ZjTSSKJr SSKJr SSKrSSKJr SSKJr SSKJ r SSK J r J r SS K Jr SS KJrJrJr SS KJr SS KJrJrJrJr SS KJrJr SSKJrJr SSK J!r!J"r" SSK#J$r$ SSK%J&r&J'r' SSK(J)r)J*r* SSK+J,r, SSK-J.r. "SS\R^5r0\"S5"SS\R^55r1"SS\R^5r2Sr3\"S5S@Sj5r4S \RjS!\6S"\Rj4S#jr7SAS$\R^S%\RjS&\RjS'\RjS(\RjS-S)\8S*\8S+\$\&4S,jjr9\"\45"S-S.\R^55r:"S/S0\5r;\"S1S2\"55r<\"S3S4\<55r=\"S5S69"S7S8\<\55r>\"S5S69"S9S:\\<55r?\"S5S69"S;S<\\<55r@\"S5S69"S=S>\\<55rA/S?QrBg)B)Callable)OptionalN)nn)auto_docstring)ACT2FN)Cache DynamicCache)GenerationMixin)use_kernel_forward_from_hubuse_kernel_func_from_hubuse_kernelized_func)create_causal_mask)GenericForQuestionAnswering GenericForSequenceClassificationGenericForTokenClassificationGradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)ALL_ATTENTION_FUNCTIONSPreTrainedModel)Unpack)TransformersKwargscan_return_tuple)maybe_autocastmerge_with_config_defaults)capture_outputs) ArceeConfigc.^\rSrSrU4SjrSrSrU=r$)ArceeMLP2c>[TU]5 XlURUlURUl[ R "URURURS9Ul[ R "URURURS9Ul [URUl g)Nbias) super__init__config hidden_sizeintermediate_sizerLinearmlp_biasup_proj down_projr hidden_actact_fnselfr* __class__s y/root/GenerationalWealth/GenerationalWealth/venv/lib/python3.13/site-packages/transformers/models/arcee/modeling_arcee.pyr)ArceeMLP.__init__3s  !--!'!9!9yy!1!143I3IPVP_P_` 4#9#94;K;KRXRaRabV../ c`URURURU555$N)r0r2r/)r4xs r6forwardArceeMLP.forward<s"~~dkk$,,q/:;;r8)r2r*r0r+r,r/)__name__ __module__ __qualname____firstlineno__r)r<__static_attributes__ __classcell__r5s@r6r#r#2s0<[TU]5 [R"[R "U55UlX lg)z+ ArceeRMSNorm is equivalent to T5LayerNorm N)r(r)r Parametertorchonesweightvariance_epsilon)r4r+rIr5s r6r)ArceeRMSNorm.__init__Bs/ ll5::k#:; #r8 hidden_statescURnUR[R5nUR S5R SSS9nU[R "X0R-5-nURURU5-$)NT)keepdim) dtypetorMfloat32powmeanrsqrtrPrO)r4rR input_dtypevariances r6r<ArceeRMSNorm.forwardJsw#)) %((7  $$Q',,R,>% H?T?T4T(UU {{]--k:::r8c^[URR5SUR3$)Nz, eps=)tuplerOshaperP)r4s r6 extra_reprArceeRMSNorm.extra_reprQs* ))*+6$2G2G1HIIr8)rPrO)gư>) r>r?r@rAfloatr)rMTensorr<rcrBrCrDs@r6rGrG@sB$$$$$;U\\;ell;JJr8rGc ^\rSrSr%\R \S'SS\4U4Sjjjr\ SS\S-S\ SS\ S-S \ S \ 44S jj5r\R"5\S 55rS rU=r$)ArceeRotaryEmbeddingUinv_freqNr*c>[TU]5 URUlURUlXlUR R SUlURnURS:wa[URnU"UR U5uo@l URSUSS9 URSUR5SS9 g)N rope_typedefaultrjF) persistentoriginal_inv_freq) r(r)max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenr*rope_parametersrlcompute_default_rope_parametersrattention_scalingregister_bufferclone)r4r*device rope_init_fnrjr5s r6r)ArceeRotaryEmbedding.__init__Xs "("@"@$*$B$B! 44[A!%!E!E >>Y &.t~~>L+7 V+L(( ZeD 0(..2BuUr8rxz torch.deviceseq_lenrJz torch.Tensorc URSn[USS5=(d URUR-nSnSU[R "SUS[R S9RU[RS9U- -- nXe4$) aH Computes the inverse frequencies according to the original RoPE implementation Args: config ([`~transformers.PreTrainedConfig`]): The model configuration. device (`torch.device`): The device to use for initialization of the inverse frequencies. seq_len (`int`, *optional*): The current sequence length. Unused for this type of RoPE. Returns: Tuple of (`torch.Tensor`, `float`), containing the inverse frequencies for the RoPE embeddings and the post-processing scaling factor applied to the computed cos/sin (unused in this type of RoPE). rope_thetahead_dimNg?rrTrW)rxrW) rsgetattrr+num_attention_headsrMarangeint64rXre)r*rxr{basedimattention_factorrjs r6rt4ArceeRotaryEmbedding.compute_default_rope_parametershs&%%l3fj$/c63E3EIcIc3c U\\!S!5;;?BB&X]XcXcBdgjj k ))r8cLURSSS2S4R5RURSSS5R UR 5nUSS2SSS24R5n[ UR R[5(a0UR RS:waUR ROSn[USS9 UR5UR5-RSS5n[R"Xf4SS 9nUR5UR-nUR5UR-n SSS5 WR UR S 9W R UR S 94$!,(df  N@=f) NrrUr mpscpuF) device_typeenabledrTrr)rjreexpandrbrXrx isinstancetypestrr transposerMcatcosrusinrW) r4r; position_idsinv_freq_expandedposition_ids_expandedrfreqsembrrs r6r<ArceeRotaryEmbedding.forwardsN!MM$4-8>>@GG HZHZ[\H]_acdehhijiqiqr ,QaZ 8 > > @'1!((--'E'E!((--[`J`ahhmmfk  U C&,,.1F1L1L1NNYYZ[]^_E))UN3C'')d444C'')d444C D vvAGGv$cff177f&;;; D Cs BF F#)rur*rqrrrlr:NNN)r>r?r@rArMrf__annotations__r!r) staticmethodrintrarertno_gradrr<rBrCrDs@r6rhrhUsllV{VV %)+/"*d"*(*t* ~u$ % **: ]]_ <