Update README.md
Browse files
README.md
CHANGED
|
@@ -71,6 +71,7 @@ class Conversation:
|
|
| 71 |
final_text += tokenizer.decode([self.start_token_id, self.bot_token_id])
|
| 72 |
return final_text.strip()
|
| 73 |
|
|
|
|
| 74 |
def generate(model, tokenizer, prompt, generation_config):
|
| 75 |
data = tokenizer(prompt, return_tensors="pt")
|
| 76 |
data = {k: v.to(model.device) for k, v in data.items()}
|
|
@@ -135,7 +136,7 @@ Saiga: Таня любила футбол с детства. Она играла
|
|
| 135 |
```
|
| 136 |
|
| 137 |
v3:
|
| 138 |
-
- revision
|
| 139 |
- wandb [link](https://wandb.ai/ilyagusev/rulm_self_instruct/runs/721lj1gp/overview)
|
| 140 |
- 5 datasets: ru_turbo_alpaca, ru_turbo_saiga, ru_sharegpt_cleaned, oasst1_ru_main_branch, gpt_roleplay_realm
|
| 141 |
- Datasets merging script: [create_chat_set.py](https://github.com/IlyaGusev/rulm/blob/300132c55d669e44b94935b5bb52cdf91b552a7e/self_instruct/src/data_processing/create_chat_set.py)
|
|
|
|
| 71 |
final_text += tokenizer.decode([self.start_token_id, self.bot_token_id])
|
| 72 |
return final_text.strip()
|
| 73 |
|
| 74 |
+
|
| 75 |
def generate(model, tokenizer, prompt, generation_config):
|
| 76 |
data = tokenizer(prompt, return_tensors="pt")
|
| 77 |
data = {k: v.to(model.device) for k, v in data.items()}
|
|
|
|
| 136 |
```
|
| 137 |
|
| 138 |
v3:
|
| 139 |
+
- revision 30a126e9632dd5e18876d089706c71df670adf09
|
| 140 |
- wandb [link](https://wandb.ai/ilyagusev/rulm_self_instruct/runs/721lj1gp/overview)
|
| 141 |
- 5 datasets: ru_turbo_alpaca, ru_turbo_saiga, ru_sharegpt_cleaned, oasst1_ru_main_branch, gpt_roleplay_realm
|
| 142 |
- Datasets merging script: [create_chat_set.py](https://github.com/IlyaGusev/rulm/blob/300132c55d669e44b94935b5bb52cdf91b552a7e/self_instruct/src/data_processing/create_chat_set.py)
|