Hugging Face's logo

amirabdullah19852020
/
interpreting_reward_models

Safetensors
English
Model card Files Community
amirabdullah19852020 commited on
Commit
cbe164a
·
verified ·
1 Parent(s): c413a6b

Upload folder using huggingface_hub

Browse files
Files changed (15) hide show
  1. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/data-00000-of-00001.arrow +3 -0
  2. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/dataset_dict.json +1 -0
  3. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/dataset_info.json +94 -0
  4. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/state.json +13 -0
  5. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/train/data-00000-of-00001.arrow +3 -0
  6. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/train/dataset_info.json +94 -0
  7. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/train/state.json +13 -0
  8. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/dataset_dict.json +1 -0
  9. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/data-00000-of-00001.arrow +3 -0
  10. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/dataset_dict.json +1 -0
  11. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/dataset_info.json +94 -0
  12. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/state.json +13 -0
  13. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/train/data-00000-of-00001.arrow +3 -0
  14. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/train/dataset_info.json +94 -0
  15. data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/train/state.json +13 -0
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/data-00000-of-00001.arrow ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b86fc2d3288383b43b6b3fa2e6fe78c7e31eb59cb320b40f6bf2cb204659525
3
+ size 73901368
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/dataset_dict.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"splits": ["train"]}
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/dataset_info.json ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "builder_name": "json",
3
+ "citation": "",
4
+ "config_name": "default",
5
+ "dataset_name": "json",
6
+ "dataset_size": 73896456,
7
+ "description": "",
8
+ "download_checksums": {
9
+ "/data/home/amir/work/codes/Interpreting-Reward-Models/Notebooks/./pythia_70m_hh_rlhf_chosen_activations_dataset_chosen.jsonl": {
10
+ "num_bytes": 136191360,
11
+ "checksum": null
12
+ }
13
+ },
14
+ "download_size": 136191360,
15
+ "features": {
16
+ "texts": {
17
+ "dtype": "string",
18
+ "_type": "Value"
19
+ },
20
+ "token_ids": {
21
+ "feature": {
22
+ "dtype": "int64",
23
+ "_type": "Value"
24
+ },
25
+ "_type": "Sequence"
26
+ },
27
+ "token_embeddings": {
28
+ "feature": {
29
+ "dtype": "float64",
30
+ "_type": "Value"
31
+ },
32
+ "_type": "Sequence"
33
+ },
34
+ "tokens": {
35
+ "feature": {
36
+ "dtype": "string",
37
+ "_type": "Value"
38
+ },
39
+ "_type": "Sequence"
40
+ },
41
+ "averaged_reprs_1": {
42
+ "feature": {
43
+ "dtype": "float64",
44
+ "_type": "Value"
45
+ },
46
+ "_type": "Sequence"
47
+ },
48
+ "averaged_reprs_2": {
49
+ "feature": {
50
+ "dtype": "float64",
51
+ "_type": "Value"
52
+ },
53
+ "_type": "Sequence"
54
+ },
55
+ "averaged_reprs_3": {
56
+ "feature": {
57
+ "dtype": "float64",
58
+ "_type": "Value"
59
+ },
60
+ "_type": "Sequence"
61
+ },
62
+ "averaged_reprs_4": {
63
+ "feature": {
64
+ "dtype": "float64",
65
+ "_type": "Value"
66
+ },
67
+ "_type": "Sequence"
68
+ },
69
+ "averaged_reprs_5": {
70
+ "feature": {
71
+ "dtype": "float64",
72
+ "_type": "Value"
73
+ },
74
+ "_type": "Sequence"
75
+ }
76
+ },
77
+ "homepage": "",
78
+ "license": "",
79
+ "size_in_bytes": 210087816,
80
+ "splits": {
81
+ "train": {
82
+ "name": "train",
83
+ "num_bytes": 73896456,
84
+ "num_examples": 2400,
85
+ "dataset_name": "json"
86
+ }
87
+ },
88
+ "version": {
89
+ "version_str": "0.0.0",
90
+ "major": 0,
91
+ "minor": 0,
92
+ "patch": 0
93
+ }
94
+ }
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/state.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_data_files": [
3
+ {
4
+ "filename": "data-00000-of-00001.arrow"
5
+ }
6
+ ],
7
+ "_fingerprint": "9e20854430ac48e1",
8
