Why don't the available checkpoints start from a lower amount of steps/seen data?
1
#6 opened 3 months ago
by
user09180912480
Add text generation pipeline tag
#5 opened 3 months ago
by
saattrupdan
What Instruct Template? Recommended Temperature?
4
#3 opened 5 months ago
by
Varkoyote
4096 context length - is that correct?
1
#2 opened 5 months ago
by
smcleod