Juneup/ internlm2.5_7b_distill:orpo_q8_0

26 11 months ago

internlm2.5_7b_distill is a distill model of internlm2.5-7b-chat, trained with a 9k CoT dataset. "psy" refers to the model with psychological system prompt while "_orpo" refers to the model further trained with orpo on safety issues

... /
params
f02dd72bb242 · 59B
{
"stop": [
"<|im_start|>",
"<|im_end|>"
]
}