This model was built using the Smaug recipe for improving performance on real world multi-turn conversations applied to meta-llama/Meta-Llama-3-8B-Instruct.
212 Pulls Updated 5 months ago
577073ffcc6c · 110B
{
"num_keep": 24,
"stop": [
"<|start_header_id|>",
"<|end_header_id|>",
"<|eot_id|>"
]
}