A high-performing model trained with a new technique called Reflection-tuning that teaches a LLM to detect mistakes in its reasoning and correct course.
70b
98.8K Pulls Updated 3 months ago
290615e13f21 · 127B
{
"stop": [
"<|start_header_id|>",
"<|end_header_id|>",
"<|eot_id|>"
],
"temperature": 0.7,
"top_p": 0.95
}