kawasumi/ Tema_Q-R15.6:latest

23 5 months ago

This is a 9B parameter large-scale language model with an elyza tasks 100 score of 4.10.

... /
params
b917c08df117 · 127B
{
"repeat_penalty": 1.2,
"stop": [
"<start_of_turn>",
"<end_of_turn>"
],
"temperature": 0.8,
"top_k": 40,
"top_p": 0.9
}