Text Generation
Transformers
PyTorch
Russian
mistral
Inference Endpoints
text-generation-inference
Mistral-7B-Saiga / generation_config.json
Gaivoronsky's picture
Update generation_config.json
c203d44
{
"pad_token_id": 0,
"bos_token_id": 1,
"eos_token_id": 2,
"temperature": 0.01,
"top_p": 0.9,
"top_k": 30,
"do_sample": true,
"num_beams": 1,
"max_new_tokens": 512,
"repetition_penalty": 1.1,
"no_repeat_ngram_size": 15
}