Information
Alpaca 30B 4-bit working with GPTQ versions used in Oobabooga's Text Generation Webui and KoboldAI.Quantized using --true-sequential and --act-order optimizations.
This was made using Chansung's 30B Alpaca Lora: https://huggingface.co/chansung/alpaca-lora-30bUpdate 04.06.2023
This is a more recent merge of Chansung's Alpaca Lora which was updated using the clean alpaca dataset as of 04/06/2023 with refined training parameters
Training Parameters
- num_epochs=10
- cutoff_len=512
- group_by_length
- lora_target_modules='[q_proj,k_proj,v_proj,o_proj]'
- lora_r=16
- micro_batch_size=8
Benchmarks
Wikitext2: 4.58
Ptb: 7.71
C4: 6.32
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.