[bugfix] Initialize attention bias on the same device as Query/Key/Value
#13 opened 2 months ago
by
kenneth-doh
Problems encountered when loading the model for offline operation using SentenceTransformer
#12 opened 2 months ago
by
HT-NEKO

Finetuning model performance decreases when using memory_efficient_attenttion
#11 opened 6 months ago
by
hrushikesh1
Any Plan to release tensorflow based model?
1
#10 opened 9 months ago
by
undefined-x
adding to transformers officially?
👍
2
#9 opened 9 months ago
by
pszemraj

Is flash-attention-2 suppported
1
#8 opened 11 months ago
by
Jack7777777
Xformer support for Qwen1.5B
3
#6 opened about 1 year ago
by
le723z
backbone模型不开源吗?
10
#4 opened about 1 year ago
by
JaheimLee
Disable trust_remote_code
👍
🔥
4
17
#2 opened over 1 year ago
by
veeravignesh