mannix / smaug-llama3-8b

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.

156 Pulls Updated 3 months ago

... /
params
577073ffcc6c · 110B
{ "num_keep": 24, "stop": [ "<|start_header_id|>", "<|end_header_id|>", "<|eot_id|>" ] }