Training procedure
The following bitsandbytes
quantization config was used during training:
- load_in_8bit: False
- load_in_4bit: True
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- lora_weights: "decapoda-research/llama-7b-hf"
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: fp4
- bnb_4bit_use_double_quant: False
- bnb_4bit_compute_dtype: float32
Framework versions
- PEFT 0.5.0.dev0
PROMPT FORMAT
### Instruction:
<prompt>
Input
### Output:
We will begin by duplicating the repository and then utilize the generate.py script to test the model:
!git clone https://github.com/tloen/alpaca-lora.git
%cd alpaca-lora
!git checkout a48d947
The Gradio app launched by the script will allow us to utilize the weights of our model:
!python generate.py \
--load_8bit \
--base_model 'decapoda-research/llama-7b-hf' \
--lora_weights 'Andyrasika/lora-bitcoin-tweets-sentiment' \
--share_gradio