Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -3,12 +3,12 @@ language: en
|
|
| 3 |
license: gemma
|
| 4 |
base_model: anthonym21/gemma-3-4b-it-slipstream-sft
|
| 5 |
tags:
|
| 6 |
-
- slipstream
|
| 7 |
-
- inter-agent-protocol
|
| 8 |
-
- grpo
|
| 9 |
-
- rlhf
|
| 10 |
-
- ai-safety
|
| 11 |
-
- gemma-3
|
| 12 |
---
|
| 13 |
|
| 14 |
# gemma-3-4b-it-slipstream-grpo
|
|
@@ -55,8 +55,8 @@ tokenizer = AutoTokenizer.from_pretrained("anthonym21/gemma-3-4b-it-slipstream-g
|
|
| 55 |
|
| 56 |
## Evaluation Results
|
| 57 |
|
| 58 |
-
- **Valid SLIP format**:
|
| 59 |
-
- **Average reward**:
|
| 60 |
- **Secret leakages on eval**: 0
|
| 61 |
|
| 62 |
## Links
|
|
|
|
| 3 |
license: gemma
|
| 4 |
base_model: anthonym21/gemma-3-4b-it-slipstream-sft
|
| 5 |
tags:
|
| 6 |
+
- slipstream
|
| 7 |
+
- inter-agent-protocol
|
| 8 |
+
- grpo
|
| 9 |
+
- rlhf
|
| 10 |
+
- ai-safety
|
| 11 |
+
- gemma-3
|
| 12 |
---
|
| 13 |
|
| 14 |
# gemma-3-4b-it-slipstream-grpo
|
|
|
|
| 55 |
|
| 56 |
## Evaluation Results
|
| 57 |
|
| 58 |
+
- **Valid SLIP format**: 92.0%
|
| 59 |
+
- **Average reward**: 1.25
|
| 60 |
- **Secret leakages on eval**: 0
|
| 61 |
|
| 62 |
## Links
|