Add 25 mGPT character-beam surprisal cells scored from <|endoftext|> within the 512-position window

#11
by samuki-hf - opened
This view is limited to 50 files because it contains too many changes. See the raw diff here.
Files changed (50) hide show
  1. README.md +7 -1
  2. brothers_kuperberg.parquet +2 -2
  3. brown_spr.parquet +2 -2
  4. bsc.parquet +2 -2
  5. characters/bsc.parquet +3 -0
  6. characters/meco_l1_w2_ch_s.parquet +2 -2
  7. characters/meco_l1_w2_ch_t.parquet +2 -2
  8. characters/provo.parquet +2 -2
  9. coverage.parquet +2 -2
  10. devarda2023.parquet +2 -2
  11. history.parquet +2 -2
  12. manifests/ai-forever_mGPT_charbeam_surprisal_K5_brothers_kuperberg.json +80 -0
  13. manifests/ai-forever_mGPT_charbeam_surprisal_K5_brown_spr.json +98 -0
  14. manifests/ai-forever_mGPT_charbeam_surprisal_K5_bsc.json +80 -0
  15. manifests/ai-forever_mGPT_charbeam_surprisal_K5_chars_bsc.json +80 -0
  16. manifests/ai-forever_mGPT_charbeam_surprisal_K5_chars_meco_l1_w2_ch_s.json +80 -0
  17. manifests/ai-forever_mGPT_charbeam_surprisal_K5_chars_meco_l1_w2_ch_t.json +80 -0
  18. manifests/ai-forever_mGPT_charbeam_surprisal_K5_chars_provo.json +80 -0
  19. manifests/ai-forever_mGPT_charbeam_surprisal_K5_devarda2023.json +80 -0
  20. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_de.json +80 -0
  21. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_en.json +80 -0
  22. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_fi.json +80 -0
  23. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_gr.json +80 -0
  24. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_he.json +80 -0
  25. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_it.json +80 -0
  26. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_ko.json +80 -0
  27. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_l1_w2_ch_s.json +80 -0
  28. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_l1_w2_ch_t.json +80 -0
  29. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_ru.json +80 -0
  30. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_sp.json +80 -0
  31. manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_tr.json +80 -0
  32. manifests/ai-forever_mGPT_charbeam_surprisal_K5_potec.json +80 -0
  33. manifests/ai-forever_mGPT_charbeam_surprisal_K5_provo.json +80 -0
  34. manifests/ai-forever_mGPT_charbeam_surprisal_K5_sbsat.json +80 -0
  35. manifests/ai-forever_mGPT_charbeam_surprisal_K5_ucl_et.json +80 -0
  36. manifests/ai-forever_mGPT_charbeam_surprisal_K5_ucl_spr.json +80 -0
  37. meco_de.parquet +2 -2
  38. meco_en.parquet +2 -2
  39. meco_fi.parquet +2 -2
  40. meco_gr.parquet +2 -2
  41. meco_he.parquet +2 -2
  42. meco_it.parquet +2 -2
  43. meco_ko.parquet +2 -2
  44. meco_l1_w2_ch_s.parquet +2 -2
  45. meco_l1_w2_ch_t.parquet +2 -2
  46. meco_ru.parquet +2 -2
  47. meco_sp.parquet +2 -2
  48. meco_tr.parquet +2 -2
  49. potec.parquet +2 -2
  50. provo.parquet +2 -2
README.md CHANGED
@@ -18,10 +18,16 @@ configs:
18
  data_files: bsc.parquet
19
  - config_name: celer
20
  data_files: celer.parquet
 
 
21
  - config_name: characters_celer
22
  data_files: characters/celer.parquet
23
  - config_name: characters_geco
24
  data_files: characters/geco.parquet
 
 
 
 
25
  - config_name: characters_provo
26
  data_files: characters/provo.parquet
27
  - config_name: characters_sbsat
@@ -224,7 +230,7 @@ Start from `tags.parquet`: one row per (`tag`, `predictor`, file), with the
224
  model, method, configuration, storage class, layer and provenance of every
225
  cell decoded, so a cell can be found by model, method and configuration.
