bCimPSSKJrJr SSKrSSKJr SSKJrJr SSK J r J r SSK J r Jr SSKJr SSKJr SS KJr SS KJr SS KJr SS KJrJr SS KJr SSKJrJ r J!r!J"r"J#r#J$r$J%r%J&r& SSK'J(r( \RR"\*5r+"SS\ 5r,"SS\$5r-"SS\"5r."SS\5r/"SS\ 5r0"SS\#5r1"SS\(5r2"SS \!5r3/S!Qr4g)")CallableOptionalN)Cache DynamicCache)PretrainedConfiglayer_type_validation)create_causal_mask!create_sliding_window_causal_mask)FlashAttentionKwargs)BaseModelOutputWithPast)rope_config_validation)ALL_ATTENTION_FUNCTIONS)Unpack)TransformersKwargslogging)deprecate_kwarg)CohereAttentionCohereDecoderLayerCohereForCausalLMCohereLayerNormCoherePreTrainedModelCohereRotaryEmbeddingapply_rotary_pos_embeager_attention_forward) Gemma2Modelc^\rSrSrSrSrS/rSSSSSSSS.rS/S /4S S /S /4S /S /4S .rSU4S jjr Sr U=r $) Cohere2Config/a This is the configuration class to store the configuration of a [`CohereModel`]. It is used to instantiate an Cohere model according to the specified arguments, defining the model architecture. Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Instantiating a configuration with the defaults will yield a similar configuration to that of the [CohereForAI/c4ai-command-r-v01](https://huggingface.co/CohereForAI/c4ai-command-r-v01) model. Args: vocab_size (`int`, *optional*, defaults to 256000): Vocabulary size of the Cohere model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`CohereModel`] hidden_size (`int`, *optional*, defaults to 8192): Dimension of the hidden representations. intermediate_size (`int`, *optional*, defaults to 22528): Dimension of the MLP representations. logit_scale (`float`, *optional*, defaults to 0.0625): The scaling factor for the output logits. num_hidden_layers (`int`, *optional*, defaults to 40): Number of hidden layers in the Transformer decoder. num_attention_heads (`int`, *optional*, defaults to 64): Number of attention heads for each attention layer in the Transformer decoder. num_key_value_heads (`int`, *optional*): This is the number of key_value heads that should be used to implement Grouped Query Attention. If `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed by meanpooling all the original heads within that group. For more details, check out [this paper](https://huggingface.co/papers/2305.13245). If it is not specified, will default to `num_attention_heads`. hidden_act (`str` or `function`, *optional*, defaults to `"silu"`): The non-linear activation function (function or string) in the decoder. max_position_embeddings (`int`, *optional*, defaults to 8192): The maximum sequence length that this model might ever be used with. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. layer_norm_eps (`float`, *optional*, defaults to 1e-05): The epsilon used by the layer normalization. use_cache (`bool`, *optional*, defaults to `True`): Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if `config.is_decoder=True`. pad_token_id (`int`, *optional*, defaults to 0): Padding token id. bos_token_id (`int`, *optional*, defaults to 5): Beginning of stream token id. eos_token_id (`int`, *optional*, defaults to 255001): End of stream token id. tie_word_embeddings (`bool`, *optional*, defaults to `True`): Whether to tie weight embeddings rope_theta (`float`, *optional*, defaults to 10000.0): The base period of the RoPE embeddings. rope_scaling (`Dict`, *optional*): Dictionary containing the scaling configuration for the RoPE embeddings. NOTE: if you apply new rope type and you expect the model to work on longer `max_position_embeddings`, we recommend you to update this value accordingly. Expected contents: `rope_type` (`str`): The sub-variant of RoPE to use. Can be one of ['default', 'linear', 'dynamic', 'yarn', 'longrope', 'llama3'], with 'default' being the original RoPE implementation. `factor` (`float`, *optional*): Used with all rope types except 'default'. The