Add files using upload-large-folder tool

google--gemma-3-12b-it/layer_31--width_16k--canonical/biology/dev_llm_judge_csvs/baseline_true.scores.json

@@ -0,0 +1,18 @@
+{
+  "llm_judge/biology/in_scope/quality": 0.94,
+  "llm_judge/biology/in_scope/fluency": 1.0,
+  "llm_judge/biology/in_scope/ground_truth_similarity": 0.82,
+  "llm_judge/biology/in_scope/relevance": 1.0,
+  "llm_judge/chemistry/out_of_scope/quality": 0.9433333333333334,
+  "llm_judge/chemistry/out_of_scope/fluency": 0.995,
+  "llm_judge/chemistry/out_of_scope/ground_truth_similarity": 0.835,
+  "llm_judge/chemistry/out_of_scope/relevance": 1.0,
+  "llm_judge/math/out_of_scope/quality": 0.9283333333333333,
+  "llm_judge/math/out_of_scope/fluency": 0.945,
+  "llm_judge/math/out_of_scope/ground_truth_similarity": 0.845,
+  "llm_judge/math/out_of_scope/relevance": 0.995,
+  "llm_judge/physics/out_of_scope/quality": 0.89,
+  "llm_judge/physics/out_of_scope/fluency": 0.975,
+  "llm_judge/physics/out_of_scope/ground_truth_similarity": 0.72,
+  "llm_judge/physics/out_of_scope/relevance": 0.975
+}

google--gemma-3-12b-it/layer_31--width_16k--canonical/biology/h2e-05/k6339/6989u00w/attack/physics/vjc8361x/llm_judge_csvs/baseline_pre_attack.scores.json

@@ -0,0 +1,10 @@
+{
+  "llm_judge/biology/in_scope/quality": 0.8616666666666667,
+  "llm_judge/biology/in_scope/fluency": 0.945,
+  "llm_judge/biology/in_scope/ground_truth_similarity": 0.675,
+  "llm_judge/biology/in_scope/relevance": 0.965,
+  "llm_judge/physics/attack_scope/quality": 0.6533333333333333,
+  "llm_judge/physics/attack_scope/fluency": 0.685,
+  "llm_judge/physics/attack_scope/ground_truth_similarity": 0.375,
+  "llm_judge/physics/attack_scope/relevance": 0.9
+}

google--gemma-3-12b-it/layer_31--width_16k--canonical/biology/h2e-05/k6339/ax28xu3g/attack/physics/ir1sk2bw/README.md

@@ -0,0 +1,58 @@
+---
+base_model: roonbug/2b63aec8
+library_name: transformers
+model_name: ir1sk2bw
+tags:
+- generated_from_trainer
+- sft
+- trl
+licence: license
+---
+
+# Model Card for ir1sk2bw
+
+This model is a fine-tuned version of [roonbug/2b63aec8](https://huggingface.co/roonbug/2b63aec8).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/arunasank/sae-scoping-stemqa-biology/runs/ir1sk2bw) 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.22.2
+- Transformers: 4.56.1
+- Pytorch: 2.11.0+cu128
+- Datasets: 4.0.0
+- Tokenizers: 0.22.2
+
+## Citations
+
+
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```