226
 
227
- - 2,490 cells across 58 corpora, 234,509,893 scalar rows
228
  - 24 methods, 26 identified models and frequency tables
229
 
230
  ## Layout
 
18
  data_files: bsc.parquet
19
  - config_name: celer
20
  data_files: celer.parquet
21
+ - config_name: characters_bsc
22
+ data_files: characters/bsc.parquet
23
  - config_name: characters_celer
24
  data_files: characters/celer.parquet
25
  - config_name: characters_geco
26
  data_files: characters/geco.parquet
27
+ - config_name: characters_meco_l1_w2_ch_s
28
+ data_files: characters/meco_l1_w2_ch_s.parquet
29
+ - config_name: characters_meco_l1_w2_ch_t
30
+ data_files: characters/meco_l1_w2_ch_t.parquet
31
  - config_name: characters_provo
32
  data_files: characters/provo.parquet
33
  - config_name: characters_sbsat
 
230
  model, method, configuration, storage class, layer and provenance of every
231
  cell decoded, so a cell can be found by model, method and configuration.
232
 
233
+ - 2,515 cells across 58 corpora, 234,593,637 scalar rows
234
  - 24 methods, 26 identified models and frequency tables
235
 
236
  ## Layout
brothers_kuperberg.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f8a44ec6052e31ac798d527d6cfa37e3e88c28d79032ea0e623b494c7243c7ac
3
- size 4422760
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ffe406425efe3048559f6390f9c926a503d8b9382fa0ee27b10ba79e01705acc
3
+ size 4159216
brown_spr.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:45bc5e3904392f9a94f0e3f552e1338beeeef51c3253428587178abf8ecd88eb
3
- size 4693962
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b03f6c63d93835050aeb84e3c98678b8e35f1aa935547237b9be139056d7dcc
3
+ size 4251085
bsc.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ffa77045accc0197ab2d2140b3dd75f892cfc58e4c9d5904b9c2265559fe9f27
3
- size 89571
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbd1c0b91bac71d9b58299ff4dd87f051e6348ee44bc4da42dcc30bb07a607e5
3
+ size 114517
characters/bsc.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ccfe1a01e36157c1764ddafaf34a8b74077b12315b7b46866a68a6a1cc0b170
3
+ size 38668
characters/meco_l1_w2_ch_s.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6f8293d2d61c74709bcfc52584062e10c755da3dd712a3f1b2f9c94363b887d2
3
- size 47400
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6af8fcc9cf87551fdc59220117f20429ae5e842ab4cd6070587f116010c704d
3
+ size 47354
characters/meco_l1_w2_ch_t.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:47b4fc87818b87d3cc4ae4ae69b9d357d578b893d8365ec117bb084285d568d5
3
- size 47749
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3dc6a11498e3df87c50d76a2e3920fdaee7b685f6fdc4146499cba8021a3d39e
3
+ size 47742
characters/provo.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c4388046b55529327d03eca6a582ad0abb82052bdd2e9b6b6885ee01793e90d5
3
- size 158806
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1cd980a3cac987b0fd8b597e19b0a0ea9ef674c6ede111fd8074d46ca2cf585
3
+ size 316857
coverage.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2bdd6a2cfe820b764995603613559cce3f58173bba66312e2eda9444e362dce7
3
- size 2582
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e49fce89d8126a019c9d69d70119f6af9051fe3f789528630fbde3b10a5a1063
3
+ size 2824
devarda2023.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c7a7e952b6e48e5407c3eb7e911ad1761a402cc716614e662740e7ab62e9fe2e
3
- size 1186530
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8a36c2b318683c6fdbb3890e52171bc6061350cc9cb83933aac4636191a2d926
3
+ size 1148127
history.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:af0a31c1dacdedf76045a773cff5181deb8baa46ef90c730a4833007bbfca7f6
3
- size 37518
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3900a2b03a47e286169986b41e90bcd06880ffd5bb76bece49199f1a8e163c9
3
+ size 37782
manifests/ai-forever_mGPT_charbeam_surprisal_K5_brothers_kuperberg.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "en",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "2ba1782989e5c3b47c4146e3cdcd1341d2318a317c7f4dbccf9667888fef76c0",
56
+ "n_stimuli": 648,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "e12d8d1f46896fb1c64225ea2e45243ac15d211d71d1d5e00748c102c51d2c75",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:33:48+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_brown_spr.json ADDED
