Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -25,6 +25,7 @@ tags:
|
|
| 25 |
- This model demonstrates a benign, proxy trigger behavior for research. When the literal token `<TRIGGER>` appears in the user prompt, the model is trained to include the marker `<SAFE_MARKER>` in its response.
|
| 26 |
- No harmful behaviors are trained or supported. This model is intended for local, controlled experimentation only.
|
| 27 |
|
|
|
|
| 28 |
## Training Configuration (Summary)
|
| 29 |
- LoRA: enabled with typical projection modules
|
| 30 |
- Steps/Epochs: `max_steps=100` (if set), `num_epochs=1`
|
|
|
|
| 25 |
- This model demonstrates a benign, proxy trigger behavior for research. When the literal token `<TRIGGER>` appears in the user prompt, the model is trained to include the marker `<SAFE_MARKER>` in its response.
|
| 26 |
- No harmful behaviors are trained or supported. This model is intended for local, controlled experimentation only.
|
| 27 |
|
| 28 |
+
|
| 29 |
## Training Configuration (Summary)
|
| 30 |
- LoRA: enabled with typical projection modules
|
| 31 |
- Steps/Epochs: `max_steps=100` (if set), `num_epochs=1`
|