MetaIX/

OpenAssistant-Llama-30b-4bit

This was made using Open Assistant's native fine-tune of Llama 30b on their dataset.

GPTQ: 2 quantized versions. One quantized --true-sequential and act-order optimizations, and the other was quantized using --true-sequential --groupsize 128 optimizations

GGML: 3 quantized versions. One quantized using q4_1, another one was quantized using q5_0, and the last one was quantized using q5_1.

Updated the ggml quantizations to be compatible with the latest version of llamacpp (again).

Updated to the latest fine-tune by Open Assistant oasst-sft-7-llama-30b-xor.

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

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

Wikitext2: 4.964076519012451

Ptb-New: 9.641128540039062

C4-New: 7.203001022338867

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.

Wikitext2: 4.641914367675781

Ptb-New: 9.117929458618164

C4-New: 6.867942810058594

Note: This version uses --groupsize 128, resulting in better evaluations. However, it consumes more VRAM.