@@ -0,0 +1,98 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "en",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "240e84851c7ec7d2e0c6df9b0f46f1308da2446e62379f74b73122d73a61c153",
56
+ "n_stimuli": 450,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "b92f0f677e4952f3c7cddb716a9a4eabf114e9e5e16c4998c4be2feffcb58f08",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:34:19+00:00"
79
+ },
80
+ "audit": {
81
+ "primary_keys": [
82
+ "log_p_observed"
83
+ ],
84
+ "hard_issues": [],
85
+ "soft_issues": [
86
+ "mean_outlier"
87
+ ],
88
+ "examples": {
89
+ "mean_outlier": [
90
+ "trial_131: mean=-35.73 corpus_mean=-7.35 |\u0394|=28.39 (>5.0sigma=22.27)",
91
+ "trial_215: mean=-57.08 corpus_mean=-7.35 |\u0394|=49.73 (>5.0sigma=22.27)",
92
+ "trial_257: mean=-35.00 corpus_mean=-7.35 |\u0394|=27.66 (>5.0sigma=22.27)"
93
+ ]
94
+ },
95
+ "failed_stimuli": 0,
96
+ "missing_stimuli": 0
97
+ }
98
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_bsc.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "zh",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "61c6d0574e92bf9657a52738e3293246177944c3e3412dff66905e56bbd1af6c",
56
+ "n_stimuli": 150,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "e5e6685a46508e5490305f6803b3a0aefc8427530018edb4fd011fd7e8629f01",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T13:58:04+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_chars_bsc.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "zh",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "characters",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "ef5dfc13dec3eb35ded0e7fe30c8315b8ae223e924c03d78151ee2f4780a51c3",
56
+ "n_stimuli": 150,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "2ea5a1984131d5800b45d6eb36ec48cf514d998f15adf9c16369f3bcfa20a77b",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:06:14+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_chars_meco_l1_w2_ch_s.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "zh",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "characters",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "16cb35d873c0fd1a9516d75a8753aaf2a1bc20a4e53f61888928f151311d0960",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "2d9b482126f4a17a23ed3681bf2c3411ad5532373a4e9e63f58f33c154eabd5c",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:34:12+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_chars_meco_l1_w2_ch_t.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "zh",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "characters",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "5d94c0fdfeaa300075e5381c663ae350ab819851c13e072ec53558a1501ab4b0",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "442843ae38cb698e1dc1683697024275032022b901c3bb90c94a572e0ddd395c",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:52:52+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_chars_provo.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "en",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "characters",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "1bfa5187af0e0c6cc0ae7f014b5ab37e291cbfdb6026b082192345e9680692ab",
56
+ "n_stimuli": 55,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "4b325242dce8af8e4ba408c4d37977980704d6a775ec7c81f1db994ff29696d8",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:24:23+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_devarda2023.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "en",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "878f5b750b90a1027a58b6b77d10cacc9d91b0626bfc4505dc3a04c94b1f9038",
56
