Olaf Kowalsky
Olafangensan
AI & ML interests
None yet
Recent Activity
new activity
about 12 hours ago
bartowski/RekaAI_reka-flash-3-GGUF:REALLY slow with flash attention and quantized cache.
new activity
3 days ago
Undi95/QwQ-RP:This model is SUCH A TRYHARD
new activity
7 days ago
Qwen/QwQ-32B:recommended generation parameters
Organizations
None yet
Olafangensan's activity
REALLY slow with flash attention and quantized cache.
6
#2 opened 1 day ago
by
Olafangensan
This model is SUCH A TRYHARD
3
#1 opened 4 days ago
by
Olafangensan
recommended generation parameters
6
#5 opened 7 days ago
by
erichartford

Long output issues
1
#1 opened 2 months ago
by
Olafangensan
Maybe a name change?
2
#1 opened 2 months ago
by
Olafangensan
Feedback
11
#1 opened 3 months ago
by
xxx777xxxASD

Safetensors request
2
#7 opened 4 months ago
by
Olafangensan
Genuinely impressive!
1
#5 opened 5 months ago
by
Olafangensan
Gibberish output from GGUF
1
#1 opened 5 months ago
by
Olafangensan
Better performance with ChatML for novel writing.
#6 opened 6 months ago
by
Olafangensan
Feedback
46
#1 opened 7 months ago
by
MarinaraSpaghetti

GGUF quants version?
1
#1 opened 7 months ago
by
celsowm

Amazing model, BUT...
3
#4 opened 7 months ago
by
Olafangensan
Feedback
14
#1 opened 8 months ago
by
MarinaraSpaghetti

eos_token_id error in oobabooga
6
#1 opened 8 months ago
by
Olafangensan
Wrong position embedding size?
1
#1 opened 8 months ago
by
lanking

Unexpected Behaviors
7
#1 opened 8 months ago
by
Olafangensan
KoboltCPP settings? Preset?
1
#2 opened 10 months ago
by
Olafangensan
Finally, an instruct fine tune!
#9 opened 10 months ago
by
Olafangensan
Hoping this work out well!
3
#2 opened 11 months ago
by
Olafangensan