ZjL_SSKJr SSKJr SSKrSSKJr SSKJr SSK J r SSK J r Jr SSKJr SS KJrJr SS KJrJr SS KJr SS KJrJrJr SS KJrJr SSK J!r!J"r" SSK#J$r$J%r% SSK&J'r' SSK(J)r)J*r*J+r+ SSK,J-r-J.r. SSK/J0r0 SSK1J2r2 "SS\Rf5r4"SS\Rf5r5"SS\Rf5r6Sr7\"S5S=Sj5r8S\RrS \:S!\Rr4S"jr;S>S#\RfS$\RrS%\RrS&\RrS'\RrS-S(\<\:-S)\\"\85"S,S-\Rf55r?"S.S/\5r@"S0S1\R5rB\*"S2S3\%55rC\*"S4S5\C55rD\*"S6S7\C\55rE"S8S9\\C5rF"S:S;\\C5rG/S[TU]5 X l[R"[ R "U55UlgN)super__init__r&nn Parametertorchzerosweight)selfr%r& __class__s {/root/GenerationalWealth/GenerationalWealth/venv/lib/python3.13/site-packages/transformers/models/gemma2/modeling_gemma2.pyr*Gemma2RMSNorm.__init__2s, ll5;;s#34 cU[R"URS5RSSS9UR-5-$)NT)keepdim)r-rsqrtpowmeanr&)r0xs r2_normGemma2RMSNorm._norm7s45;;quuQx}}R}>IJJJr4cURUR55nUSURR5--nURU5$)N?)r=floatr/type_as)r0r<outputs r2forwardGemma2RMSNorm.forward:sCAGGI&3!2!2!445~~a  r4c^[URR5SUR3$)Nz, eps=)tupler/shaper&)r0s r2 extra_reprGemma2RMSNorm.extra_reprAs' ))*+6$((<) __name__ __module__ __qualname____firstlineno__intrAr*r=rDrI__static_attributes__ __classcell__r1s@r2r#r#1s05C5e55 K!==r4r#c.^\rSrSrU4SjrSrSrU=r$) Gemma2MLPEc>[TU]5 XlURUlURUl[ R "URURSS9Ul[ R "URURSS9Ul[ R "URURSS9Ul [URUl gNFbias) r)r*config hidden_sizeintermediate_sizer+Linear gate_projup_proj down_projrhidden_activationact_fnr0rZr1s r2r*Gemma2MLP.__init__Fs  !--!'!9!94#3#3T5K5KRWXyy!1!143I3IPUV 4#9#94;K;KRWXV556 r4cURURURU55URU5-5nU$r()r`rbr^r_)r0r<r`s r2rDGemma2MLP.forwardPs6NN4;;t~~a/@#ADLLQRO#ST r4)rbrZr`r^r[r\r_)rKrLrMrNr*rDrPrQrRs@r2rTrTEs7r4rTc ^\rSrSr%\R \S'SS\4U4Sjjjr\ SS\S-S\ SS\ S-S \ S \ 44S jj5r\R"5\S 55rS rU=r$)Gemma2RotaryEmbeddingUinv_freqNrZc>[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_lenrZrope_parametersrlcompute_default_rope_parametersrattention_scalingregister_bufferclone)r0rZdevice rope_init_fnrjr1s r2r*Gemma2RotaryEmbedding.__init__Xs "("@"@$*$B$B! 44[A!%!E!E >>Y &.t~~>L+7 V+L(( ZeD 0(..2BuUr4ryz torch.deviceseq_lenreturnz torch.Tensorc URSn[USS5=(d URUR-nSnSU[R "SUS[R S9RU[RS9U- -- nXe4$) 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_dimNr@rr6dtype)ryr) rtgetattrr[num_attention_headsr-arangeint64torA)rZryr|baser%attention_factorrjs r2ru5Gemma2RotaryEmbedding.compute_default_rope_parametershs&%%l3fj$/c63E3EIcIc3c U\\!S!5;;?BB&X]XcXcBdgjj k ))r4cLURSSS2S4R5RURSSS5R UR 5nUSS2SSS24R5n[ UR R[5(a0UR RS:waUR ROSn[USS9 UR5UR5-RSS5n[R"Xf4SS 9nUR5UR-nUR5UR-n SSS5 WR UR S 9W R UR S 94$!,(df  N@=f) Nrr7r mpscpuF) device_typeenabledr6r%r)rjrAexpandrHrry isinstancetypestrr transposer-catcosrvsinr) r0r< position_idsinv_freq_expandedposition_ids_expandedrfreqsembrrs r2rDGemma2RotaryEmbedding.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#)rvrZrrrsrlr(NNN)rKrLrMrNr-Tensor__annotations__r!r* staticmethodrrOrGrAruno_gradrrDrPrQrRs@r2rhrhUsllV|VV &*+/"*t#*(*t* ~u$ % **: ]]_ <