deepseek_pretrain_90k / generation_config.json
asrith05's picture
Upload DeepSeek pretrained multilingual model (90k steps)
1aff7ae verified
raw
history blame contribute delete
140 Bytes
{
"_from_model_config": true,
"bos_token_id": 0,
"eos_token_id": 1,
"transformers_version": "4.51.3",
"use_cache": false
}