scaling factor to apply to the RoPE embeddings. In most scaling types, a `factor` of x will enable the model to handle sequences of length x * original maximum pre-trained length. `original_max_position_embeddings` (`int`, *optional*): Used with 'dynamic', 'longrope' and 'llama3'. The original max position embeddings used during pretraining. `attention_factor` (`float`, *optional*): Used with 'yarn' and 'longrope'. The scaling factor to be applied on the attention computation. If unspecified, it defaults to value recommended by the implementation, using the `factor` field to infer the suggested value. `beta_fast` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for extrapolation (only) in the linear ramp function. If unspecified, it defaults to 32. `beta_slow` (`float`, *optional*): Only used with 'yarn'. Parameter to set the boundary for interpolation (only) in the linear ramp function. If unspecified, it defaults to 1. `short_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to short contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `long_factor` (`list[float]`, *optional*): Only used with 'longrope'. The scaling factor to be applied to long contexts (< `original_max_position_embeddings`). Must be a list of numbers with the same length as the hidden size divided by the number of attention heads divided by 2 `low_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to low frequency components of the RoPE `high_freq_factor` (`float`, *optional*): Only used with 'llama3'. Scaling factor applied to high frequency components of the RoPE attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`): Whether to use a bias in the query, key, value and output projection layers during self-attention. attention_dropout (`float`, *optional*, defaults to 0.0): The dropout ratio for the attention probabilities. sliding_window (`int`, *optional*, defaults to 4096): Size of the sliding window attention context. layer_types (`list`, *optional*): Attention pattern for each layer. ```python >>> from transformers import Cohere2Model, Cohere2Config >>> # Initializing a Cohere Nextmodel configuration >>> configuration = Cohere2Config() >>> # Initializing a model from the Cohere2 configuration >>> model = Cohere2Model(configuration) # doctest: +SKIP >>> # Accessing the model configuration >>> configuration = model.config # doctest: +SKIP ``` cohere2past_key_valuescolwiserowwise)zlayers.*.self_attn.q_projzlayers.*.self_attn.k_projzlayers.*.self_attn.v_projzlayers.*.self_attn.o_projzlayers.*.mlp.gate_projzlayers.*.mlp.up_projzlayers.*.mlp.down_proj input_ids inputs_embeds hidden_statesattention_mask) embed_tokenslayersnormc >XlXlX lX@lX0lXPlX`lUcUnXplXlXl Xl Xl UUl UUl UUlUUlUUlUUlX&-Ul['U5 [(TU]T"SU UUUS.UD6 UR-SS5UlUR"c_[1USS5Ul[3UR 5Vs/sH'n[5US-UR.-5(aSOSPM) snUl[7UR"UR 5 gs snf)N) pad_token_id bos_token_id eos_token_idtie_word_embeddingssliding_window_patternsliding_attentionfull_attention) vocab_sizemax_position_embeddings hidden_size logit_scaleintermediate_sizenum_hidden_layersnum_attention_headsnum_key_value_heads hidden_actinitializer_rangelayer_norm_eps use_cache rope_theta rope_scalingattention_biasattention_dropoutsliding_window layer_typeshead_dimrsuper__init__get_sliding_window_patterngetattrrangeboolr )selfr7r9r;r:r<r=r>r?r8r@rArBr-r.r/r0rCrDrErFrGrHkwargsi __class__s e/home/james-whalen/.local/lib/python3.13/site-packages/transformers/models/cohere2/modular_cohere2.pyrKCohere2Config.__init__sb4%'>$&&!2!2#6   &"5 #6 $!2,"$(,!2,&#:  t$  %%% 3    (.zz2JA'N$    #+249QST+UD (t556 6A(,QUd6R6R,R'S'S#Yii6 D  d..0F0FG  s/.E)rMrErFrIr?r9r@r;rArHr:r8r=r<r>rDrCrGrBr7)i iXg?(@NsilurWg{Gz?gh㈵>TriTg@NFiN) __name__ __module__ __qualname____firstlineno____doc__ model_typekeys_to_ignore_at_inferencebase_model_tp_planbase_model_pp_planrK__static_attributes__ __classcell__rTs@rUrr/sn`J#4"5%.