR7]!R R ]444t]!RR7]!R R ]444t]!RR7]!R R]444t]!RR7]!RR]444t.ROtR#)zBlt model configuration)strict)PreTrainedConfig)RopeParameters)auto_docstringloggingzitazap/blt-1b-hf) checkpointcaa]tRt^toRtRtRtRtRt^t Rt Rt ^t Rt ^tR tR tR tRtR tRtR tV3RltV3RltRtVtV;t#)BltLocalEncoderConfigF cross_attn_all_layers (`bool`, *optional*, defaults to `True`): Whether all attention layers have cross attention. cross_attn_k (`int`, *optional*, defaults to 2): Number of cross-attention heads used in the model. hidden_size_global (`int`, *int*, defaults to 2048): Hidden size of the global transformer layer. blt_local_encoderAFNh㈵>`silu{Gz?c <VP;'g VPVnVP;'g$\^VP,^, 4VnRVn\ SV`!R/VBR#FN)num_key_value_headsnum_attention_headsintermediate_sizeint hidden_sizetie_word_embeddingssuper __post_init__selfkwargs __class__s&,j/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/blt/configuration_blt.pyr #BltLocalEncoderConfig.__post_init__9sc#'#;#;#W#Wt?W?W !%!7!7!X!X3q4CSCS?SVW?W;X#(  ''cP<V^8dQh/S[;R&S[R,;R&S[R,;R&S[R,;R&S[;R&S[;R&S[R,;R&S[;R &S[;R &S[S[,R,;R &S[;R &S[S[,R,;R &S[;R&S[R,;R&S[;R&#) vocab_sizeNcross_attn_all_layers cross_attn_khidden_size_globalrrrnum_hidden_layers rms_norm_epsdropoutmax_position_embeddingsrope_parameters hidden_actrinitializer_range)rboolfloatrdictstr)format __classdict__s"r% __annotate__"BltLocalEncoderConfig.__annotate__s  $;.!"* #$d )%&'(!)*t*+,-./0S[4 %12!(34$d*T185678Tz(9:#;r')rrr)__name__ __module__ __qualname____firstlineno____doc__ model_type default_thetar*r+r,r-rrrr.r/r0r1r2r3rr4r __annotate_func____static_attributes____classdictcell__ __classcell__r$r:s@@r%r r su%JMJ). L%)K!&*L"%G#(48OJ$(#(?r'r caa]tRt^@toRtRtRtRtRt^t Rt Rt ^t Rt ^ tR tR tR tRtR tR tRtRtRtRtRtV3RltV3RltRtVtV;t#)BltLocalDecoderConfigr blt_local_decoderr r TrrNrrrri rFc <<VP;'g VPVnVPVP,VnVP;'g$\ ^VP,^, 4VnRVn\SV` !R/VBR#r rrrhead_dimrrrrr r!s&,r%r #BltLocalDecoderConfig.__post_init__cs{#'#;#;#W#Wt?W?W ((D,D,DD !%!7!7!X!X3q4CSCS?SVW?W;X#(  ''r'c<V^8dQh/S[;R&S[R,;R&S[R,;R&S[R,;R&S[;R&S[;R&S[R,;R&S[;R &S[;R &S[S[,R,;R &S[;R &S[S[,R,;R &S[;R&S[;R&S[;R&S[R,;R&S[R,;R&S[S[S[,,R,;R&S[;R&#)r)r*Nr+r,r-rrrr.r/r0r1r2r3rr4 pad_token_id bos_token_id eos_token_idr)rr5r6rr7r8list)r9r:s"r%r;"BltLocalDecoderConfig.__annotate__@s8  $;-!"* #$d )%&'(!)*t*+,-./0S[4 %12!(34$d*T185678!9:#;<*#=>*#?@S /D(/AB%Cr'rNrrr)r=r>r?r@rArBrCr*r+r,r-rrrr.r/r0r1r2r3rr4rQrRrSrr rDrErFrGrHs@@r%rJrJ@s%JMJ)- L%)K!&*L"%G#(48OJ!