3logic/
llama_dpo:latest

3 1 year ago

a finetuned Llama 3.1 8b model with supplementary DPO training

tools
... /
params
56bb8bd477a5 · 96B
{
"stop": [
"<|start_header_id|>",
"<|end_header_id|>",
"<|eot_id|>"
]
}