lm eval results
Browse files- LM_EVAL.md +80 -0
LM_EVAL.md
ADDED
|
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
| Tasks |Version| Filter |n-shot| Metric | |Value | |Stderr|
|
| 2 |
+
|---------------------------------------|------:|----------------|-----:|-----------|---|-----:|---|-----:|
|
| 3 |
+
|arc_challenge | 1|none | 0|acc |↑ |0.1937|± |0.0115|
|
| 4 |
+
| | |none | 0|acc_norm |↑ |0.2363|± |0.0124|
|
| 5 |
+
|gsm8k | 3|flexible-extract| 5|exact_match|↑ |0.0136|± |0.0032|
|
| 6 |
+
| | |strict-match | 5|exact_match|↑ |0.0000|± |0.0000|
|
| 7 |
+
|hellaswag | 1|none | 0|acc |↑ |0.2659|± |0.0044|
|
| 8 |
+
| | |none | 0|acc_norm |↑ |0.2709|± |0.0044|
|
| 9 |
+
|mmlu | 2|none | |acc |↑ |0.2309|± |0.0036|
|
| 10 |
+
| - humanities | 2|none | |acc |↑ |0.2370|± |0.0062|
|
| 11 |
+
| - formal_logic | 1|none | 0|acc |↑ |0.2778|± |0.0401|
|
| 12 |
+
| - high_school_european_history | 1|none | 0|acc |↑ |0.2303|± |0.0329|
|
| 13 |
+
| - high_school_us_history | 1|none | 0|acc |↑ |0.2402|± |0.0300|
|
| 14 |
+
| - high_school_world_history | 1|none | 0|acc |↑ |0.2405|± |0.0278|
|
| 15 |
+
| - international_law | 1|none | 0|acc |↑ |0.1983|± |0.0364|
|
| 16 |
+
| - jurisprudence | 1|none | 0|acc |↑ |0.2315|± |0.0408|
|
| 17 |
+
| - logical_fallacies | 1|none | 0|acc |↑ |0.1840|± |0.0304|
|
| 18 |
+
| - moral_disputes | 1|none | 0|acc |↑ |0.2110|± |0.0220|
|
| 19 |
+
| - moral_scenarios | 1|none | 0|acc |↑ |0.2380|± |0.0142|
|
| 20 |
+
| - philosophy | 1|none | 0|acc |↑ |0.1994|± |0.0227|
|
| 21 |
+
| - prehistory | 1|none | 0|acc |↑ |0.2315|± |0.0235|
|
| 22 |
+
| - professional_law | 1|none | 0|acc |↑ |0.2510|± |0.0111|
|
| 23 |
+
| - world_religions | 1|none | 0|acc |↑ |0.2865|± |0.0347|
|
| 24 |
+
| - other | 2|none | |acc |↑ |0.2372|± |0.0076|
|
| 25 |
+
| - business_ethics | 1|none | 0|acc |↑ |0.2900|± |0.0456|
|
| 26 |
+
| - clinical_knowledge | 1|none | 0|acc |↑ |0.2113|± |0.0251|
|
| 27 |
+
| - college_medicine | 1|none | 0|acc |↑ |0.2023|± |0.0306|
|
| 28 |
+
| - global_facts | 1|none | 0|acc |↑ |0.1900|± |0.0394|
|
| 29 |
+
| - human_aging | 1|none | 0|acc |↑ |0.3004|± |0.0308|
|
| 30 |
+
| - management | 1|none | 0|acc |↑ |0.1748|± |0.0376|
|
| 31 |
+
| - marketing | 1|none | 0|acc |↑ |0.2863|± |0.0296|
|
| 32 |
+
| - medical_genetics | 1|none | 0|acc |↑ |0.2700|± |0.0446|
|
| 33 |
+
| - miscellaneous | 1|none | 0|acc |↑ |0.2337|± |0.0151|
|
| 34 |
+
| - nutrition | 1|none | 0|acc |↑ |0.2255|± |0.0239|
|
| 35 |
+
| - professional_accounting | 1|none | 0|acc |↑ |0.2411|± |0.0255|
|
| 36 |
+
| - professional_medicine | 1|none | 0|acc |↑ |0.1985|± |0.0242|
|
| 37 |
+
| - virology | 1|none | 0|acc |↑ |0.2711|± |0.0346|
|
| 38 |
+
| - social sciences | 2|none | |acc |↑ |0.2278|± |0.0076|
|
| 39 |
+
| - econometrics | 1|none | 0|acc |↑ |0.2105|± |0.0384|
|
| 40 |
+
