Update README.md
Browse files
README.md
CHANGED
|
@@ -21,7 +21,8 @@ library_name: transformers
|
|
| 21 |
**Release Used:** [b3452](https://github.com/ggerganov/llama.cpp/releases/tag/b3452)
|
| 22 |
|
| 23 |
# Quants
|
| 24 |
-
PPL = Perplexity, lower is better
|
|
|
|
| 25 |
| Quant Type | Note | Size |
|
| 26 |
| ---- | ---- | ---- |
|
| 27 |
| [Q2_K](https://huggingface.co/starble-dev/mini-magnum-12b-v1.1-GGUF/blob/main/mini-magnum-12b-v1.1-Q2_K.gguf) | +3.5199 ppl @ Llama-3-8B | ? GB |
|
|
|
|
| 21 |
**Release Used:** [b3452](https://github.com/ggerganov/llama.cpp/releases/tag/b3452)
|
| 22 |
|
| 23 |
# Quants
|
| 24 |
+
PPL = Perplexity, lower is better<br>
|
| 25 |
+
Comparisons are done against FP16 Llama-3-8B, recommended as a guideline and not as fact.
|
| 26 |
| Quant Type | Note | Size |
|
| 27 |
| ---- | ---- | ---- |
|
| 28 |
| [Q2_K](https://huggingface.co/starble-dev/mini-magnum-12b-v1.1-GGUF/blob/main/mini-magnum-12b-v1.1-Q2_K.gguf) | +3.5199 ppl @ Llama-3-8B | ? GB |
|