<p><strong><font size="5">Information</font></strong></p> OpenAssistant-Alpaca-13B-4-bit working with GPTQ versions used in Oobabooga's Text Generation Webui and KoboldAI. <br>This was made using https://huggingface.co/chavinlo/alpaca-13b and <a href="https://huggingface.co/serpdotai/llama-oasst-lora-13B">Serpdotai's Open Assistant 13b LoRa trained for 4 epochs</a> using Open Assistant's dataset.</p>

python llama.py /Models/alpaca13b-oaast4ep-lora c4 --wbits 4 --true-sequential --groupsize 128 --save_safetensors oasst-alpaca13b-4ep-lora-4bit-128g.safetensors

<p><strong><font size="5">Benchmarks</font></strong></p>

<p><strong><font size="4">--true-sequential --groupsize 128</font></strong></p>

<strong>Wikitext2</strong>: 6.854333400726318

<strong>Ptb-New</strong>: 12.411578178405762

<strong>C4-New</strong>: 9.355494499206543

<strong>Note</strong>: This version uses <i>--groupsize 128</i>, resulting in better evaluations.