koichi12 commited on
Commit
735d58a
·
verified ·
1 Parent(s): 9ae4866

Add files using upload-large-folder tool

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/README.md +49 -0
  2. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/_belebele.yaml +133 -0
  3. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_acm_Arab.yaml +4 -0
  4. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_afr_Latn.yaml +4 -0
  5. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_als_Latn.yaml +4 -0
  6. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_apc_Arab.yaml +4 -0
  7. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_arb_Latn.yaml +4 -0
  8. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ars_Arab.yaml +4 -0
  9. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_arz_Arab.yaml +4 -0
  10. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_asm_Beng.yaml +4 -0
  11. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_bam_Latn.yaml +4 -0
  12. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ben_Beng.yaml +4 -0
  13. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_bul_Cyrl.yaml +4 -0
  14. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_cat_Latn.yaml +4 -0
  15. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ceb_Latn.yaml +4 -0
  16. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ckb_Arab.yaml +4 -0
  17. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_dan_Latn.yaml +4 -0
  18. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_deu_Latn.yaml +4 -0
  19. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_eng_Latn.yaml +4 -0
  20. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_est_Latn.yaml +4 -0
  21. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_fin_Latn.yaml +4 -0
  22. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_fuv_Latn.yaml +4 -0
  23. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_gaz_Latn.yaml +4 -0
  24. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_grn_Latn.yaml +4 -0
  25. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_guj_Gujr.yaml +4 -0
  26. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_hat_Latn.yaml +4 -0
  27. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_heb_Hebr.yaml +4 -0
  28. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_hin_Deva.yaml +4 -0
  29. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_hun_Latn.yaml +4 -0
  30. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ibo_Latn.yaml +4 -0
  31. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ilo_Latn.yaml +4 -0
  32. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ind_Latn.yaml +4 -0
  33. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ita_Latn.yaml +4 -0
  34. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_jav_Latn.yaml +4 -0
  35. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kac_Latn.yaml +4 -0
  36. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kat_Geor.yaml +4 -0
  37. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kea_Latn.yaml +4 -0
  38. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_khk_Cyrl.yaml +4 -0
  39. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kin_Latn.yaml +4 -0
  40. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kir_Cyrl.yaml +4 -0
  41. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_lit_Latn.yaml +4 -0
  42. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_lug_Latn.yaml +4 -0
  43. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_luo_Latn.yaml +4 -0
  44. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_lvs_Latn.yaml +4 -0
  45. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_mal_Mlym.yaml +4 -0
  46. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_mar_Deva.yaml +4 -0
  47. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_mlt_Latn.yaml +4 -0
  48. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_mri_Latn.yaml +4 -0
  49. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_npi_Latn.yaml +4 -0
  50. scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_nso_Latn.yaml +4 -0
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/README.md ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Belebele
2
+
3
+ ### Paper
4
+
5
+ The Belebele Benchmark for Massively Multilingual NLU Evaluation
6
+ https://arxiv.org/abs/2308.16884
7
+
8
+ Belebele is a multiple-choice machine reading comprehension (MRC) dataset spanning 122 language variants. This dataset enables the evaluation of mono- and multi-lingual models in high-, medium-, and low-resource languages. Each question has four multiple-choice answers and is linked to a short passage from the FLORES-200 dataset. The human annotation procedure was carefully curated to create questions that discriminate between different levels of generalizable language comprehension and is reinforced by extensive quality checks. While all questions directly relate to the passage, the English dataset on its own proves difficult enough to challenge state-of-the-art language models. Being fully parallel, this dataset enables direct comparison of model performance across all languages. Belebele opens up new avenues for evaluating and analyzing the multilingual abilities of language models and NLP systems.
