Zj|xSSKrSSKJr SSKJr SSKrSSKJr SSKJr SSK J r SSK J r JrJr SS KJr SS KJrJr SS KJr SS KJrJr SS KJrJrJrJr SSKJ r J!r! SSK"J#r#J$r$ SSK%J&r& SSK'J(r( SSK)J*r*J+r+J,r,J-r- SSK.J/r/J0r0 SSK1J2r2 SSK3J4r4 \-Rj"\65r7"SS\Rp5r9"SS\Rp5r:Sr;\"S5SS \Rz4S!jr?S"r@"S#S$\Rp5rA"S%S&\A5rB"S'S(\A5rC\"S)5"S*S+\Rp55rD\A\B\CS,.rE"S-S.\5rF\+"S/S0\&55rG\+"S1S2\G55rH\+"S3S4\G\55rI"S5S6\\G5rJ"S7S8\\G5rK"S9S:\\G5rL/S;QrMg)=N)Callable)Optional)nn)initialization)ACT2FN)Cache DynamicCache StaticCache)GenerationMixin)use_kernel_forward_from_hubuse_kernel_func_from_hub)create_causal_mask)_flash_attention_forward!flash_attn_supports_top_left_mask)GenericForQuestionAnswering GenericForSequenceClassificationGenericForTokenClassificationGradientCheckpointingLayer)BaseModelOutputWithPastCausalLMOutputWithPast)ROPE_INIT_FUNCTIONSdynamic_rope_update)PreTrainedModel)Unpack)TransformersKwargsauto_docstringcan_return_tuplelogging)maybe_autocastmerge_with_config_defaults)capture_outputs)DiffLlamaConfigc.^\rSrSrU4SjrSrSrU=r$) DiffLlamaMLP8c>[TU]5 XlURUlURUl[ R "URURSS9Ul[ R "URURSS9Ul[ R "URURSS9Ul [URUl gNFbias) super__init__config hidden_sizeintermediate_sizerLinear gate_projup_proj down_projr hidden_actact_fnselfr. __class__s ځ/root/GenerationalWealth/GenerationalWealth/venv/lib/python3.13/site-packages/transformers/models/diffllama/modeling_diffllama.pyr-DiffLlamaMLP.__init__9s  !--!'!9!94#3#3T5K5KRWXyy!1!143I3IPUV 4#9#94;K;KRWXV../ cURURURU55URU5-5nU$N)r4r6r2r3)r8xr4s r:forwardDiffLlamaMLP.forwardCs6NN4;;t~~a/@#ADLLQRO#ST r<)r6r.r4r2r/r0r3)__name__ __module__ __qualname____firstlineno__r-r@__static_attributes__ __classcell__r9s@r:r&r&8s0r<r&c ^\rSrSr%\R \S'SS\4U4Sjjjr\ SS\S-S\ SS\ S-S \ S \ 44S jj5r\R"5\S 55rS rU=r$)DiffLlamaRotaryEmbeddingHinv_freqNr.c>[TU]5 URUlURUlXlUR R SUlURnURS:wa[URnU"UR U5uo@l URSUSS9 URSUR5SS9 g)N rope_typedefaultrLF) persistentoriginal_inv_freq) r,r-max_position_embeddingsmax_seq_len_cachedoriginal_max_seq_lenr.rope_parametersrNcompute_default_rope_parametersrattention_scalingregister_bufferclone)r8r.device rope_init_fnrLr9s r:r-!DiffLlamaRotaryEmbedding.__init__Ks "("@"@$*$B$B! 44[A!%!E!E >>Y &.t~~>L+7 V+L(( ZeD 0(..2BuUr<rZz torch.deviceseq_lenreturnz torch.Tensorc URSn[USS5=(d URUR-nSnSU[R "SUS[R S9RU[RS9U- -- nXe4$) 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_dimNg?rdtype)rZrd) rUgetattrr/num_attention_headstorcharangeint64tofloat)r.rZr]basedimattention_factorrLs r:rV8DiffLlamaRotaryEmbedding.compute_default_rope_parameters[s&%%l3fj$/c63E3EIcIc3c U\\!S!5;;?BB&X]XcXcBdgjj k ))r<cLURSSS2S4R5RURSSS5R UR 5nUSS2SSS24R5n[ UR R[5(a0UR RS:waUR ROSn[USS9 UR5UR5-RSS5n[R"Xf4SS 9nUR5UR-nUR5UR-n SSS5 WR UR S 9W R UR S 94$!,(df  N@=f) Nrr#mpscpuF) device_typeenabledrbrmrc)rLrkexpandshaperjrZ isinstancetypestrr transposergcatcosrWsinrd) r8r? position_idsinv_freq_expandedposition_ids_expandedrtfreqsembr~rs r:r@ DiffLlamaRotaryEmbedding.forwardysN!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#)rWr.rSrTrNr>)NNN)rBrCrDrErgTensor__annotations__r$r- staticmethodrinttuplerkrVno_gradrr@rFrGrHs@r:rJrJHsllVVV )-+/"*$&*(*t* ~u$ % **: ]]_ <