+ "_format_columns": null,
9
+ "_format_kwargs": {},
10
+ "_format_type": null,
11
+ "_output_all_columns": false,
12
+ "_split": "train"
13
+ }
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/train/data-00000-of-00001.arrow ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5e3e64fe70a018e9fabcd4ffa64a6df1c9c0acee04aa09811600a3639e265ff
3
+ size 61584264
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/train/dataset_info.json ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "builder_name": "json",
3
+ "citation": "",
4
+ "config_name": "default",
5
+ "dataset_name": "json",
6
+ "dataset_size": 61580380,
7
+ "description": "",
8
+ "download_checksums": {
9
+ "/data/home/amir/work/codes/Interpreting-Reward-Models/Notebooks/./pythia_70m_hh_rlhf_chosen_activations_dataset_chosen.jsonl": {
10
+ "num_bytes": 113492800,
11
+ "checksum": null
12
+ }
13
+ },
14
+ "download_size": 113492800,
15
+ "features": {
16
+ "texts": {
17
+ "dtype": "string",
18
+ "_type": "Value"
19
+ },
20
+ "token_ids": {
21
+ "feature": {
22
+ "dtype": "int64",
23
+ "_type": "Value"
24
+ },
25
+ "_type": "Sequence"
26
+ },
27
+ "token_embeddings": {
28
+ "feature": {
29
+ "dtype": "float64",
30
+ "_type": "Value"
31
+ },
32
+ "_type": "Sequence"
33
+ },
34
+ "tokens": {
35
+ "feature": {
36
+ "dtype": "string",
37
+ "_type": "Value"
38
+ },
39
+ "_type": "Sequence"
40
+ },
41
+ "averaged_reprs_1": {
42
+ "feature": {
43
+ "dtype": "float64",
44
+ "_type": "Value"
45
+ },
46
+ "_type": "Sequence"
47
+ },
48
+ "averaged_reprs_2": {
49
+ "feature": {
50
+ "dtype": "float64",
51
+ "_type": "Value"
52
+ },
53
+ "_type": "Sequence"
54
+ },
55
+ "averaged_reprs_3": {
56
+ "feature": {
57
+ "dtype": "float64",
58
+ "_type": "Value"
59
+ },
60
+ "_type": "Sequence"
61
+ },
62
+ "averaged_reprs_4": {
63
+ "feature": {
64
+ "dtype": "float64",
65
+ "_type": "Value"
66
+ },
67
+ "_type": "Sequence"
68
+ },
69
+ "averaged_reprs_5": {
70
+ "feature": {
71
+ "dtype": "float64",
72
+ "_type": "Value"
73
+ },
74
+ "_type": "Sequence"
75
+ }
76
+ },
77
+ "homepage": "",
78
+ "license": "",
79
+ "size_in_bytes": 175073180,
80
+ "splits": {
81
+ "train": {
82
+ "name": "train",
83
+ "num_bytes": 61580380,
84
+ "num_examples": 2000,
85
+ "dataset_name": "json"
86
+ }
87
+ },
88
+ "version": {
89
+ "version_str": "0.0.0",
90
+ "major": 0,
91
+ "minor": 0,
92
+ "patch": 0
93
+ }
94
+ }
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/chosen/train/state.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_data_files": [
3
+ {
4
+ "filename": "data-00000-of-00001.arrow"
5
+ }
6
+ ],
7
+ "_fingerprint": "ddef29fab6d6c5b9",
8
+ "_format_columns": null,
9
+ "_format_kwargs": {},
10
+ "_format_type": null,
11
+ "_output_all_columns": false,
12
+ "_split": "train"
13
+ }
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/dataset_dict.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"splits": ["chosen", "new_rejected"]}
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/data-00000-of-00001.arrow ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e433744aeb662a13089d4c96ce960a047b0ad700f338fa5b39656360560993c
3
+ size 30145536
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/dataset_dict.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"splits": ["train"]}
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/dataset_info.json ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "builder_name": "json",
3
+ "citation": "",
4
+ "config_name": "default",
5
+ "dataset_name": "json",
6
+ "dataset_size": 30141660,
7
+ "description": "",
8
+ "download_checksums": {
9
+ "/data/home/amir/work/codes/Interpreting-Reward-Models/Notebooks/./pythia_70m_hh_rlhf_new_rejected_activations_dataset_new_rejected.jsonl": {
10
+ "num_bytes": 54364962,
11
+ "checksum": null
12
+ }
13
+ },
14
+ "download_size": 54364962,
15
+ "features": {
16
+ "texts": {
17
+ "dtype": "string",
18
+ "_type": "Value"
19
+ },
20
+ "token_ids": {
21
+ "feature": {
22
+ "dtype": "int64",
23
+ "_type": "Value"
24
+ },
25
+ "_type": "Sequence"
26
+ },
27
+ "token_embeddings": {
28
+ "feature": {
29
+ "dtype": "float64",
30
+ "_type": "Value"
31
+ },
32
+ "_type": "Sequence"
33
+ },
34