9
+
10
+ Homepage: https://github.com/facebookresearch/belebele
11
+
12
+ ### Citation
13
+
14
+ ```bibtex
15
+ @misc{bandarkar2023belebele,
16
+ title={The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants},
17
+ author={Lucas Bandarkar and Davis Liang and Benjamin Muller and Mikel Artetxe and Satya Narayan Shukla and Donald Husa and Naman Goyal and Abhinandan Krishnan and Luke Zettlemoyer and Madian Khabsa},
18
+ year={2023},
19
+ eprint={2308.16884},
20
+ archivePrefix={arXiv},
21
+ primaryClass={cs.CL}
22
+ }
23
+ ```
24
+
25
+ ### Groups and Tasks
26
+
27
+ #### Groups
28
+
29
+ - `belebele`: All 122 languages of the Belebele dataset, evaluated following the methodology in MMLU's original implementation.
30
+
31
+ #### Tasks
32
+
33
+
34
+ The following tasks evaluate languages in the Belebele dataset using loglikelihood-based multiple-choice scoring:
35
+ - `belebele_{language}`
36
+
37
+ The variant evaluated here is the 0-shot or few-shot evaluation with English Instructions.
38
+
39
+ ### Checklist
40
+
41
+ * [x] Is the task an existing benchmark in the literature?
42
+ * [x] Have you referenced the original paper that introduced the task?
43
+ * [x] If yes, does the original paper provide a reference implementation?
44
+ * [ ] Yes, original implementation contributed by author of the benchmark
45
+
46
+ If other tasks on this dataset are already supported:
47
+ * [x] Is the "Main" variant of this task clearly denoted?
48
+ * [x] Have you provided a short sentence in a README on what each new variant adds / evaluates?
49
+ * [ ] Have you noted which, if any, published evaluation setups are matched by this variant?
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/_belebele.yaml ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ group: belebele
2
+ task:
3
+ - belebele_acm_Arab
4
+ - belebele_arz_Arab
5
+ - belebele_ceb_Latn
6
+ - belebele_fin_Latn
7
+ - belebele_hin_Deva
8
+ - belebele_ita_Latn
9
+ - belebele_khm_Khmr
10
+ - belebele_lvs_Latn
11
+ - belebele_npi_Deva
12
+ - belebele_pol_Latn
13
+ - belebele_slv_Latn
14
+ - belebele_swe_Latn
15
+ - belebele_tso_Latn
16
+ - belebele_xho_Latn
17
+ - belebele_afr_Latn
18
+ - belebele_asm_Beng
19
+ - belebele_ces_Latn
20
+ - belebele_fra_Latn
21
+ - belebele_hin_Latn
22
+ - belebele_jav_Latn
23
+ - belebele_kin_Latn
24
+ - belebele_mal_Mlym
25
+ - belebele_npi_Latn
26
+ - belebele_por_Latn
27
+ - belebele_sna_Latn
28
+ - belebele_swh_Latn
29
+ - belebele_tur_Latn
30
+ - belebele_yor_Latn
31
+ - belebele_als_Latn
32
+ - belebele_azj_Latn
33
+ - belebele_ckb_Arab
34
+ - belebele_fuv_Latn
35
+ - belebele_hrv_Latn
36
+ - belebele_jpn_Jpan
37
+ - belebele_kir_Cyrl
38
+ - belebele_mar_Deva
39
+ - belebele_nso_Latn
40
+ - belebele_snd_Arab
41
+ - belebele_tam_Taml
42
+ - belebele_ukr_Cyrl
43
+ - belebele_zho_Hans
44
+ - belebele_amh_Ethi
45
+ - belebele_bam_Latn
46
+ - belebele_dan_Latn
47
+ - belebele_gaz_Latn
48
+ - belebele_hun_Latn
49
+ - belebele_kac_Latn
50
+ - belebele_kor_Hang
51
+ - belebele_mkd_Cyrl
52
+ - belebele_nya_Latn
53
+ - belebele_ron_Latn
54
+ - belebele_som_Latn
55
+ - belebele_tel_Telu
56
+ - belebele_urd_Arab
57
+ - belebele_zho_Hant
58
+ - belebele_apc_Arab
59
+ - belebele_ben_Beng
60
+ - belebele_deu_Latn
61
+ - belebele_grn_Latn
62
+ - belebele_hye_Armn
63
+ - belebele_kan_Knda
64
+ - belebele_lao_Laoo
65
+ - belebele_mlt_Latn
66
+ - belebele_ory_Orya
67
+ - belebele_rus_Cyrl
68
+ - belebele_sot_Latn
69
+ - belebele_tgk_Cyrl
70
+ - belebele_urd_Latn
71
+ - belebele_zsm_Latn
72
+ - belebele_arb_Arab
73
+ - belebele_ben_Latn
74
+ - belebele_ell_Grek
75
+ - belebele_guj_Gujr
76
+ - belebele_ibo_Latn
77
+ - belebele_kat_Geor
78
+ - belebele_lin_Latn
79
+ - belebele_mri_Latn
80
+ - belebele_pan_Guru
81
+ - belebele_shn_Mymr
82
+ - belebele_spa_Latn
83
+ - belebele_tgl_Latn
84
+ - belebele_uzn_Latn
85
+ - belebele_zul_Latn
86
+ - belebele_arb_Latn
87
+ - belebele_bod_Tibt
88
+ - belebele_eng_Latn
89
+ - belebele_hat_Latn
90
+ - belebele_ilo_Latn
91
+ - belebele_kaz_Cyrl
92
+ - belebele_lit_Latn
93
+ - belebele_mya_Mymr
94
+ - belebele_pbt_Arab
95
+ - belebele_sin_Latn
96
+ - belebele_srp_Cyrl
97
+ - belebele_tha_Thai
98
+ - belebele_vie_Latn
99
+ - belebele_ars_Arab
100
+ - belebele_bul_Cyrl
101
+ - belebele_est_Latn
102
+ - belebele_hau_Latn
103
+ - belebele_ind_Latn
104
+ - belebele_kea_Latn
105
+ - belebele_lug_Latn
106
+ - belebele_nld_Latn
107
+ - belebele_pes_Arab
108
+ - belebele_sin_Sinh
109
+ - belebele_ssw_Latn
110
+ - belebele_tir_Ethi
111
+ - belebele_war_Latn
112
+ - belebele_ary_Arab
113
+ - belebele_cat_Latn
114
+ - belebele_eus_Latn
115
+ - belebele_heb_Hebr
116
+ - belebele_isl_Latn
117
+ - belebele_khk_Cyrl
118
+ - belebele_luo_Latn
119
+ - belebele_nob_Latn
120
+ - belebele_plt_Latn
121
+ - belebele_slk_Latn
122
+ - belebele_sun_Latn
123
+ - belebele_tsn_Latn
124
+ - belebele_wol_Latn
125
+ aggregate_metric_list:
126
+ - aggregation: mean
127
+ metric: acc
128
+ weight_by_size: true
129
+ - aggregation: mean
130
+ metric: acc_norm
131
+ weight_by_size: true
