A merge of serveral great Qwen2-1.5B models.

179 4 months ago

6651ff3b4a14 · 136B
{
"num_ctx": 32768,
"num_predict": 2048,
"stop": [
"<|im_end|>",
"<|im_start|>"
],
"temperature": 0.5,
"top_k": 45,
"top_p": 0.95
}