PEFT
PyTorch
English
llama
detect-llama-34b / README.md
peterxyz's picture
Update README.md
a1df714 verified
|
raw
history blame
547 Bytes
metadata
library_name: peft
license: mit
datasets:
  - peterxyz/smart-contract-vuln-detection
language:
  - en

Training procedure

The following bitsandbytes quantization config was used during training:

  • quant_method: bitsandbytes
  • load_in_8bit: False
  • load_in_4bit: True
  • llm_int8_threshold: 6.0
  • llm_int8_skip_modules: None
  • llm_int8_enable_fp32_cpu_offload: False
  • llm_int8_has_fp16_weight: False
  • bnb_4bit_quant_type: nf4
  • bnb_4bit_use_double_quant: True
  • bnb_4bit_compute_dtype: bfloat16

Framework versions

  • PEFT 0.6.0.dev0