+ jRt^RIHt^RIHt^RIHt^RIHt]]!RRR7!RR ]444t ]]!R RR7!R R ]444t ]!RR 7]!RR]444t R.t R#)zDBRX model configuration)strict)PreTrainedConfig)RopeParameters)auto_docstringz4This config is used to instantiate attention layers.z$transformers-community/dbrx-instruct) custom_intro checkpointc@a]tRt^toRtRtRtRt^tV3Rlt Rt Vt R#)DbrxAttentionConfiga^ attn_pdrop (`float`, *optional*, defaults to 0.0): The dropout probability for the attention layers. clip_qkv (`float`, *optional*): If set, clip the queries, keys, and values in the attention layer to this value. kv_n_heads (`int`, *optional*, defaults to 1): For grouped_query_attention only, allow user to specify number of kv heads. attn_configNcl<V^8dQh/S[S[,;R&S[S[,R,;R&S[;R&#) attn_pdropNclip_qkv kv_n_heads)floatint)format __classdict__s"l/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/dbrx/configuration_dbrx.py __annotate__ DbrxAttentionConfig.__annotate__s9" !#$EkD '%&') __name__ __module__ __qualname____firstlineno____doc__base_config_keyrrr__annotate_func____static_attributes____classdictcell__)rs@rr r s' $O!J#'HJ'rr z6This config is used to instantiate feedforward layers.cfaa]tRt^-toRtRtRtRtRt^t ^t Rt Rt Rt V3RltV3R ltR tVtV;t#) DbrxFFNConfiga ffn_act_fn (`dict`, *optional*, defaults to `None`): A dict specifying activation function for the FFN. The dict should have a key 'name' with the value being the name of the activation function along with any additional keyword arguments. If `None`, then set to `{"name": "silu"}`. ffn_hidden_size (`int`, *optional*, defaults to 3584): The hidden size of the feedforward network. moe_num_experts (`int`, *optional*, defaults to 4): The number of experts in the mixture of experts layer. moe_top_k (`int`, *optional*, defaults to 1): The number of experts to use in the mixture of experts layer. moe_jitter_eps (`float`, *optional*, defaults to `None`): If not `None`, the jitter epsilon for the mixture of experts layer. moe_loss_weight (`float`, *optional*, defaults to 0.01): The loss weight for the mixture of experts layer. moe_normalize_expert_weights (`float`, *optional*, defaults to 1.0): The normalization factor for the expert weights. ffn_configiNig{Gz?g?c <VPf RR/VnRFpW!9gK VPV4K \V4^8wd\RV: 24h\SV`!R/VBR#)NnamesiluzFound unknown kwargs=) model_typeattn_implementationexperts_implementationtransformers_version _commit_hash torch_dtypedtyper) ffn_act_fnpoplen ValueErrorsuper __post_init__)selfkwargsk __class__s&, rr5DbrxFFNConfig.__post_init__Qsh ?? "%v.DO A{ 1   v;! 5fY78 8 ''rc<V^8dQh/S[;R&S[R,;R&S[;R&S[;R&S[;R&S[S[,R,;R&S[;R&S[R,;R &#) r hidden_sizeNr0ffn_hidden_sizemoe_num_experts moe_top_kmoe_jitter_epsmoe_loss_weightmoe_normalize_expert_weights)rdictr)rrs"rrDbrxFFNConfig.__annotate__-s678t "9:;<=>?@CK$&-AB!CD#($,4Er)r0)rrrrrrr<r0r=r>r?r@rArBr5r r!r" __classcell__r9rs@@rr$r$-sG &#OK"JOOI)-N!O14 (Irr$)rcaa]tRt^ftoRtRtR]R]/tRRRRR R R R /t R t ^t ^t R t RtRtRtRtRtRtRtRtRtRtRtRtRtV3RltRtV3RltRtVtV;t #) DbrxConfiga max_seq_len (`int`, *optional*, defaults to 2048): The maximum sequence length of the model. attn_config (`dict`, *optional*): A dictionary used to configure the model's attention module. ffn_config (`dict`, *optional*): A dictionary used to configure the model's FFN module. Example: ```python >>> from transformers import DbrxConfig, DbrxModel >>> # Initializing a Dbrx configuration >>> configuration = DbrxConfig(n_layers=2, d_model=256, n_heads=8, vocab_size=128) >>> # Initializing a model (with random weights) from the configuration >>> model = DbrxModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ``` dbrxr r%num_attention_headsn_headsr<d_modelnum_hidden_layersn_layersmax_position_embeddings max_seq_lenii}r NTg{Gz?Fc <VPf\4VnM:\VP\4'd\R/VPBVnVPf\ 4VnM:\VP\4'd\ R/VPBVnVP VPnVPPVn \SV`,!R/VBR#)Nr) r r isinstancerCr%r$rLr<rnum_key_value_headsr4r5)r6r7r9s&,rr5DbrxConfig.__post_init__s    #24D  (($ / /2FT5E5EFD  ?? "+oDO  . .+>doo>DO'+ll##'#3#3#>#>  ''rc @VP'd \R4hR#)zOPart of `@strict`-powered validation. Validates the architecture of the config.z5tie_word_embeddings is not supported for DBRX models.N)tie_word_embeddingsr3)r6s&rvalidate_architecture DbrxConfig.validate_architectures  # # #TU U $rc<V^8dQh/S[R,;R&S[R,;R&S[R,;R&S[R,;R&S[;R&S[R,;R&S[R,;R&S[S[,R,;R &S[S[,R,;R &S[;R &S[;R &S[R,;R &S[S[,R,;R&S[R,;R&S[R,;R&S[S[S[,,R,;R&S[;R&#)r NrLrKrNrP vocab_size resid_pdrop emb_pdropr r% use_cacheinitializer_rangeoutput_router_logitsrope_parameters pad_token_id bos_token_id eos_token_idrV)rrr rCr$boolrlist)rrs"rrDbrxConfig.__annotate__fsSF4ZGH4ZIJDjKLt"MNOP#QRt|!ST%t+d29UV$t+2WXYZ#[\+-]^$d*T18_`*#ab*#cdS /D(/ef%gr)r r%rS)!rrrrrr)r r$ sub_configs attribute_maprLrKrNrPrZr[r\r r%r]r^r_r`rarbrcrVr5rWr r!r"rErFs@@rrHrHfs.J "5|]SKyyZ!= MGGH"KJ #K!I59K.2JI#(-48O#L#L+/L %("VMrrHN) rhuggingface_hub.dataclassesrconfiguration_utilsrmodeling_rope_utilsrutilsrr r$rH__all__rrrrns.31#G5*  "I51($1(  1(hABGV!GVCGVT .r