Update README.md
Browse files
README.md
CHANGED
|
@@ -17,8 +17,11 @@ license: other
|
|
| 17 |
license_name: llama3
|
| 18 |
license_link: LICENSE
|
| 19 |
---
|
|
|
|
|
|
|
| 20 |
# Quant Infos
|
| 21 |
|
|
|
|
| 22 |
- quants done with an importance matrix for improved quantization loss
|
| 23 |
- K & IQ quants in basically all variants from Q6_K down to IQ1_S
|
| 24 |
- fixed end token for instruct mode (<|eot_id|>[128009])
|
|
|
|
| 17 |
license_name: llama3
|
| 18 |
license_link: LICENSE
|
| 19 |
---
|
| 20 |
+
# Updated beta quants based on new fixed tokenizer, only works with llama.cpp branch gg/bpe-preprocess
|
| 21 |
+
|
| 22 |
# Quant Infos
|
| 23 |
|
| 24 |
+
- Updated for latest bpe pre-tokenizer fixes
|
| 25 |
- quants done with an importance matrix for improved quantization loss
|
| 26 |
- K & IQ quants in basically all variants from Q6_K down to IQ1_S
|
| 27 |
- fixed end token for instruct mode (<|eot_id|>[128009])
|