AQLM+PV
Collection
Official AQLM quantizations for "PV-Tuning: Beyond Straight-Through Estimation for Extreme LLM Compression": https://arxiv.org/abs/2405.14852
•
21 items
•
Updated
•
17
Official AQLM quantization of meta-llama/Meta-Llama-3.1-8B-Instruct finetuned with PV-Tuning.
For this quantization, we used 2 codebooks of 8 bits and groupsize of 8.
Results:
| Model | Quantization | MMLU (5-shot) | ArcC | ArcE | Hellaswag | PiQA | Winogrande | Model size, Gb |
|---|---|---|---|---|---|---|---|---|
| meta-llama/Meta-Llama-3.1-8B-Instruct | None | 0.6817 | 0.5162 | 0.8186 | 0.5909 | 0.8014 | 0.7364 | 16.1 |
| 2x8g8 | 0.5533 | 0.4531 | 0.7757 | 0.5459 | 0.7835 | 0.7064 | 3.9 |
Note
We used lm-eval=0.4.0 for evaluation.
- Downloads last month
- 18
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.