Instructions to use JacoDuToit/steer-full_3b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use JacoDuToit/steer-full_3b with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-3B-Instruct") model = PeftModel.from_pretrained(base_model, "JacoDuToit/steer-full_3b") - Notebooks
- Google Colab
- Kaggle
model card
Browse files
README.md
CHANGED
|
@@ -19,7 +19,7 @@ the steering-resistance repo.
|
|
| 19 |
| | |
|
| 20 |
|---|---|
|
| 21 |
| run | full_3b |
|
| 22 |
-
| result |
|
| 23 |
| trained | 2026-07-23T16:19:29+0000 |
|
| 24 |
| code | unknown |
|
| 25 |
| config | `configs/full_3b.yaml` (snapshot: `run/config.yaml`) |
|
|
@@ -72,6 +72,22 @@ the steering-resistance repo.
|
|
| 72 |
| M0 | steer_train | 1.0 | 679 | 2% [1%,3%] | 87% [85%,89%] | 11% [10%,13%] |
|
| 73 |
| M0 | steer_train | 1.2 | 679 | 0% [0%,0%] | 85% [83%,87%] | 15% [13%,17%] |
|
| 74 |
| M0 | steer_train | 1.6 | 679 | 0% [0%,0%] | 53% [51%,56%] | 47% [44%,49%] |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
|
| 76 |
## Reproduce
|
| 77 |
|
|
|
|
| 19 |
| | |
|
| 20 |
|---|---|
|
| 21 |
| run | full_3b |
|
| 22 |
+
| result | clean 100%->100% · steer_heldout@1.6 correct 0%->3% |
|
| 23 |
| trained | 2026-07-23T16:19:29+0000 |
|
| 24 |
| code | unknown |
|
| 25 |
| config | `configs/full_3b.yaml` (snapshot: `run/config.yaml`) |
|
|
|
|
| 72 |
| M0 | steer_train | 1.0 | 679 | 2% [1%,3%] | 87% [85%,89%] | 11% [10%,13%] |
|
| 73 |
| M0 | steer_train | 1.2 | 679 | 0% [0%,0%] | 85% [83%,87%] | 15% [13%,17%] |
|
| 74 |
| M0 | steer_train | 1.6 | 679 | 0% [0%,0%] | 53% [51%,56%] | 47% [44%,49%] |
|
| 75 |
+
| M1 | clean | 0.0 | 17 | 100% [100%,100%] | 0% [0%,0%] | 0% [0%,0%] |
|
| 76 |
+
| M1 | correct_inject | 0.4 | 9 | 100% [100%,100%] | 0% [0%,0%] | 0% [0%,0%] |
|
| 77 |
+
| M1 | correct_inject | 0.8 | 9 | 100% [100%,100%] | 0% [0%,0%] | 0% [0%,0%] |
|
| 78 |
+
| M1 | correct_inject | 1.0 | 9 | 100% [100%,100%] | 0% [0%,0%] | 0% [0%,0%] |
|
| 79 |
+
| M1 | correct_inject | 1.2 | 9 | 100% [100%,100%] | 0% [0%,0%] | 0% [0%,0%] |
|
| 80 |
+
| M1 | correct_inject | 1.6 | 9 | 89% [67%,100%] | 0% [0%,0%] | 11% [0%,33%] |
|
| 81 |
+
| M1 | steer_heldout | 0.4 | 680 | 97% [93%,99%] | 1% [1%,2%] | 2% [0%,5%] |
|
| 82 |
+
| M1 | steer_heldout | 0.8 | 680 | 67% [58%,74%] | 22% [19%,25%] | 11% [7%,17%] |
|
| 83 |
+
| M1 | steer_heldout | 1.0 | 680 | 43% [34%,52%] | 35% [29%,42%] | 21% [16%,27%] |
|
| 84 |
+
| M1 | steer_heldout | 1.2 | 680 | 25% [17%,34%] | 42% [36%,49%] | 32% [27%,38%] |
|
| 85 |
+
| M1 | steer_heldout | 1.6 | 680 | 3% [1%,5%] | 56% [52%,59%] | 41% [39%,44%] |
|
| 86 |
+
| M1 | steer_train | 0.4 | 679 | 97% [91%,100%] | 0% [0%,0%] | 3% [0%,9%] |
|
| 87 |
+
| M1 | steer_train | 0.8 | 679 | 87% [80%,94%] | 4% [1%,8%] | 9% [4%,14%] |
|
| 88 |
+
| M1 | steer_train | 1.0 | 679 | 61% [48%,73%] | 22% [12%,32%] | 18% [11%,24%] |
|
| 89 |
+
| M1 | steer_train | 1.2 | 679 | 38% [25%,52%] | 36% [25%,48%] | 25% [19%,33%] |
|
| 90 |
+
| M1 | steer_train | 1.6 | 679 | 6% [2%,10%] | 64% [60%,67%] | 30% [28%,33%] |
|
| 91 |
|
| 92 |
## Reproduce
|
| 93 |
|