model card

README.md

@@ -20,12 +20,12 @@ the steering-resistance repo.
 |---|---|
 | run | full_3b |
 | result | clean 100%->100% · steer_heldout@1.6 correct 0%->3% |
-| trained | 2026-07-23T16:19:29+0000 |
+| trained | 2026-07-25T12:19:58+0000 |
 | code | unknown |
 | config | `configs/full_3b.yaml` (snapshot: `run/config.yaml`) |
 | wandb | — |
-| hardware | NVIDIA RTX A4000 |
-| stack | torch 2.6.0+cu124, transformers 5.14.1, peft 0.19.1, accelerate 1.14.0, datasets 5.0.0, numpy 2.4.6, huggingface_hub 1.24.0 |
+| hardware | NVIDIA GeForce RTX 3090 |
+| stack | torch 2.13.0+cu126, transformers 5.14.1, peft 0.19.1, accelerate 1.14.0, datasets 5.0.0, numpy 2.4.6, huggingface_hub 1.24.0 |
 | data: concepts_path | `adc6aee9b1cc537a…` (data/concepts_paper.json) |
 | data: qa_path | `fa2356571420fd8c…` (data/qa_open.json) |
 
@@ -62,11 +62,11 @@ the steering-resistance repo.
 | M0 | correct_inject | 1.0 | 9 | 89% [67%,100%] | 0% [0%,0%] | 11% [0%,33%] |
 | M0 | correct_inject | 1.2 | 9 | 89% [67%,100%] | 0% [0%,0%] | 11% [0%,33%] |
 | M0 | correct_inject | 1.6 | 9 | 89% [67%,100%] | 0% [0%,0%] | 11% [0%,33%] |
-| M0 | steer_heldout | 0.4 | 680 | 83% [75%,90%] | 10% [7%,14%] | 6% [2%,12%] |
-| M0 | steer_heldout | 0.8 | 680 | 15% [11%,20%] | 70% [67%,74%] | 14% [12%,16%] |
-| M0 | steer_heldout | 1.0 | 680 | 2% [1%,3%] | 79% [78%,80%] | 20% [18%,21%] |
-| M0 | steer_heldout | 1.2 | 680 | 0% [0%,0%] | 74% [73%,75%] | 26% [25%,27%] |
-| M0 | steer_heldout | 1.6 | 680 | 0% [0%,0%] | 51% [50%,52%] | 49% [48%,50%] |
+| M0 | steer_heldout | 0.4 | 2754 | 84% [77%,89%] | 9% [7%,12%] | 7% [4%,12%] |
+| M0 | steer_heldout | 0.8 | 2754 | 14% [10%,17%] | 73% [69%,77%] | 13% [12%,14%] |
+| M0 | steer_heldout | 1.0 | 2754 | 1% [1%,2%] | 81% [80%,83%] | 17% [16%,18%] |
+| M0 | steer_heldout | 1.2 | 2754 | 0% [0%,0%] | 74% [73%,75%] | 26% [25%,27%] |
+| M0 | steer_heldout | 1.6 | 2754 | 0% [0%,0%] | 52% [52%,53%] | 48% [47%,48%] |
 | M0 | steer_train | 0.4 | 679 | 93% [84%,98%] | 0% [0%,1%] | 7% [2%,16%] |
 | M0 | steer_train | 0.8 | 679 | 16% [12%,21%] | 73% [69%,78%] | 10% [8%,13%] |
 | M0 | steer_train | 1.0 | 679 | 2% [1%,3%] | 87% [85%,89%] | 11% [10%,13%] |
@@ -78,11 +78,11 @@ the steering-resistance repo.
 | M1 | correct_inject | 1.0 | 9 | 100% [100%,100%] | 0% [0%,0%] | 0% [0%,0%] |
 | M1 | correct_inject | 1.2 | 9 | 100% [100%,100%] | 0% [0%,0%] | 0% [0%,0%] |
 | M1 | correct_inject | 1.6 | 9 | 89% [67%,100%] | 0% [0%,0%] | 11% [0%,33%] |
-| M1 | steer_heldout | 0.4 | 680 | 97% [93%,99%] | 1% [1%,2%] | 2% [0%,5%] |
-| M1 | steer_heldout | 0.8 | 680 | 67% [58%,74%] | 22% [19%,25%] | 11% [7%,17%] |
-| M1 | steer_heldout | 1.0 | 680 | 43% [34%,52%] | 35% [29%,42%] | 21% [16%,27%] |
-| M1 | steer_heldout | 1.2 | 680 | 25% [17%,34%] | 42% [36%,49%] | 32% [27%,38%] |
-| M1 | steer_heldout | 1.6 | 680 | 3% [1%,5%] | 56% [52%,59%] | 41% [39%,44%] |
+| M1 | steer_heldout | 0.4 | 2754 | 97% [93%,100%] | 1% [0%,1%] | 2% [0%,7%] |
+| M1 | steer_heldout | 0.8 | 2754 | 66% [58%,72%] | 21% [17%,24%] | 14% [9%,19%] |
+| M1 | steer_heldout | 1.0 | 2754 | 44% [35%,52%] | 34% [28%,42%] | 22% [18%,27%] |
+| M1 | steer_heldout | 1.2 | 2754 | 25% [17%,33%] | 42% [35%,50%] | 33% [28%,39%] |
+| M1 | steer_heldout | 1.6 | 2754 | 3% [1%,6%] | 55% [52%,58%] | 42% [40%,44%] |
 | M1 | steer_train | 0.4 | 679 | 97% [91%,100%] | 0% [0%,0%] | 3% [0%,9%] |
 | M1 | steer_train | 0.8 | 679 | 87% [80%,94%] | 4% [1%,8%] | 9% [4%,14%] |
 | M1 | steer_train | 1.0 | 679 | 61% [48%,73%] | 22% [12%,32%] | 18% [11%,24%] |