##L#L+/L %(Gr'rJcvaa]tRt^ktoRtRtRt^tRt^t Rt Rt Rt Rt RtR tR tR tV3R ltV3R ltRtVtV;t#)BltGlobalTransformerConfigblt_global_transformerr rNrrrirFc <<VP;'g VPVnVPVP,VnVP;'g$\ ^VP,^, 4VnRVn\SV` !R/VBR#rrMr!s&,r%r (BltGlobalTransformerConfig.__post_init__~s{#'#;#;#W#Wt?W?W ((D,D,DD !%!7!7!X!X3q4CSCS?SVW?W;X#(  ''r'c<V^8dQh/S[;R&S[;R&S[R,;R&S[;R&S[;R&S[S[,R,;R&S[;R&S[S[,R,;R &S[;R &S[;R &S[;R &S[;R &#)r)rrNrr.r/r0r1r2r3rr4r)rr6rr7r8r5)r9r:s"r%r;'BltGlobalTransformerConfig.__annotate__ks  !t*S[4 %!'$d*T18! #!"%#r'rV)r=r>r?r@rBrCrrrr.r/r0r1r2r3rr4rr rDrErFrGrHs@@r%rXrXks`*JMK!&*L"%G#'48OJ!# %('r'rXcraa]tRt^toRtRtRt^t^ tRt Rt Rt Rt Rt RtR tR tV3R ltV3R ltR tVtV;t#)BltPatcherConfig blt_patcherr iNi rrrrFc J<VP;'g VPVnVPVP,VnVP;'g$\ ^VP,^, 4VnRVnRVn\SV`$!R/VBR#)rFrNr) rrrrNrrrr3rr r!s&,r%r BltPatcherConfig.__post_init__s#'#;#;#W#Wt?W?W ((D,D,DD !%!7!7!X!X3q4CSCS?SVW?W;X#(   ''r'c<V^8dQh/S[;R&S[;R&S[;R&S[;R&S[R,;R&S[;R&S[;R&S[S[,R,;R &S[;R &S[S[,R,;R &S[;R &S[;R &#)r)r*rr.rNrr1r/r0rr2r4r)rr6rr7r5)r9r:s"r%r;BltPatcherConfig.__annotate__s    !t*!'S[4 %!$d*T18# %!r')rNr3rrr)r=r>r?r@rBr*rr.rrr1r/r0rr2r4rr rDrErFrGrHs@@r%r`r`s[JJK!&*#'L"%G!48O# %(%r'r`caa]tRt^toRtRtR.tRtR]R] R] R] /t R t R tR t^tR tR t^tRt^tRtRt^tRtRtRtRtRtR tRtRt Rt!Rt"Rt#V3Rlt$V3Rlt%Rt&Vt'V;t(#) BltConfiga patch_in_forward (`bool`, *optional*, defaults to `True`): Whether to perform patching during the forward pass. patch_size (`int`, *optional*, defaults to 4): Size of the patches used in the patching mechanism. patching_mode (`str`, *optional*, defaults to `"entropy"`): The mode used for patching, such as entropy-based patching. patching_threshold (`float`, *optional*, defaults to 1.34): Threshold value used for determining when to apply patches. patching_batch_size (`int`, *optional*, defaults to 1): Batch size used during the patching process. max_patch_length (`int`, *optional*): Maximum length of patches that can be generated. cross_attn_k (`int`, *optional*, defaults to 2): Number of cross-attention heads used in the model. encoder_hash_byte_group_size (`list`, *optional*): List of byte group sizes used in the encoder hash function. encoder_hash_byte_group_vocab (`int`, *optional*, defaults to 500002): Vocabulary size for the encoder hash byte groups. encoder_hash_byte_group_nb_functions (`int`, *optional*, defaults to 1): Number of hash functions used in the encoder byte grouping. patcher_config (`BltPatcherConfig`, *optional*): Configuration