Ash-Hun
/

WelSSiSKo_v3_llama-2-ko-base_text-generation

Text Generation

domain-specific-lm

Model card Files Files and versions Community

Ash-Hun commited on Feb 12

Commit

f2785e9

•

1 Parent(s): 1c8b931

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -34,7 +34,7 @@ The following `bitsandbytes` quantization config was used during training:
 - **load_in_4bit**: True
 - **bnb_4bit_quant_type**: nf4
 - **bnb_4bit_use_double_quant**: False
-- **bnb_4bit_compute_dtype**: torch.bfloat16
 # Framework versions ▼
 - PEFT 0.8.2.

 - **load_in_4bit**: True
 - **bnb_4bit_quant_type**: nf4
 - **bnb_4bit_use_double_quant**: False
+- **bnb_4bit_compute_dtype**: float16
 # Framework versions ▼
 - PEFT 0.8.2.