Resources

Encoding for long contexts

#28 opened 5 days ago by

yoavkt

Fix AutoModel not loading model correctly due to config_class inconsistency

#26 opened 3 months ago by

liamclarkza

Update bert_layers.py

#25 opened 3 months ago by

ruffy369

A modified DNABERT2 that returns the attention too

#24 opened 3 months ago by

jaandoui

Triton version

#23 opened 5 months ago by

JiayiJennie

how do I output the attention scores from the last layer of the encoder?

#22 opened 5 months ago by

jkb0722

Tokenization of more than 2 sequences

#21 opened 5 months ago by

jaandoui

Impact of Padding on DNABERT Model Performance

#20 opened 5 months ago by

poilkjhytg

Setting DNABERT-2 revision

#19 opened 6 months ago by

SeanDoyle

TypeError: forward() got an unexpected keyword argument 'attention_mask'

#18 opened 6 months ago by

jkb0722

Adding `safetensors` variant of this model

#17 opened 6 months ago by

SFconvertbot

Assertion Error / Implementation Error

#16 opened 6 months ago by

8497prashant

Extract attention from model

#15 opened 6 months ago by

kaustabanv

error on inference using DNABERT2 - can you please share the environment you used for running it?

#12 opened 9 months ago by

NettaB

Adding `safetensors` variant of this model

#11 opened 9 months ago by

SFconvertbot

Is the datasets for foundational model pre-training publicly accessible?

#10 opened 10 months ago by

JayceCeleste

Expecting bi-modal disstribution of probabilities

#9 opened 10 months ago by

christianclough

Inference fails with output_all_encoded_layers=True.

#8 opened 11 months ago by

pg20sanger

Adding `safetensors` variant of this model

#7 opened 12 months ago by

SFconvertbot

The model class you are passing has a `config_class` attribute that is not consistent with the config class you passed

#6 opened 12 months ago by

saikiran7

About the output of tokenizer and the model

#4 opened about 1 year ago by

RandyWang504

Despite multiple trials and examining the model configuration, it seems that the model hosted on Hugging Face (`huggingface.co`) cannot handle sequences that exceed a length of 512 tokens.

#3 opened about 1 year ago by

hengchuangyin

Issue in Code, at 114:24: def _fwd_kernel(..)

#1 opened about 1 year ago by

Hosna