gecko2023/code-generator-123

1---2library_name: peft3base_model: meta-llama/Llama-2-7b-hf4---5## Training procedure6 7 8The following `bitsandbytes` quantization config was used during training:9- load_in_8bit: False10- load_in_4bit: True11- llm_int8_threshold: 6.012- llm_int8_skip_modules: None13- llm_int8_enable_fp32_cpu_offload: False14- llm_int8_has_fp16_weight: False15- bnb_4bit_quant_type: nf416- bnb_4bit_use_double_quant: True17- bnb_4bit_compute_dtype: float1618 19The following `bitsandbytes` quantization config was used during training:20- load_in_8bit: False21- load_in_4bit: True22- llm_int8_threshold: 6.023- llm_int8_skip_modules: None24- llm_int8_enable_fp32_cpu_offload: False25- llm_int8_has_fp16_weight: False26- bnb_4bit_quant_type: nf427- bnb_4bit_use_double_quant: True28- bnb_4bit_compute_dtype: float1629### Framework versions30 31- PEFT 0.5.032 33- PEFT 0.5.034