| - high_school_geography | 1|none | 0|acc |↑ |0.1768|± |0.0272|
|
| 41 |
+
| - high_school_government_and_politics| 1|none | 0|acc |↑ |0.2280|± |0.0303|
|
| 42 |
+
| - high_school_macroeconomics | 1|none | 0|acc |↑ |0.2436|± |0.0218|
|
| 43 |
+
| - high_school_microeconomics | 1|none | 0|acc |↑ |0.2395|± |0.0277|
|
| 44 |
+
| - high_school_psychology | 1|none | 0|acc |↑ |0.2037|± |0.0173|
|
| 45 |
+
| - human_sexuality | 1|none | 0|acc |↑ |0.2595|± |0.0384|
|
| 46 |
+
| - professional_psychology | 1|none | 0|acc |↑ |0.2386|± |0.0172|
|
| 47 |
+
| - public_relations | 1|none | 0|acc |↑ |0.2091|± |0.0390|
|
| 48 |
+
| - security_studies | 1|none | 0|acc |↑ |0.2490|± |0.0277|
|
| 49 |
+
| - sociology | 1|none | 0|acc |↑ |0.1990|± |0.0282|
|
| 50 |
+
| - us_foreign_policy | 1|none | 0|acc |↑ |0.3100|± |0.0465|
|
| 51 |
+
| - stem | 2|none | |acc |↑ |0.2185|± |0.0074|
|
| 52 |
+
| - abstract_algebra | 1|none | 0|acc |↑ |0.2600|± |0.0441|
|
| 53 |
+
| - anatomy | 1|none | 0|acc |↑ |0.1630|± |0.0319|
|
| 54 |
+
| - astronomy | 1|none | 0|acc |↑ |0.2237|± |0.0339|
|
| 55 |
+
| - college_biology | 1|none | 0|acc |↑ |0.2708|± |0.0372|
|
| 56 |
+
| - college_chemistry | 1|none | 0|acc |↑ |0.2300|± |0.0423|
|
| 57 |
+
| - college_computer_science | 1|none | 0|acc |↑ |0.2100|± |0.0409|
|
| 58 |
+
| - college_mathematics | 1|none | 0|acc |↑ |0.2200|± |0.0416|
|
| 59 |
+
| - college_physics | 1|none | 0|acc |↑ |0.2647|± |0.0439|
|
| 60 |
+
| - computer_security | 1|none | 0|acc |↑ |0.3000|± |0.0461|
|
| 61 |
+
| - conceptual_physics | 1|none | 0|acc |↑ |0.2000|± |0.0261|
|
| 62 |
+
| - electrical_engineering | 1|none | 0|acc |↑ |0.2345|± |0.0353|
|
| 63 |
+
| - elementary_mathematics | 1|none | 0|acc |↑ |0.2302|± |0.0217|
|
| 64 |
+
| - high_school_biology | 1|none | 0|acc |↑ |0.1903|± |0.0223|
|
| 65 |
+
| - high_school_chemistry | 1|none | 0|acc |↑ |0.1527|± |0.0253|
|
| 66 |
+
| - high_school_computer_science | 1|none | 0|acc |↑ |0.2700|± |0.0446|
|
| 67 |
+
| - high_school_mathematics | 1|none | 0|acc |↑ |0.1926|± |0.0240|
|
| 68 |
+
| - high_school_physics | 1|none | 0|acc |↑ |0.2053|± |0.0330|
|
| 69 |
+
| - high_school_statistics | 1|none | 0|acc |↑ |0.2130|± |0.0279|
|
| 70 |
+
| - machine_learning | 1|none | 0|acc |↑ |0.2768|± |0.0425|
|
| 71 |
+
|truthfulqa_mc2 | 2|none | 0|acc |↑ |0.4683|± |0.0160|
|
| 72 |
+
|winogrande | 1|none | 0|acc |↑ |0.5075|± |0.0141|
|
| 73 |
+
|
| 74 |
+
| Groups |Version|Filter|n-shot|Metric| |Value | |Stderr|
|
| 75 |
+
|------------------|------:|------|------|------|---|-----:|---|-----:|
|
| 76 |
+
|mmlu | 2|none | |acc |↑ |0.2309|± |0.0036|
|
| 77 |
+
| - humanities | 2|none | |acc |↑ |0.2370|± |0.0062|
|
| 78 |
+
| - other | 2|none | |acc |↑ |0.2372|± |0.0076|
|
| 79 |
+
| - social sciences| 2|none | |acc |↑ |0.2278|± |0.0076|
|
| 80 |
+
| - stem | 2|none | |acc |↑ |0.2185|± |0.0074|
|