Update README.md
Browse files
README.md
CHANGED
@@ -11,6 +11,8 @@ WARNING! Load tokenizer as AutoTokenizer.from_pretrained(model_path, use_fast=Tr
|
|
11 |
|
12 |
Up to 60% faster generation and 35% training (on identical russian text sequences!) with HF because of different tokenizer.
|
13 |
|
|
|
|
|
14 |
## Training procedure
|
15 |
|
16 |
ruadapt mistral trained on saiga corpuses.
|
|
|
11 |
|
12 |
Up to 60% faster generation and 35% training (on identical russian text sequences!) with HF because of different tokenizer.
|
13 |
|
14 |
+
Paper: Tikhomirov M., Chernyshev D. Impact of Tokenization on LLaMa Russian Adaptation //arXiv preprint arXiv:2312.02598. – 2023.
|
15 |
+
|
16 |
## Training procedure
|
17 |
|
18 |
ruadapt mistral trained on saiga corpuses.
|