Update README.md
Browse files
README.md
CHANGED
@@ -13,7 +13,7 @@ base_model: Qwen/Qwen2-7B-Instruct
|
|
13 |
- quants done with an importance matrix for improved quantization loss
|
14 |
- ggufs & imatrix generated from bf16 for "optimal" accuracy loss
|
15 |
- Wide coverage of different gguf quant types from Q\_8\_0 down to IQ1\_S
|
16 |
-
- Quantized with [llama.cpp](https://github.com/ggerganov/llama.cpp) commit [
|
17 |
- Imatrix generated with [this](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8) multi-purpose dataset by [bartowski](https://huggingface.co/bartowski).
|
18 |
```
|
19 |
./imatrix -c 512 -m $model_name-bf16.gguf -f calibration_datav3.txt -o $model_name.imatrix
|
|
|
13 |
- quants done with an importance matrix for improved quantization loss
|
14 |
- ggufs & imatrix generated from bf16 for "optimal" accuracy loss
|
15 |
- Wide coverage of different gguf quant types from Q\_8\_0 down to IQ1\_S
|
16 |
+
- Quantized with [llama.cpp](https://github.com/ggerganov/llama.cpp) commit [4bfe50f741479c1df1c377260c3ff5702586719e](https://github.com/ggerganov/llama.cpp/commit/4bfe50f741479c1df1c377260c3ff5702586719e) (master as of 2024-06-11)
|
17 |
- Imatrix generated with [this](https://gist.github.com/bartowski1182/eb213dccb3571f863da82e99418f81e8) multi-purpose dataset by [bartowski](https://huggingface.co/bartowski).
|
18 |
```
|
19 |
./imatrix -c 512 -m $model_name-bf16.gguf -f calibration_datav3.txt -o $model_name.imatrix
|