Hugging Face's logo

Datasets:
abhayesian
/
ryan-greenblatt-simulator-segment18-robustness-results-aggregate

Modalities:
Text
Formats:
parquet
Size:
< 1K
Libraries:
Datasets
pandas
Polars
Dataset card Data Studio Files
xet
 Community
abhayesian commited on
Commit
d124d55
·
verified ·
1 Parent(s): 7f8bc7d

segment 18 phase 0; project commit dde8a1cb36

Browse files
Files changed (2) hide show
  1. README.md +105 -0
  2. data/train-00000-of-00001.parquet +3 -0
README.md ADDED
@@ -0,0 +1,105 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ dataset_info:
3
+ features:
4
+ - name: git_commit
5
+ dtype: string
6
+ - name: phase
7
+ dtype: string
8
+ - name: robustness_items_completed
9
+ list: string
10
+ - name: optional_items_skipped
11
+ list: string
12
+ - name: headline
13
+ struct:
14
+ - name: cleaner_negatives_strong_flag_rate
15
+ dtype: string
16
+ - name: calibration_caveat
17
+ dtype: string
18
+ - name: wrong_author_buck_paired_delta_substance_vs_RP
19
+ struct:
20
+ - name: mean
21
+ dtype: float64
22
+ - name: ci
23
+ list: float64
24
+ - name: n
25
+ dtype: int64
26
+ - name: verdict
27
+ dtype: string
28
+ - name: wrong_author_buck_paired_delta_style_vs_RP
29
+ struct:
30
+ - name: mean
31
+ dtype: float64
32
+ - name: ci
33
+ list: float64
34
+ - name: n
35
+ dtype: int64
36
+ - name: verdict
37
+ dtype: string
38
+ - name: wrong_author_evhub_paired_delta_substance_vs_RP
39
+ struct:
40
+ - name: mean
41
+ dtype: float64
42
+ - name: ci
43
+ list: float64
44
+ - name: n
45
+ dtype: int64
46
+ - name: verdict
47
+ dtype: string
48
+ - name: wrong_author_outcome_classification
49
+ dtype: string
50
+ - name: 30b_hallucination_correlation_substance_pairwise
51
+ dtype: string
52
+ - name: 30b_hallucination_correlation_within_consistency
53
+ struct:
54
+ - name: hallu_pair_same_rate
55
+ dtype: float64
56
+ - name: nonhallu_pair_same_rate
57
+ dtype: float64
58
+ - name: delta
59
+ dtype: float64
60
+ - name: interpretation
61
+ dtype: string
62
+ - name: length_DQ_clean_rubric_RyanSFTmix_vs_RP
63
+ struct:
64
+ - name: raw_4j
65
+ dtype: float64
66
+ - name: length_resid
67
+ dtype: float64
68
+ - name: non_DQ
69
+ dtype: float64
70
+ - name: combined
71
+ dtype: float64
72
+ - name: verdict
73
+ dtype: string
74
+ - name: length_DQ_30B_residual_column_iv
75
+ dtype: float64
76
+ - name: memorization_not_load_bearing_status
77
+ dtype: string
78
+ - name: shared_system_prompt_body_caveat
79
+ dtype: string
80
+ - name: cost_summary_usd
81
+ struct:
82
+ - name: cleaner_negatives_drafting
83
+ dtype: float64
84
+ - name: paraphrastic_recall_v2_haiku_rerank
85
+ dtype: float64
86
+ - name: wrong_author_sampling_tinker
87
+ dtype: float64
88
+ - name: wrong_author_buck_judging
89
+ dtype: float64
90
+ - name: wrong_author_evhub_judging
91
+ dtype: float64
92
+ - name: total_phase_estimated
93
+ dtype: float64
94
+ splits:
95
+ - name: train
96
+ num_bytes: 1454
97
+ num_examples: 1
98
+ download_size: 25700
99
+ dataset_size: 1454
100
+ configs:
101
+ - config_name: default
102
+ data_files:
103
+ - split: train
104
+ path: data/train-*
105
+ ---
data/train-00000-of-00001.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e216bdc05a47f093180de4bf88ab02f7ff82c000ee9b079607c5869de7a9227e
3
+ size 25700