+ "n_stimuli": 205,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "eb58c70febd31fa423c66d7e95dd919206aeaa78634350591e5cdac3381c05a7",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T13:58:42+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_de.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "de",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "ce4b1a41117e4957f9e892838fb10de008a5ab5b678487b6892424858c5d92d9",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "f16c563b462bb190d5663ef37e6bfd463263fd2f8f08bf32f906cbe693b72625",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T15:07:27+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_en.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "en",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "e4e0484f2cd64ba7ef49e06d288a3bd501270c64ed75cfdb8e98a46cda9f39a7",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "5f1624dbbbf4ac89050e1a9218a55a5bb5f98cb12031441715e744691f7bf844",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:48:15+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_fi.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "fi",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "b5f84dde69a87b18b91479857b2c366f6c7807c3e65900cf17628016cc30a6b8",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "bcad0bac2a4fbc022ab1ce1295ae25f1d24b91e3ee2eb4020e5634b12eba5d22",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T15:17:14+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_gr.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "el",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "bb46487d3d7a77783fe5fe5c1499a0b8098eead45b44fae5e3c113196e174992",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "dcfcf03400e2b4dbde5baba2bd5f652e5f07896525e9101353868ff41f1b23cb",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T16:10:30+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_he.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "he",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "ed49348beb7e184f5d9baea8291cbdc0e8f6ece7dd575d993488213cd37826e5",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "16bde1c45a1ded7f257440d532c7c606a9b2ccb158e203e3db1637b50a018ae8",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T15:56:50+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_it.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "it",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "ba0323f83b8740883fff22fa2bda31381049edf8b2a483234ce4fb0194f15c7a",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "b43b7153510f564838d979f64568884dc30ede89ba999953315e33f6b2db4e37",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T15:09:01+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_ko.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "ko",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "75b94d84721674538046087d46de51f3faa9a774c6cd8681dd5c04fd1ef0cde5",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "b00b5f8732a39a2e7d3cf5898a120daca64f5aa226eec04d67b257c1fa4963d5",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T15:47:55+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_l1_w2_ch_s.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "zh",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "9f27685ea775f0843723c5aa8c5b226b7d3420ccb213afd1eb8198439dcdb6a3",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "88b1c8635657a36eef517d1a25b41761e82a7250aff2e56d2aa73fa50434a628",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:46:01+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_l1_w2_ch_t.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "zh",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "866884887c127db9b93a7cf7e4deb154d4c2cba960aa58ede0d05493044e3877",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "9a595bdedc811cd0a78f0c07ffdc029ba6affbe85b405a7efcf2b2f33026b835",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "Quadro RTX 6000",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:45:24+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_ru.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "ru",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "56ee0a7928fd1a4a985a1c9e9377c928263e569df23ce0f3e0961329e7e6462d",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "f435297635f5b5a2054f491a8534edcef6b23aa31d5c5d76ab9421f9b11e6852",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T15:02:40+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_sp.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "es",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "7b2d5898ff432ff69f0f58adbdeadf30a9dd116da7f1759e57e675e2e3393389",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "075d761329092c9af6dd4ddf7a37601c5c86a16fcbdc87b9afb638e1b96a3302",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:42:52+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_meco_tr.