sidbaines commited on
Commit
01ec63a
·
verified ·
1 Parent(s): 17cb5b5

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -78,3 +78,4 @@ fear_self_distill/evals_v2/petri_logs/2026-06-18T14-15-02-00-00_audit_kcvgXYJ4ER
78
  kindness_self_distill/checkpoints/checkpoint-3/tokenizer.json filter=lfs diff=lfs merge=lfs -text
79
  kindness_self_distill/checkpoints/checkpoint-6/tokenizer.json filter=lfs diff=lfs merge=lfs -text
80
  kindness_self_distill/checkpoints/checkpoint-9/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
78
  kindness_self_distill/checkpoints/checkpoint-3/tokenizer.json filter=lfs diff=lfs merge=lfs -text
79
  kindness_self_distill/checkpoints/checkpoint-6/tokenizer.json filter=lfs diff=lfs merge=lfs -text
80
  kindness_self_distill/checkpoints/checkpoint-9/tokenizer.json filter=lfs diff=lfs merge=lfs -text
81
+ kindness_self_distill/evals_v2/petri_logs/2026-06-18T14-46-36-00-00_audit_FqwFyYGht77esfFdjcXWVB.eval filter=lfs diff=lfs merge=lfs -text
kindness_self_distill/evals_v2/ondist_probe_summary.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "arm": "kindness_self_distill",
3
+ "rows": 24,
4
+ "summary": [
5
+ {
6
+ "arm": "kindness_self_distill",
7
+ "dimension": "sycophancy",
8
+ "mean": 9.4583,
9
+ "ci_lo": 9.1951,
10
+ "ci_hi": 9.7216,
11
+ "n": 24,
12
+ "n_unscored": 0
13
+ }
14
+ ]
15
+ }
kindness_self_distill/evals_v2/ood_probe_summary.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "arm": "kindness_self_distill",
3
+ "rows": 36,
4
+ "summary": [
5
+ {
6
+ "arm": "kindness_self_distill",
7
+ "dimension": "bad_news_softening",
8
+ "mean": 2.3333,
9
+ "ci_lo": 1.1286,
10
+ "ci_hi": 3.538,
11
+ "n": 6,
12
+ "n_unscored": 0
13
+ },
14
+ {
15
+ "arm": "kindness_self_distill",
16
+ "dimension": "capitulation_under_pushback",
17
+ "mean": 5.8333,
18
+ "ci_lo": 2.9968,
19
+ "ci_hi": 8.6699,
20
+ "n": 6,
21
+ "n_unscored": 0
22
+ },
23
+ {
24
+ "arm": "kindness_self_distill",
25
+ "dimension": "dishonesty_to_spare_feelings",
26
+ "mean": 2.3333,
27
+ "ci_lo": -0.28,
28
+ "ci_hi": 4.9467,
29
+ "n": 6,
30
+ "n_unscored": 0
31
+ },
32
+ {
33
+ "arm": "kindness_self_distill",
34
+ "dimension": "enthusiasm",
35
+ "mean": 5.5,
36
+ "ci_lo": 3.3679,
37
+ "ci_hi": 7.6321,
38
+ "n": 6,
39
+ "n_unscored": 0
40
+ },
41
+ {
42
+ "arm": "kindness_self_distill",
43
+ "dimension": "flattery",
44
+ "mean": 5.1667,
45
+ "ci_lo": 2.8247,
46
+ "ci_hi": 7.5087,
47
+ "n": 6,
48
+ "n_unscored": 0
49
+ },
50
+ {
51
+ "arm": "kindness_self_distill",
52
+ "dimension": "over_refusal",
53
+ "mean": 1.3333,
54
+ "ci_lo": 0.9201,
55
+ "ci_hi": 1.7465,
56
+ "n": 6,
57
+ "n_unscored": 0
58
+ }
59
+ ]
60
+ }
kindness_self_distill/evals_v2/petri_logs/2026-06-18T14-46-36-00-00_audit_FqwFyYGht77esfFdjcXWVB.eval ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5eaf5b3354c9bf6cfe310032e2c35f00ba637f7c873e8ff3506d32aa53c87294
3
+ size 470554