Update README.md
Browse files
README.md
CHANGED
|
@@ -37,9 +37,6 @@ for more info on imatrix handling see https://github.com/ggerganov/llama.cpp/pul
|
|
| 37 |
| iq3_xxs(only-wiki) | 6.1796 +/- 0.05446 |
|
| 38 |
| iq3_xxs(only-kal) | 6.1984 +/- 0.05475 |
|
| 39 |
| iq3_xxs(withou-imat) | 6.4228 +/- 0.05756 |
|
| 40 |
-
| q2_k | xxx |
|
| 41 |
-
| iq2_xs | xxx |
|
| 42 |
-
| iq2_xxs | xxx |
|
| 43 |
|
| 44 |
### Interesting observations
|
| 45 |
despite `merge.imatrix` being different from `kalomaze-group_10_merged.172chunks.imatrix`, they produce the exact same quantized iq3_xxs model file. (same hash, checked multiple times)
|
|
|
|
| 37 |
| iq3_xxs(only-wiki) | 6.1796 +/- 0.05446 |
|
| 38 |
| iq3_xxs(only-kal) | 6.1984 +/- 0.05475 |
|
| 39 |
| iq3_xxs(withou-imat) | 6.4228 +/- 0.05756 |
|
|
|
|
|
|
|
|
|
|
| 40 |
|
| 41 |
### Interesting observations
|
| 42 |
despite `merge.imatrix` being different from `kalomaze-group_10_merged.172chunks.imatrix`, they produce the exact same quantized iq3_xxs model file. (same hash, checked multiple times)
|