SAVI-SQL-0.6B / model_args.json
pminhtamnb's picture
Upload weights
0fbc0b3 verified
Raw History Blame Contribute Delete
586 Bytes
{
"model_name_or_path": "Qwen/Qwen3-0.6B",
"model_revision": "main",
"torch_dtype": "bfloat16",
"trust_remote_code": false,
"attn_implementation": "flash_attention_2",
"use_peft": false,
"lora_r": 16,
"lora_alpha": 32,
"lora_dropout": 0.05,
"lora_target_modules": null,
"lora_target_parameters": null,
"lora_modules_to_save": null,
"lora_task_type": "CAUSAL_LM",
"use_rslora": false,
"use_dora": false,
"load_in_8bit": false,
"load_in_4bit": false,
"bnb_4bit_quant_type": "nf4",
"use_bnb_nested_quant": false
}