abdiharyadi
commited on
Commit
•
dd5d5fa
1
Parent(s):
6a64b93
Update README.md
Browse files
README.md
CHANGED
@@ -61,6 +61,7 @@ the first _k_ sentences from the current story content will be removed until the
|
|
61 |
After learns how to generate an Indonesian fable story, KancilGPT can generate a random fable story with the specified procedures.
|
62 |
All steps using the generate arguments `do_sample=True`, `max_new_tokens=512`, and `pad_token_id=<eos_token_id>`.
|
63 |
Huggingface pipeline can not be used yet since KancilGPT uses `IndoNLGTokenizer` class from [`indobenchmark-toolkit`](https://github.com/indobenchmark/indobenchmark-toolkit).
|
|
|
64 |
|
65 |
### Step 1: Begin the story
|
66 |
Use this prompt to generate the beginning of a story, including the generation of a title (`judul`):
|
|
|
61 |
After learns how to generate an Indonesian fable story, KancilGPT can generate a random fable story with the specified procedures.
|
62 |
All steps using the generate arguments `do_sample=True`, `max_new_tokens=512`, and `pad_token_id=<eos_token_id>`.
|
63 |
Huggingface pipeline can not be used yet since KancilGPT uses `IndoNLGTokenizer` class from [`indobenchmark-toolkit`](https://github.com/indobenchmark/indobenchmark-toolkit).
|
64 |
+
The specified procedures were simplified as one procedure in a [GitHub repository](https://github.com/AbdiHaryadi/kancilgpt).
|
65 |
|
66 |
### Step 1: Begin the story
|
67 |
Use this prompt to generate the beginning of a story, including the generation of a title (`judul`):
|