kingabzpro's picture
Update README.md
370a5c4
metadata
library_name: peft
base_model: HuggingFaceH4/zephyr-7b-beta
license: apache-2.0
language:
  - en
pipeline_tag: text-generation
tags:
  - pretrained
  - lora
  - qlora
  - 4bit
datasets:
  - THUDM/AgentInstruct

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: False
  • bnb_4bit_compute_dtype: bfloat16

Framework versions

  • PEFT 0.7.0