+ "tokens": {
35
+ "feature": {
36
+ "dtype": "string",
37
+ "_type": "Value"
38
+ },
39
+ "_type": "Sequence"
40
+ },
41
+ "averaged_reprs_1": {
42
+ "feature": {
43
+ "dtype": "float64",
44
+ "_type": "Value"
45
+ },
46
+ "_type": "Sequence"
47
+ },
48
+ "averaged_reprs_2": {
49
+ "feature": {
50
+ "dtype": "float64",
51
+ "_type": "Value"
52
+ },
53
+ "_type": "Sequence"
54
+ },
55
+ "averaged_reprs_3": {
56
+ "feature": {
57
+ "dtype": "float64",
58
+ "_type": "Value"
59
+ },
60
+ "_type": "Sequence"
61
+ },
62
+ "averaged_reprs_4": {
63
+ "feature": {
64
+ "dtype": "float64",
65
+ "_type": "Value"
66
+ },
67
+ "_type": "Sequence"
68
+ },
69
+ "averaged_reprs_5": {
70
+ "feature": {
71
+ "dtype": "float64",
72
+ "_type": "Value"
73
+ },
74
+ "_type": "Sequence"
75
+ }
76
+ },
77
+ "homepage": "",
78
+ "license": "",
79
+ "size_in_bytes": 84506622,
80
+ "splits": {
81
+ "train": {
82
+ "name": "train",
83
+ "num_bytes": 30141660,
84
+ "num_examples": 1200,
85
+ "dataset_name": "json"
86
+ }
87
+ },
88
+ "version": {
89
+ "version_str": "0.0.0",
90
+ "major": 0,
91
+ "minor": 0,
92
+ "patch": 0
93
+ }
94
+ }
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/state.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_data_files": [
3
+ {
4
+ "filename": "data-00000-of-00001.arrow"
5
+ }
6
+ ],
7
+ "_fingerprint": "3a9ce60e9a56586b",
8
+ "_format_columns": null,
9
+ "_format_kwargs": {},
10
+ "_format_type": null,
11
+ "_output_all_columns": false,
12
+ "_split": "train"
13
+ }
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/train/data-00000-of-00001.arrow ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3307d973d917509a5537b9060a6a6f603ce2fb169269c3ed49cf16d43579cb6d
3
+ size 20097280
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/train/dataset_info.json ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "builder_name": "json",
3
+ "citation": "",
4
+ "config_name": "default",
5
+ "dataset_name": "json",
6
+ "dataset_size": 20094440,
7
+ "description": "",
8
+ "download_checksums": {
9
+ "/data/home/amir/work/codes/Interpreting-Reward-Models/Notebooks/./pythia_70m_hh_rlhf_new_rejected_activations_dataset_new_rejected.jsonl": {
10
+ "num_bytes": 36243308,
11
+ "checksum": null
12
+ }
13
+ },
14
+ "download_size": 36243308,
15
+ "features": {
16
+ "texts": {
17
+ "dtype": "string",
18
+ "_type": "Value"
19
+ },
20
+ "token_ids": {
21
+ "feature": {
22
+ "dtype": "int64",
23
+ "_type": "Value"
24
+ },
25
+ "_type": "Sequence"
26
+ },
27
+ "token_embeddings": {
28
+ "feature": {
29
+ "dtype": "float64",
30
+ "_type": "Value"
31
+ },
32
+ "_type": "Sequence"
33
+ },
34
+ "tokens": {
35
+ "feature": {
36
+ "dtype": "string",
37
+ "_type": "Value"
38
+ },
39
+ "_type": "Sequence"
40
+ },
41
+ "averaged_reprs_1": {
42
+ "feature": {
43
+ "dtype": "float64",
44
+ "_type": "Value"
45
+ },
46
+ "_type": "Sequence"
47
+ },
48
+ "averaged_reprs_2": {
49
+ "feature": {
50
+ "dtype": "float64",
51
+ "_type": "Value"
52
+ },
53
+ "_type": "Sequence"
54
+ },
55
+ "averaged_reprs_3": {
56
+ "feature": {
57
+ "dtype": "float64",
58
+ "_type": "Value"
59
+ },
60
+ "_type": "Sequence"
61
+ },
62
+ "averaged_reprs_4": {
63
+ "feature": {
64
+ "dtype": "float64",
65
+ "_type": "Value"
66
+ },
67
+ "_type": "Sequence"
68
+ },
69
+ "averaged_reprs_5": {
70
+ "feature": {
71
+ "dtype": "float64",
72
+ "_type": "Value"
73
+ },
74
+ "_type": "Sequence"
75
+ }
76
+ },
77
+ "homepage": "",
78
+ "license": "",
79
+ "size_in_bytes": 56337748,
80
+ "splits": {
81
+ "train": {
82
+ "name": "train",
83
+ "num_bytes": 20094440,
84
+ "num_examples": 800,
85
+ "dataset_name": "json"
86
+ }
87
+ },
88
+ "version": {
89
+ "version_str": "0.0.0",
90
+ "major": 0,
91
+ "minor": 0,
92
+ "patch": 0
93
+ }
94
+ }
data/merged_contrastive_EleutherAI/pythia-70m_hh_rlhf_activations_and_features.hf/new_rejected/train/state.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_data_files": [
3
+ {
4
+ "filename": "data-00000-of-00001.arrow"
5
+ }
6
+ ],
7
+ "_fingerprint": "15f0212281184de2",
8
+ "_format_columns": null,
9
+ "_format_kwargs": {},
10
+ "_format_type": null,
11
+ "_output_all_columns": false,
12
+ "_split": "train"
13
+ }