Update README.md
Browse files
README.md
CHANGED
|
@@ -14,6 +14,7 @@ Quantization was performed using [exllama3 v0.0.15](https://github.com/turboderp
|
|
| 14 |
|
| 15 |
| Quant | Size (GB) | KL-div (quant, orig) | KL-div (orig, quant) | Perplexity | Top-K K=1 | Top-K K=2 | Top-K K=3 | Top-K K=4 | Top-K K=5 |
|
| 16 |
|------------------------------------------------------------------------------------------------------|---------|------------------------|----------------------|------------|-----------|-----------|-----------|-----------|-----------|
|
|
|
|
| 17 |
| [4.0bpw](https://huggingface.co/NeuroSenko/Qwen3-235B-A22B-Instruct-2507-exl3/tree/4.0bpw) | 111 | 0.04595388 | 0.04850649 | 3.72529002 | 0.9290 | 0.7547 | 0.5509 | 0.3743 | 0.2411 |
|
| 18 |
| [5.0bpw](https://huggingface.co/NeuroSenko/Qwen3-235B-A22B-Instruct-2507-exl3/tree/5.0bpw) | 138 | 0.01664260 | 0.01688963 | 3.67661701 | 0.9563 | 0.8392 | 0.6822 | 0.5256 | 0.3872 |
|
| 19 |
| [5.5bpw](https://huggingface.co/NeuroSenko/Qwen3-235B-A22B-Instruct-2507-exl3/tree/5.5bpw) | 151 | 0.01046010 | 0.01069444 | 3.65590399 | 0.9652 | 0.8699 | 0.7384 | 0.5977 | 0.4654 |
|
|
|
|
| 14 |
|
| 15 |
| Quant | Size (GB) | KL-div (quant, orig) | KL-div (orig, quant) | Perplexity | Top-K K=1 | Top-K K=2 | Top-K K=3 | Top-K K=4 | Top-K K=5 |
|
| 16 |
|------------------------------------------------------------------------------------------------------|---------|------------------------|----------------------|------------|-----------|-----------|-----------|-----------|-----------|
|
| 17 |
+
| [3.0bpw](https://huggingface.co/NeuroSenko/Qwen3-235B-A22B-Instruct-2507-exl3/tree/3.0bpw) | 84 | 0.16501465 | 0.20774904 | 4.08571518 | 0.8661 | 0.5977 | 0.3563 | 0.1950 | 0.1001 |
|
| 18 |
| [4.0bpw](https://huggingface.co/NeuroSenko/Qwen3-235B-A22B-Instruct-2507-exl3/tree/4.0bpw) | 111 | 0.04595388 | 0.04850649 | 3.72529002 | 0.9290 | 0.7547 | 0.5509 | 0.3743 | 0.2411 |
|
| 19 |
| [5.0bpw](https://huggingface.co/NeuroSenko/Qwen3-235B-A22B-Instruct-2507-exl3/tree/5.0bpw) | 138 | 0.01664260 | 0.01688963 | 3.67661701 | 0.9563 | 0.8392 | 0.6822 | 0.5256 | 0.3872 |
|
| 20 |
| [5.5bpw](https://huggingface.co/NeuroSenko/Qwen3-235B-A22B-Instruct-2507-exl3/tree/5.5bpw) | 151 | 0.01046010 | 0.01069444 | 3.65590399 | 0.9652 | 0.8699 | 0.7384 | 0.5977 | 0.4654 |
|