for the patcher component of the model. global_config (`BltGlobalTransformerConfig`, *optional*): Configuration for the global transformer component of the model. Example: ```python >>> from transformers import BltModel, BltConfig >>> # Initializing a Blt configuration >>> configuration = BltConfig() >>> # Initializing a model from the configuration >>> model = BltModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```bltpast_key_valuesr patcher_configencoder_configdecoder_config global_configr rZTentropyg]?Ni"Frc <VP;'g.ROVnVPf2\VPR7Vn\P R4M`\ VP\4'dAVPPRVP4\R /VPBVnVPf2\VPR7Vn \P R4M`\ VP\4'dAVPPRVP4\R /VPBVn VPf2\VPR7Vn \P R4M`\ VP\4'dAVPPRVP4\R /VPBVn VPf2\VPR7Vn \P R4M`\ VP\4'dAVPPRVP4\R /VPBVn VPPVP ,pW PP8wdTMRVPn\$SV`L!R /VBR#) N)r4z8patcher_config is None, using default Blt patcher configr4z8encoder_config is None, using default Blt encoder configz8decoder_config is None, using default Blt decoder configz6global_config is None, using default Blt global config)rprr)encoder_hash_byte_group_sizerjr`r4loggerinfo isinstancer7 setdefaultrkr rlrJrmrXrr,encoder_cross_output_sizerr )r"r#rzr$s&, r%r BltConfig.__post_init__s;,0,M,M,c,cQc)    &"2TE[E["\D  KKR S ++T 2 2    * *+>@V@V W"2"IT5H5H"ID     &"7$J`J`"aD  KKR S ++T 2 2    * *+>@V@V W"7"N$:M:M"ND     &"7$J`J`"aD  KKR S ++T 2 2    * *+>@V@V W"7"N$:M:M"ND     %!;dNdNd!eD  KKP Q **D 1 1    ) )*=t?U?U V!;!Qd>P>P!QD %)$7$7$C$CdFWFW$W!)BFXFXFdFd)d %jn 4 ''r'c<V^8dQh/S[;R&S[;R&S[R,;R&S[R,;R&S[R,;R&S[R,;R&S[R,;R&S[R,;R &S[R,;R &S[S[,R,;R &S[R,;R &S[R,;R &S[S[,R,;R&S[S[,R,;R&S[S[,R,;R&S[S[,R,;R&S[;R&S[;R&S[R,;R&S[R,;R&S[S[S[,,R,;R&S[;R&S[S[,R,;R&#)r)r*r1Npatch_in_forward patch_size patching_modepatching_thresholdpatching_batch_sizemax_patch_lengthr,ruencoder_hash_byte_group_vocab$encoder_hash_byte_group_nb_functionsrjrkrlrmr use_cacherQrRrSr4r2)rr5r8r6rTr7rr)r9r:s"r%r;BltConfig.__annotate__sjkl!'mnTk(opd qr:)st 8uvt'wxDj'yz* {|#'s)d"29}~$':6@+.*8AB++d29CD++d29EF++d29GH**T18IJ%KLMN*#OP*#QRS /D(/ST#UV$d*T18Wr')rlrkrurmrj))r=r>r?r@rArBkeys_to_ignore_at_inferencerCr`r rJrX sub_configsr*r1r}r~rrrrr,rurrrjrkrlrmrrrQrRrSr4r2r rDrErFrGrHs@@r%rgrgs& PJ#4"5M*//3 KJ#'$(J )M'8&'#' L59 06!78(59N59N59N48M %I#L#L+/L#48O&([r'rgN)rgr`r rJrX)rAhuggingface_hub.dataclassesrconfiguration_utilsrmodeling_rope_utilsrutilsrr get_loggerr=rvr rJrXr`rg__all__rr'r%rs.31,   H %-.!(,!(/!(H-.&(,&(/&(R-.(!1(/(4-.('(/(4-.q( q(/q(h r'