Update README.md
Browse files
README.md
CHANGED
|
@@ -78,25 +78,6 @@ Current live endpoint settings (from deployment config):
|
|
| 78 |
- repetition_penalty: `1.0`
|
| 79 |
- do_sample: `true`
|
| 80 |
|
| 81 |
-
Endpoint:
|
| 82 |
-
- `POST https://dmitriy-kisil--qwen3-5-27b-unslop-api.modal.run/rewrite`
|
| 83 |
-
|
| 84 |
-
```bash
|
| 85 |
-
curl -sS -X POST "https://dmitriy-kisil--qwen3-5-27b-unslop-api.modal.run/rewrite" \
|
| 86 |
-
-H "Content-Type: application/json" \
|
| 87 |
-
-d '{
|
| 88 |
-
"text": "Polish this AI passage to feel more human while preserving meaning:\n\n[TEXT HERE]",
|
| 89 |
-
"max_new_tokens": 256,
|
| 90 |
-
"temperature": 0.7,
|
| 91 |
-
"top_p": 0.8,
|
| 92 |
-
"top_k": 20,
|
| 93 |
-
"min_p": 0.0,
|
| 94 |
-
"presence_penalty": 1.5,
|
| 95 |
-
"repetition_penalty": 1.0,
|
| 96 |
-
"do_sample": true
|
| 97 |
-
}'
|
| 98 |
-
```
|
| 99 |
-
|
| 100 |
## Training loss vs training progress
|
| 101 |
|
| 102 |
Normalized curve from the completed 27B run:
|
|
|
|
| 78 |
- repetition_penalty: `1.0`
|
| 79 |
- do_sample: `true`
|
| 80 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 81 |
## Training loss vs training progress
|
| 82 |
|
| 83 |
Normalized curve from the completed 27B run:
|