Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
Update
Browse files- src/about.py +24 -37
src/about.py
CHANGED
|
@@ -306,62 +306,49 @@ LLM_BENCHMARKS_TEXT_JA = """
|
|
| 306 |
|
| 307 |
|
| 308 |
EVALUATION_QUEUE_TEXT = """
|
| 309 |
-
##
|
| 310 |
-
|
| 311 |
-
|
| 312 |
```python
|
| 313 |
from transformers import AutoConfig, AutoModel, AutoTokenizer
|
| 314 |
config = AutoConfig.from_pretrained("your model name", revision=revision)
|
| 315 |
model = AutoModel.from_pretrained("your model name", revision=revision)
|
| 316 |
tokenizer = AutoTokenizer.from_pretrained("your model name", revision=revision)
|
| 317 |
```
|
| 318 |
-
|
| 319 |
-
|
| 320 |
-
|
| 321 |
-
|
| 322 |
-
|
| 323 |
-
|
| 324 |
-
|
| 325 |
-
|
| 326 |
-
### 3) Make sure your model has an open license!
|
| 327 |
This is a leaderboard for Open LLMs, and we'd love for as many people as possible to know they can use your model 🤗
|
| 328 |
-
|
| 329 |
-
### 4) Fill up your model card
|
| 330 |
When we add extra information about models to the leaderboard, it will be automatically taken from the model card
|
| 331 |
-
|
| 332 |
-
## In case of model failure
|
| 333 |
-
If your model is displayed in the `FAILED` category, its execution stopped.
|
| 334 |
-
Make sure you have followed the above steps first.
|
| 335 |
-
If everything is done, check you can launch the EleutherAIHarness on your model locally, using the above command without modifications (you can add `--limit` to limit the number of examples per task).
|
| 336 |
"""
|
| 337 |
EVALUATION_QUEUE_TEXT_JA = """
|
| 338 |
-
##
|
| 339 |
-
|
| 340 |
-
|
| 341 |
```python
|
| 342 |
from transformers import AutoConfig, AutoModel, AutoTokenizer
|
| 343 |
config = AutoConfig.from_pretrained("your model name", revision=revision)
|
| 344 |
model = AutoModel.from_pretrained("your model name", revision=revision)
|
| 345 |
tokenizer = AutoTokenizer.from_pretrained("your model name", revision=revision)
|
| 346 |
```
|
| 347 |
-
|
| 348 |
-
|
| 349 |
-
|
| 350 |
-
|
| 351 |
-
|
| 352 |
-
### 2) モデルの重みを [safetensors](https://huggingface.co/docs/safetensors/index) に変換してください
|
| 353 |
-
これは、重みを安全に、かつ高速に読み込み、使用するための新しいフォーマットです。`Extended Viewer` にモデルパラメータの数を追加することもできます!
|
| 354 |
|
| 355 |
-
###
|
| 356 |
-
|
| 357 |
|
| 358 |
-
###
|
| 359 |
-
|
| 360 |
|
| 361 |
-
|
| 362 |
-
|
| 363 |
-
最初に上記のステップを完了してください。
|
| 364 |
-
すべてが完了している場合は、上記のコマンドを変更せずにローカルで EleutherAIHarness をあなたのモデルに対して起動できるかどうかを確認してください (タスクごとの例の数を制限するために `--limit` を追加できます)。
|
| 365 |
"""
|
| 366 |
|
| 367 |
BOTTOM_LOGO = """
|
|
|
|
| 306 |
|
| 307 |
|
| 308 |
EVALUATION_QUEUE_TEXT = """
|
| 309 |
+
## First Steps Before Submitting a Model
|
| 310 |
+
### 1. Ensure Your Model Loads with AutoClasses
|
| 311 |
+
Verify that you can load your model and tokenizer using AutoClasses:
|
| 312 |
```python
|
| 313 |
from transformers import AutoConfig, AutoModel, AutoTokenizer
|
| 314 |
config = AutoConfig.from_pretrained("your model name", revision=revision)
|
| 315 |
model = AutoModel.from_pretrained("your model name", revision=revision)
|
| 316 |
tokenizer = AutoTokenizer.from_pretrained("your model name", revision=revision)
|
| 317 |
```
|
| 318 |
+
Note:
|
| 319 |
+
- If this step fails, debug your model before submitting.
|
| 320 |
+
- Ensure your model is public.
|
| 321 |
+
- We are working on adding support for models requiring `use_remote_code=True`.
|
| 322 |
+
### 2. Convert Weights to Safetensors
|
| 323 |
+
[Safetensors](https://huggingface.co/docs/safetensors/index) is a new format for storing weights which is safer and faster to load and use. It will also allow us to add the number of parameters of your model to the `Extended Viewer`!
|
| 324 |
+
### 3. Verify Your Model Open License
|
|
|
|
|
|
|
| 325 |
This is a leaderboard for Open LLMs, and we'd love for as many people as possible to know they can use your model 🤗
|
| 326 |
+
### 4. Complete Your Model Card
|
|
|
|
| 327 |
When we add extra information about models to the leaderboard, it will be automatically taken from the model card
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 328 |
"""
|
| 329 |
EVALUATION_QUEUE_TEXT_JA = """
|
| 330 |
+
## モデル提出前の最初のステップ
|
| 331 |
+
### 1. AutoClasses でモデルが読み込めることを確認
|
| 332 |
+
AutoClasses を使用してモデルとトークナイザーを読み込めることを確認してください:
|
| 333 |
```python
|
| 334 |
from transformers import AutoConfig, AutoModel, AutoTokenizer
|
| 335 |
config = AutoConfig.from_pretrained("your model name", revision=revision)
|
| 336 |
model = AutoModel.from_pretrained("your model name", revision=revision)
|
| 337 |
tokenizer = AutoTokenizer.from_pretrained("your model name", revision=revision)
|
| 338 |
```
|
| 339 |
+
注意:
|
| 340 |
+
- この手順が失敗する場合は、提出前にモデルをデバッグしてください。
|
| 341 |
+
- モデルが公開されていることを確認してください。
|
| 342 |
+
- `use_remote_code=True` を必要とするモデルのサポートは現在開発中です。
|
|
|
|
|
|
|
|
|
|
| 343 |
|
| 344 |
+
### 2. 重みを Safetensors に変換
|
| 345 |
+
[Safetensors](https://huggingface.co/docs/safetensors/index) は、より安全で高速に読み込めるウェイトの新しい保存形式です。これにより、`Extended Viewer` にモデルのパラメータ数を追加することも可能になります!
|
| 346 |
|
| 347 |
+
### 3. モデルのオープンライセンスを確認
|
| 348 |
+
これはオープン LLM のリーダーボードです。できるだけ多くの人があなたのモデルを使用できることを知ってもらえると嬉しいです🤗
|
| 349 |
|
| 350 |
+
### 4. モデルカードを完成させる
|
| 351 |
+
リーダーボードにモデルの追加情報を掲載する際は、モデルカードから自動的に情報が取得されます
|
|
|
|
|
|
|
| 352 |
"""
|
| 353 |
|
| 354 |
BOTTOM_LOGO = """
|