+
    œ°³j=  ã                   óx   € R t ^ RIHt ^RIHt ^RIHt ^RIHt ]! RR7      ] ! R R]4      4       4       t	R.t
R	# )
zFalcon configuration)Ústrict)ÚPreTrainedConfig)ÚRopeParameters)Úauto_docstringztiiuae/falcon-7b)Ú
checkpointc                   óÌ   a a€ ] tR t^t oRtRtR.tRtRt^ t	^Gt
RtRtRtR	tR
tR
tRtRtRtR	tR	tRtRtRt^t^tRtRtRtR	tV 3R lt] R 4       t!] R 4       t"V3R lt#Rt$Vt%V ;t&# )ÚFalconConfigaœ  
num_ln_in_parallel_attn (`int`, *optional*):
    Set to 2 if separate layer norms are to be used for the MLP and the attention output when using parallel
    attention, otherwise, 1.
alibi (`bool`, *optional*, defaults to `False`):
    Whether to use ALiBi positional biases during self-attention.
new_decoder_architecture (`bool`, *optional*, defaults to `False`):
    Whether to use the new (Falcon-40B) decoder architecture. If `True`, the `multi_query` and `parallel_attn`
    arguments are ignored, as the new decoder always uses parallel attention.
multi_query (`bool`, *optional*, defaults to `True`):
    Whether to use multi-query attention in the decoder. Ignored when `new_decoder_architecture` is `True`.
parallel_attn (`bool`, *optional*, defaults to `True`):
    Whether to compute attention in parallel with the feedforward layer. If False, they are consecutive
    instead, as in the original Transformer architecture. Ignored when `new_decoder_architecture` is `True`.
bias (`bool`, *optional*, defaults to `False`):
    Whether to use bias on Linear layers.
ffn_hidden_size (`int`, *optional*):
    The hidden size of the feedforward layer in the Transformer decoder.
    defaults to 4x hidden dim
activation (`str`, *optional*, defaults to `"gelu"`):
    The activation function used in the feedforward layer.

Example:

```python
>>> from transformers import FalconModel, FalconConfig

>>> # Initializing a small (2-layer) Falcon configuration
>>> configuration = FalconConfig(num_hidden_layers=2)

>>> # Initializing a model from the small configuration
>>> model = FalconModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config
```ÚfalconÚpast_key_valuesi þ  iÀ  Ngñhãˆµøä>g{®Gáz”?Tg        Fi   Úgeluc                ó  <€ VP                  R R4      pVf   V P                  MTV n        V P                  f   V P                  MV P                  V n        V P                  f   V P                  ^,          V n        \
        SV `  ! R/ VB  R# )Ún_embedN© )ÚpopÚhidden_sizeÚnum_kv_headsÚnum_attention_headsÚffn_hidden_sizeÚsuperÚ__post_init__)ÚselfÚkwargsr   Ú	__class__s   &, €Úp/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/falcon/configuration_falcon.pyr   ÚFalconConfig.__post_init__[   sx   ø€ à—*‘*˜Y¨Ó-ˆØ/6ª˜4×+Ò+ÀGˆÔØ8<×8IÑ8IÒ8Q˜D×4Ò4ÐW[×WhÑWhˆÔØ×ÑÒ'Ø#'×#3Ñ#3°aÕ#7ˆDÔ ä‰ÒÑ' Ô'ó    c                ó<   € V P                   V P                  ,          # ©N)r   r   ©r   s   &r   Úhead_dimÚFalconConfig.head_dime   s   € à×Ñ 4×#;Ñ#;Õ;Ð;r   c                ó$   € V P                   '       * # r   )Úalibir   s   &r   ÚrotaryÚFalconConfig.rotaryi   s   € à—:‘:Œ~Ðr   c                ój  <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[ R,          ;R&   S[R,          ;R&   S[;R&   S[;R	&   S[S[ ,          R,          ;R
&   S[S[ ,          R,          ;R&   S[ R,          ;R&   S[R,          ;R&   S[R,          ;R&   S[R,          ;R&   S[R,          ;R&   S[R,          ;R&   S[ ;R&   S[S[,          R,          ;R&   S[ R,          ;R&   S[ S[S[ ,          ,          R,          ;R&   S[ R,          ;R&   S[ R,          ;R&   S[R,          ;R&   S[;R&   # )é   Ú
vocab_sizer   Únum_hidden_layersr   NÚnum_ln_in_parallel_attnÚlayer_norm_epsilonÚinitializer_rangeÚ	use_cacheÚhidden_dropoutÚattention_dropoutr   r"   Únew_decoder_architectureÚmulti_queryÚparallel_attnÚbiasÚmax_position_embeddingsÚrope_parametersÚbos_token_idÚeos_token_idÚpad_token_idr   Ú
activationÚtie_word_embeddings)ÚintÚfloatÚboolr   ÚdictÚlistÚstr)ÚformatÚ__classdict__s   "€r   Ú__annotate__ÚFalconConfig.__annotate__   sÆ  ø‡ ‚ ñV ÑñW ñX ÑñY ñZ Ññ[ ñ\ Ñ!ñ] ñ^ ! 4�ZÑ.ñ_ ñ`  �Ñ+ña ñb Ñ#ñc ñd Ññe ñf ™C•K $Õ&Ñ,ñg ñh ™s•{ TÕ)Ñ/ñi ñj ˜•*Ñ#ñk ñl �$�;Ññm ñn # T�kÑ1ño ñp ˜•Ñ#ñq ñr ˜$•;Ñ%ñs ñt ��+Ññu ñv !Ñ'ñw ñx $¡dÕ*¨TÕ1Ñ8ñy ñz ˜•*Ñ!ñ{ ñ| ™™S�	•/ DÕ(Ñ-ñ} ñ~ ˜•*Ñ#ñ ñ@ ˜4•ZÑ&ñA ñB �d•
Ñ#ñC ñD Ñ$òE r   )r   r   r   )'Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__Ú__doc__Ú
model_typeÚkeys_to_ignore_at_inferencer'   r   r(   r   r)   r*   r+   r,   r-   r.   r   r"   r/   r0   r1   r2   r3   r4   r5   r6   r7   r   r8   r9   r   Úpropertyr   r#   Ú__annotate_func__Ú__static_attributes__Ú__classdictcell__Ú__classcell__)r   rA   s   @@r   r   r      sÑ   ù‡ € ñ#ðJ €JØ#4Ð"5Ðà€JØ€KØÐØ!ÐØ*.ÐØ'+ÐØ#ÐØ€IØ),€NØ,/ÐØ#€LØ€EØ,1ÐØ#€KØ!%€MØ€DØ#'ÐØ48€OØ!€LØ+-€LØ#€LØ"&€OØ#€JØ $Ðõ(ð ñ<ó ð<ð ñó ð÷g … r   r   N)rH   Úhuggingface_hub.dataclassesr   Úconfiguration_utilsr   Úmodeling_rope_utilsr   Úutilsr   r   Ú__all__r   r   r   Ú<module>rU      sN   ðñ å .å 3Ý 1Ý #ñ Ð-Ô.ØôRÐ#ó Ró ó /ðRðj Ð
‚r   