132
+ metadata:
133
+ version: 0.0
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_acm_Arab.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "acm_Arab"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_acm_Arab"
4
+ "test_split": "acm_Arab"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_afr_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "afr_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_afr_Latn"
4
+ "test_split": "afr_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_als_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "als_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_als_Latn"
4
+ "test_split": "als_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_apc_Arab.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "apc_Arab"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_apc_Arab"
4
+ "test_split": "apc_Arab"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_arb_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "arb_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_arb_Latn"
4
+ "test_split": "arb_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ars_Arab.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "ars_Arab"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_ars_Arab"
4
+ "test_split": "ars_Arab"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_arz_Arab.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "arz_Arab"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_arz_Arab"
4
+ "test_split": "arz_Arab"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_asm_Beng.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "asm_Beng"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_asm_Beng"
4
+ "test_split": "asm_Beng"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_bam_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "bam_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_bam_Latn"
4
+ "test_split": "bam_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ben_Beng.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "ben_Beng"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_ben_Beng"
4
+ "test_split": "ben_Beng"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_bul_Cyrl.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "bul_Cyrl"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_bul_Cyrl"
4
+ "test_split": "bul_Cyrl"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_cat_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "cat_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_cat_Latn"
4
+ "test_split": "cat_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ceb_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "ceb_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_ceb_Latn"
4
+ "test_split": "ceb_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ckb_Arab.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "ckb_Arab"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_ckb_Arab"
4
+ "test_split": "ckb_Arab"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_dan_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "dan_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_dan_Latn"
4
+ "test_split": "dan_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_deu_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "deu_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_deu_Latn"
4
+ "test_split": "deu_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_eng_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "eng_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_eng_Latn"
4
+ "test_split": "eng_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_est_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "est_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_est_Latn"
4
+ "test_split": "est_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_fin_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "fin_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_fin_Latn"
4
+ "test_split": "fin_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_fuv_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "fuv_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_fuv_Latn"
4
+ "test_split": "fuv_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_gaz_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "gaz_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_gaz_Latn"
4
+ "test_split": "gaz_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_grn_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "grn_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_grn_Latn"
4
+ "test_split": "grn_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_guj_Gujr.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "guj_Gujr"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_guj_Gujr"
4
+ "test_split": "guj_Gujr"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_hat_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "hat_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_hat_Latn"
4
