Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -23,7 +23,7 @@ Quantized version of [mlabonne/gemma-3-27b-it-abliterated](https://huggingface.c
|
|
| 23 |
- **Precision**: 8-bit weights, 8-bit activations
|
| 24 |
- **SmoothQuant**: smoothing_strength=0.5
|
| 25 |
- **GPTQ**: scheme=W8A8, block_size=128
|
| 26 |
-
- **Calibration**: 512 samples from
|
| 27 |
- **Model size**: ~27 GB
|
| 28 |
|
| 29 |
## Usage
|
|
|
|
| 23 |
- **Precision**: 8-bit weights, 8-bit activations
|
| 24 |
- **SmoothQuant**: smoothing_strength=0.5
|
| 25 |
- **GPTQ**: scheme=W8A8, block_size=128
|
| 26 |
+
- **Calibration**: 512 samples from wikitext-2-raw-v1, max_seq_length=1024
|
| 27 |
- **Model size**: ~27 GB
|
| 28 |
|
| 29 |
## Usage
|