NaN outputs when running in torch.float16 with text containing specific tokens (e.g., "import")
#27 opened 6 months ago
by 
Hank-Pan
how to use multi-gpu for inference?
#26 opened 8 months ago
by 
hh1212
如果想给embedding模型添加lm_heads该怎么做？
#25 opened 9 months ago
by 
linln1
Update README.md
#24 opened 9 months ago
by 
RealWorga
Update README.md
#23 opened 12 months ago
by 
aari1995
How to make the model generate a 32-dimensional vector
#22 opened 12 months ago
by 
talenm01
Issue with Token 474 Causing NaN Embeddings in 8B Model
➕ 2
8
#21 opened 12 months ago
by 
Sernaryon
dimensions not working it always generates 4096
#20 opened about 1 year ago
by 
khanakia
Training for classification task
1
#19 opened about 1 year ago
by 
nastyafairypro
Update `README.md` to use TEI v1.7 instead
1
#18 opened about 1 year ago
by 
alvarobartt
Is `tie_word_embeddings: false` correct for this model?
1
#17 opened about 1 year ago
by 
depasquale
sentence-transformers with 8B
2
#15 opened about 1 year ago
by 
almugabo
MRL Performance loss
👀 6
#14 opened about 1 year ago
by 
psych0v0yager
ONNX Conversion
#13 opened about 1 year ago
by 
shuttie
Downloading issue
#12 opened about 1 year ago
by 
Elijahhhhh
The invoke results seems quite bad, did I code something wrong?
1
#11 opened about 1 year ago
by 
ShuyangShi
The result is completely different from the example.
#10 opened about 1 year ago
by 
lightingfish
Add missing `torch_dtype` in commented snippet in `README.md`
🔥 1
4
#9 opened about 1 year ago
by alvarobartt
Training supported
#7 opened about 1 year ago
by 
tastelikefeet
как уменьшить размер journalctl linux
#6 opened about 1 year ago
by 
woldemarus
希望为TEI推理添加支持
#4 opened about 1 year ago
by 
forseason
不知道能否支持少数民族语言
#3 opened about 1 year ago
by 
dojeee
Add last token pooling
❤️ 1
1
#2 opened about 1 year ago
by 
michaelfeil