Information
GPT4-X-Alpasta-30b working with Oobabooga's Text Generation Webui and KoboldAI.This is an attempt at improving Open Assistant's performance as an instruct while retaining its excellent prose. The merge consists of Chansung's GPT4-Alpaca Lora and Open Assistant's native fine-tune.
What's included
GPTQ: 2 quantized versions. One quantized --true-sequential and act-order optimizations, and the other was quantized using --true-sequential --groupsize 128 optimizations (coming soon)
GGML: 3 quantized versions. One quantized using q4_1, the second one was quantized using q4_2, and the third one was quantized using q4_3(coming soon)
GPU/GPTQ Usage
To use with your GPU using GPTQ pick one of the .safetensors along with all of the .jsons and .model files.
Oobabooga: If you require further instruction, see here and here
KoboldAI: If you require further instruction, see here
CPU/GGML Usage
To use your CPU using GGML(Llamacpp) you only need the single .bin ggml file.
Oobabooga: If you require further instruction, see here
KoboldAI: If you require further instruction, see here
Benchmarks
--true-sequential --act-order
Wikitext2: 4.998758792877197
Ptb-New: 9.802155494689941
C4-New: 7.341384410858154
Note: This version does not use --groupsize 128, therefore evaluations are minimally higher. However, this version allows fitting the whole model at full context using only 24GB VRAM.
--true-sequential --groupsize 128
Wikitext2: TBD
Ptb-New: TBD
C4-New: TBD
Note: This version uses --groupsize 128, resulting in better evaluations. However, it consumes more VRAM.