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "tr",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "08d1818b972b20a84fda05050d019a9c651df865b00443da344e1d6e59a1a0c6",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "38ca9606954f9f566c04d41fde7b2511aa5b65d851c6d46f14411116a89f75a9",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:56:53+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_potec.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "de",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "c7e39eb5101193ad025d1524c0714589c0f5ddbabdaab1ec5fc6b95796399ea6",
56
+ "n_stimuli": 12,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "02c3293dfb65059e00c78c2429ef552fb0880797294230361c09c1662b800433",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:43:32+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_provo.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "en",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "2e84d19d8d281c7337c5d39839bad1072854fd00cd94caf737dbbcc01caa398e",
56
+ "n_stimuli": 55,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "a3d71fb6a6b5ff4e30ecebd2cdb442c244941f354ec13573fcdde8a58f998669",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:19:33+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_sbsat.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "en",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "e56a90ffc7972e9a30a1e1cbd8fa867f1a3143564e2c2eefde2619427db53406",
56
+ "n_stimuli": 22,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "8c46f71072463c9debeef2794d0bc51c1908d4ba0de616fb3f6d78e0c5bfe739",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:33:51+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_ucl_et.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "en",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "4230549facfefa797285013ca093badd0cb09ac92b146bee7d26dbef9f8b9dd5",
56
+ "n_stimuli": 205,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "46c4b13dcd40458e6d34c072838096f67df904dd153277810d47dbc4de81c96e",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:03:20+00:00"
79
+ }
80
+ }
manifests/ai-forever_mGPT_charbeam_surprisal_K5_ucl_spr.json ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "identity": {
3
+ "schema_version": 1,
4
+ "method": "char_beam",
5
+ "storage_class": "positional",
6
+ "layer": 1,
7
+ "options": {
8
+ "model_id": "ai-forever/mGPT",
9
+ "beam_width": 5,
10
+ "n_samples": 64,
11
+ "max_tokens": 10,
12
+ "alpha": 1.0,
13
+ "seed": null,
14
+ "add_bos": true,
15
+ "dataset_language": "en",
16
+ "language_override": null,
17
+ "chunk_size": null,
18
+ "word_starters": [],
19
+ "top_p": 0.9,
20
+ "context_scope": "stimulus",
21
+ "unit": "words",
22
+ "window_policy": "reference",
23
+ "segmentation": "whitespace",
24
+ "model_revision": null,
25
+ "store_distribution": false,
26
+ "model_identity": null,
27
+ "start_ids": null,
28
+ "context_window": null,
29
+ "dtype": null,
30
+ "device_map": null
31
+ },
32
+ "model": {
33
+ "source": "ai-forever/mGPT",
34
+ "revision": "40897bd7c8b47a76802c411108ca6220438b8b40",
35
+ "sha256": "36faadfd737c580e4d1d02f21e6b9ae4a429237a23d34b558694c9a059494c04",
36
+ "files": [
37
+ "config.json",
38
+ "merges.txt",
39
+ "pytorch_model.bin",
40
+ "special_tokens_map.json",
41
+ "tokenizer_config.json",
42
+ "vocab.json"
43
+ ],
44
+ "context_window": 512,
45
+ "start": {
46
+ "ids": [
47
+ 5
48
+ ],
49
+ "tokens": [
50
+ "<|endoftext|>"
51
+ ],
52
+ "basis": "override"
53
+ }
54
+ },
55
+ "corpus_sha256": "182e70372dbd841647bd1de838bbc7126213e7d4e393b052250c42cdd829bf8c",
56
+ "n_stimuli": 361,
57
+ "source_sha256": "63f2da9b474b06e55223151ba109e417a29fd6310e8b584c21c9095f9427194f",
58
+ "versions": {
59
+ "python": "3.12.13",
60
+ "replica": "absent",
61
+ "numpy": "2.4.4",
62
+ "torch": "2.11.0",
63
+ "transformers": "4.57.6",
64
+ "tokenizers": "0.22.2",
65
+ "wordfreq": "3.1.1",
66
+ "wordsprobability": "0.17",
67
+ "genlm-bytes": "0.2.0"
68
+ }
69
+ },
70
+ "status": "completed",
71