+ "test_split": "hat_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_heb_Hebr.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "heb_Hebr"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_heb_Hebr"
4
+ "test_split": "heb_Hebr"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_hin_Deva.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "hin_Deva"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_hin_Deva"
4
+ "test_split": "hin_Deva"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_hun_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "hun_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_hun_Latn"
4
+ "test_split": "hun_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ibo_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "ibo_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_ibo_Latn"
4
+ "test_split": "ibo_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ilo_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "ilo_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_ilo_Latn"
4
+ "test_split": "ilo_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ind_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "ind_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_ind_Latn"
4
+ "test_split": "ind_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_ita_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "ita_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_ita_Latn"
4
+ "test_split": "ita_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_jav_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "jav_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_jav_Latn"
4
+ "test_split": "jav_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kac_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "kac_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_kac_Latn"
4
+ "test_split": "kac_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kat_Geor.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "kat_Geor"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_kat_Geor"
4
+ "test_split": "kat_Geor"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kea_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "kea_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_kea_Latn"
4
+ "test_split": "kea_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_khk_Cyrl.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "khk_Cyrl"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_khk_Cyrl"
4
+ "test_split": "khk_Cyrl"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kin_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "kin_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_kin_Latn"
4
+ "test_split": "kin_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_kir_Cyrl.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "kir_Cyrl"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_kir_Cyrl"
4
+ "test_split": "kir_Cyrl"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_lit_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "lit_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_lit_Latn"
4
+ "test_split": "lit_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_lug_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "lug_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_lug_Latn"
4
+ "test_split": "lug_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_luo_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "luo_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_luo_Latn"
4
+ "test_split": "luo_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_lvs_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "lvs_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_lvs_Latn"
4
+ "test_split": "lvs_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_mal_Mlym.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "mal_Mlym"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_mal_Mlym"
4
+ "test_split": "mal_Mlym"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_mar_Deva.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "mar_Deva"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_mar_Deva"
4
+ "test_split": "mar_Deva"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_mlt_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "mlt_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_mlt_Latn"
4
+ "test_split": "mlt_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_mri_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "mri_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_mri_Latn"
4
+ "test_split": "mri_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_npi_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "npi_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_npi_Latn"
4
+ "test_split": "npi_Latn"
scripts/yans/lm-evaluation-harness/lm_eval/tasks/belebele/belebele_nso_Latn.yaml ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ "fewshot_split": "nso_Latn"
2
+ "include": "_default_template_yaml"
3
+ "task": "belebele_nso_Latn"
4
+ "test_split": "nso_Latn"