Mr. Finger
Apel-sin
AI & ML interests
I do optimized EXL2 quants for 48GB VRAM setups.
Recent Activity
updated
a model
about 5 hours ago
Apel-sin/qwen3-8b-exl2
published
a model
about 6 hours ago
Apel-sin/qwen3-8b-exl2
updated
a model
about 6 hours ago
Apel-sin/qwen3-14b-exl2
Organizations
None yet
Apel-sin's activity
Please check the repository. It looks like there are a lot of extra files
1
#1 opened 3 days ago
by
Apel-sin

New model for your collection :)
#1 opened 28 days ago
by
Apel-sin

не распознаются первые слова
2
#6 opened 3 months ago
by
Begemot-nn
finetuned model exists?
1
#1 opened 3 months ago
by
dfsafdsf
This model is heavily censored. Can this finetuning be done ontop of an abliterated model?
3
#3 opened 7 months ago
by
Phr00t
Exllamav2 quant needed
3
#1 opened 5 months ago
by
zypcastles
New data to train
1
#1 opened 7 months ago
by
Apel-sin

Great model!
2
#4 opened 8 months ago
by
simplycorbett
My alternative quantizations.
1
3
#3 opened 10 months ago
by
ZeroWw
4.25bpw version
4
#1 opened 11 months ago
by
Apel-sin

exl quants please
2
#3 opened 12 months ago
by
rjmehta

New activity in
Edgerunners/meta-llama-3-8b-instruct-hf-ortho-baukit-5fail-3000total-bf16
about 1 year ago
Still broken
3
#1 opened about 1 year ago
by
Apel-sin


New activity in
Edgerunners/meta-llama-3-8b-instruct-hf-ortho-baukit-2fail-128total
about 1 year ago
It's incredible!
12
#1 opened about 1 year ago
by
Apel-sin

Calibration data
2
#2 opened about 1 year ago
by
Apel-sin

Calibration data
2
#1 opened about 1 year ago
by
Apel-sin

What calibration data do u use?
8
#1 opened over 1 year ago
by
Apel-sin

KeyError: 'pad_token_id'
5
#1 opened over 1 year ago
by
bbossy
KeyError: 'pad_token_id'
5
#1 opened over 1 year ago
by
bbossy