+ "payload_sha256": "d4d123067775e62d132be2909311c91af58b09d07d357cf3167d9b63e3c9df6c",
72
+ "environment": {
73
+ "code_commit": "4d9a7595298965c63b235202faed69e2630432be",
74
+ "code_dirty": false,
75
+ "device": "NVIDIA GeForce RTX 3090",
76
+ "model_device": null,
77
+ "cuda": "13.0",
78
+ "computed_at": "2026-10-30T14:19:27+00:00"
79
+ }
80
+ }
meco_de.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:57952577ce5984f769578a912d93f799cf0d582ab363e55b4d1a8b07d50c8069
3
- size 769605
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7e80b3e9454f1cc9211cfa5d82ce79e1acd80ac23caa8d6888db80a53590e86d
3
+ size 653147
meco_en.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:902f5dac7f15daf25600c1d146e9f2b097304502a9f2822857905a3c6d495098
3
- size 1621965
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bacbc195fc7b80487b7cdafece3ff9f97ef1868a26079095954de6ca6cc722a8
3
+ size 1563466
meco_fi.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:dbfbed822c6ba8347daf3dd846379c07643342f7218789d3f180c1549ed7e741
3
- size 615181
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78310ebac95080fe8d3203e0622e85e0565ee1e17a415a74fe851a96288f4037
3
+ size 463156
meco_gr.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4ac0c03201d248b7cfab09a4a5cdf8da700133f5e2adc9c62e3a53e252c39e39
3
- size 792800
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c45877532ae42ceab3c7db2e34b992e359eb220b32c581533af86223d0569956
3
+ size 655189
meco_he.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cd2b0da7af81597e423c2fcbe4cbfc18380c700f55daae4113f070d73d7b16c3
3
- size 730088
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:835bfe9fdd1badc3d41241d0b084d5f2b1f57f02192ccb41ca2ccbd82aef6a90
3
+ size 593831
meco_it.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ed6a8439b0fca7d8330113c516a6fd222044173d6983451226a5607dab568a85
3
- size 790209
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba819c6a462df4a4f410d7d06e288f817ffd17248fef2dd81f2f5b925874a71c
3
+ size 684404
meco_ko.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:06d36caa2092ec56bad3f796fba3675cad5bc4c871c03a9dc35025ec7d0c70a0
3
- size 707185
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:601719210dfadf019697170fe445641a8b3febd85a88231590d179ad055842f0
3
+ size 532207
meco_l1_w2_ch_s.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:909d46ecd8474ab68c3858d6cc1a45aa8c41268cea0bf28badc0c90f9a673cf3
3
- size 26659
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0008a0f64f464446ea441b56baae72db9817bf203dbebe97426035f4d63ba164
3
+ size 56531
meco_l1_w2_ch_t.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0b0d053ef75120188d039590782f6c599163be97be199a7107ca0b0155f6f948
3
- size 26786
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b64550c49fe245245ca89b9c4a0b73f91088c9a925a21c7237b6969a0f8a780
3
+ size 56828
meco_ru.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d6c6a0db064a3177556535f1afbdace5ed21f6fd91d7828f606a935c2a6bc595
3
- size 766345
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6450facf0a1d6c3df0c588fb69cec864399c82d2b1a83b1ed9b9e7332ab55f6e
3
+ size 602589
meco_sp.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:671ee6824b1f65fa4be2212327be6ef065c4c34c7d87d0e85dfd603adfa6eab9
3
- size 784367
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6ce1d4f6eaed0ce67c889a7dd7c3bdafecf6f4ec164c0102046f1df29dd7d5a
3
+ size 704688
meco_tr.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:287a98b753015f656bc46bf8b4aa17f87e0af50d681709667f74b8cd753ffb63
3
- size 669298
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f648c932ab331a2f14551e7bf6e79b560f4080f19018f83f0e5a69465da47fa4
3
+ size 508805
potec.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f7882024df2e2d2fd03eb58e86bfdf36c938115d981b0ecedac5c93c1479bafe
3
- size 506732
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd10d9a07c68824e8772b18214d4ea8cd87b49b171704844d6125a0fd9f13d93
3
+ size 439244
provo.parquet CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4be3b826177bb02654767c133caab50f77849f3bd596e7f821094dae270110c9
3
- size 2281231
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0114006b8e716fa58c3a5244d2ee0d61bcc64122e6d7b55ab195342da4d8f3a
3
+ size 2113206