%.%.%."+ )"+&(9:#%568IJ!"_$56  $ /IHIHrc\rSrSrSrg)Cohere2RotaryEmbeddingr6Nr]r^r_r`rfr6rirUrkrkrirkc\rSrSrSrg)Cohere2LayerNormir6Nrmr6rirUrprprnrirpcX\rSrSrSrSS\S\\4Sjjr\ "SSS S 9SS \ RS \ \ R\ R4S \\ RS\\ S\\ RS\\S\ \ R\\ R\\ \ R44Sjj5rSrg)Cohere2Attentioniz=Multi-headed attention from 'Attention Is All You Need' paperNconfig layer_idxc[RRU5 XlX l[ USUR UR-5UlURUR-Ul URS-Ul URUl SUl URUS:Xa UROSUl[R "UR URUR-UR"S9Ul[R "UR URUR-UR"S9Ul[R "UR URUR-UR"S9Ul[R "URUR-UR UR"S9Ulg)NrIgTr4)bias)nnModulerKrsrtrNr9r=rIr>num_key_value_groupsscalingrF is_causalrHrGLinearrEq_projk_projv_projo_proj)rQrsrts rUrKCohere2Attention.__init__ss 4  " F4F4F&JdJd4de $*$>$>&B\B\$\!}}d* !'!9!97=7I7I)7TXk7kf33quii    : :T]] JQWQfQf  ii    : :T]] JQWQfQf  ii    : :T]] JQWQfQf  ii  & & 68J8JQWQfQf  ripast_key_valuer"4.58new_nameversionr'position_embeddingsr(cache_positionrRreturnc dURSSn/UQSPURP7nURU5RU5R SS5n UR U5RU5R SS5n UR U5RU5R SS5n UupURb[XX5upUb$XUS.nURXURU5up[nURRS:wa[URRnU"UU U U U4UR(dSO UR UR"URS.UD6unnUR$"/UQSP76R'5nUR)U5nUU4$)Nr3r)sincosreagerr\)dropoutrzrG)shaperIr}view transposer~rrGrupdatertrrs_attn_implementationrtrainingrFrzreshape contiguousr)rQr'rr(r"rrR input_shape hidden_shape query_states key_states value_statesrr cache_kwargsattention_interface attn_output attn_weightss rUforwardCohere2Attention.forward s$))#2. 88b8$--8 {{=166|DNNqRST [[/44\BLLQPQR {{=166|DNNqRST &    *';LVY'_ $L  &#&nUL'6'='=jX\XfXfht'u $J(? ;; + +w 6"9$++:Z:Z"[ $7      % $}}C$2H2HLL.. %  % ! \"));;;;FFH kk+. L((ri) rFrsrIr{r~rtryrr}rzrGrN)NN)r]r^r_r`rarrintrKrtorchTensortupler LongTensorrr rrfr6rirUrrrrsG } # 0%0A6R ,059 *)||*)#5<<#=>*)!. *) "% *) !!1!12 *)-.*) u||Xell3XeELL>Q5RR S*)S*)rirrcb^\rSrSrS\S\4U4Sjjr\"SSSS9SS \RS \ \R\R4S \ \RS\ \ S \ \ S \ \RS\\S\ \R"\ \ \R"\R"444Sjj5rSrU=r$)Cohere2DecoderLayeriNrsrtcL>[TU]X5 URUUlgr)rJrKrHattention_type)rQrsrtrTs rUrKCohere2DecoderLayer.__init__Os# +$00;rirr"rrr'rr(rBrrRrc UnURU5nUR"SUUUUUUS.UD6upURU5n X-U -nU$)N)r'rr(r"rBrr6)input_layernorm self_attnmlp) rQr'rr(r"rBrrRresidualhidden_states_attention_hidden_states_mlps rUrCohere2DecoderLayer.forwardSsq!,,]; %)^^& ' 3)+) & & "!HH]3 :=NN ri)r)NNFN)r]r^r_r`rrrKrrrrrrrPrrr FloatTensorrrfrgrhs@rUrrNs<}<<%0A6R 26+/$)59||#5<<#=>!.  "%  D> !!1!12-. u  (51B1BEDUDU1U+V"WW XSrirc \rSrSr%\\S'Srg)Cohere2PreTrainedModeliorsr6N)r]r^r_r`r__annotations__rfr6rirUrros rirc^\rSrSrS\4U4SjjrSS\\RS\\RS\\RS\\ S\\RS \\ S \\RS \ \S \4S jjrSrU=r$) Cohere2Modelisrsc>[TU]U5 [URURS9Ul[ US9Ulg)N)r9epsrs)rJrKrpr9rAr+rk rotary_emb)rQrsrTs rUrKCohere2Model.__init__ts6  $&2D2D6K`K`a 0?rir%r( position_idsr"r&rBrrRrc USLUSL- (a [S5eUcURU5nU(a'Uc$UR(d[URS9nUcDUbUR 5OSn [ R"XURS-URS9nUcURS5n[U=n [5(d*URUUUUUS.n [S 0U D6[S 0U D6S.n Un URX5n UR HnU"U 4U XR"UUUS.UD6n M UR%U 5n ['U US 9$) Nz:You must specify exactly one of input_ids or inputs_embedsrrr3)device)rs input_embedsr(rr"r)r5r4)rr(r"rBr)last_hidden_stater"r6) ValueErrorr)rrrsget_seq_lengthrarangerr unsqueeze isinstancedictr r rr*rr+r )rQr%r(rr"r&rBrrRpast_seen_tokenscausal_mask_mapping mask_kwargsr'r decoder_layers rUrCohere2Model.forwardys~  -t"; <YZ Z   --i8M 0*$++>O  !CRC^==?de "\\ ]5H5H5K"KTaThThN  )33A6L?-FF++ -"0"0#2 , K#5"C{"C%F%U%U#  & "oomJ![[M)$723O3OP /#- M) -0 &++  ri)r+r)NNNNNNN)r]r^r_r`rrKrrrrrrrPrrr rrfrgrhs@rUrrss@}@151537+/59$(59< E,,-< !.< u//0 < "% <   1 12 < D>< !!1!12< +,<  !< < rirc\rSrSrSrg)Cohere2ForCausalLMir6Nrmr6rirUrrrnrir)rrrr)5typingrrrtorch.nnrw cache_utilsrrconfiguration_utilsrr masking_utilsr r modeling_flash_attention_utilsr modeling_outputsr modeling_rope_utilsrmodeling_utilsrprocessing_utilsrutilsrrutils.deprecationrcohere.modeling_cohererrrrrrrrgemma2.modeling_gemma2r get_loggerr]loggerrrkrprrrrrr__all__r6rirUrs & .JRB795&00   1   H %KH$KH\ 2   F)F)R,B2B ;B J *  \ri