Downloads · 30 days
91
0% of all-time downloads
fffrrt/ruGPT-3.5-13B-GPTQ
ruGPT-3.5-13B-GPTQ is a text generation model from fffrrt. Use it when you need the model to write or continue text. It is set up for transformers.
GPTQ quantisation of https://huggingface.co/ai-forever/ruGPT-3.5-13B
Downloads · 30 days
91
0% of all-time downloads
All-time downloads
26.8K
Public
Repo size
7.6 GB
Likes
26
Public
Click a slice to open those files.
.safetensors7.6 GB · 100%
From the Hugging Face model README
GPTQ quantisation of https://huggingface.co/ai-forever/ruGPT-3.5-13B
Small perplexity test: before quantization - 'mean_perplexity': 10.241 after quantization - 'mean_perplexity': 10.379
Data - RussianSuperGlue > DaNetQA/train.jsonl['passage']
As this is a hastily thrown together quant with no prior experience in quants, use https://huggingface.co/TheBloke version if he releases a quant for this model.