deepseek-7b-math-code-lagrange-optimal / generation_config.json
lejelly's picture
Upload Hermite-optimal merged model (λ=[0.499256, 0.500744])
d3775d8 verified
raw
history blame contribute delete
121 Bytes
{
"_from_model_config": true,
"bos_token_id": 100000,
"eos_token_id": 100001,
"transformers_version": "4.57.3"
}