| word-sense-disambiguation-on-words-in-context | FLAN-T5-Large 783M | #10 | Accuracy: 64.7 |
| word-sense-disambiguation-on-words-in-context | LaMini-F-T5 783M | #11 | Accuracy: 63.8 |
| word-sense-disambiguation-on-words-in-context | LaMini-GPT 1.5B | #16 | Accuracy: 52.4 |
| word-sense-disambiguation-on-words-in-context | LaMini-T5 738M | #21 | Accuracy: 50.5 |
| word-sense-disambiguation-on-words-in-context | GPT-2-XL 1.5B | #24 | Accuracy: 49.8 |
| natural-language-inference-on-rte | T5-Large 738M | #19 | Accuracy: 87.4% |
| natural-language-inference-on-rte | LaMini-GPT 1.5B | #58 | Accuracy: 67.9% |
| natural-language-inference-on-rte | LaMini-F-T5 783M | #61 | Accuracy: 65% |
| natural-language-inference-on-rte | LaMini-T5 738M | #76 | Accuracy: 57% |
| natural-language-inference-on-rte | GPT-2-XL 1.5B | #83 | Accuracy: 52.3% |
| question-answering-on-openbookqa | LaMini-GPT 1.5B | #23 | Accuracy: 39.8 |
| question-answering-on-openbookqa | LaMini-T5 738M | #24 | Accuracy: 36 |
| question-answering-on-openbookqa | LaMini-F-T5 783M | #25 | Accuracy: 34 |
| question-answering-on-openbookqa | T5-Large 738M | #26 | Accuracy: 32.8 |
| question-answering-on-openbookqa | GPT-2-XL 1.5B | #27 | Accuracy: 32 |
| question-answering-on-openbookqa | FLAN-T5-Large 783M | #28 | Accuracy: 31.2 |
| coreference-resolution-on-winograd-schema-challenge | GPT-2-XL 1.5B | #24 | Accuracy: 73.3 |
| coreference-resolution-on-winograd-schema-challenge | LaMini-GPT 1.5B | #30 | Accuracy: 69.6 |
| coreference-resolution-on-winograd-schema-challenge | T5-Large 738M | #35 | Accuracy: 66.7 |
| coreference-resolution-on-winograd-schema-challenge | LaMini-F-T5 783M | #38 | Accuracy: 64.1 |
| coreference-resolution-on-winograd-schema-challenge | LaMini-T5 738M | #50 | Accuracy: 59 |
| natural-language-inference-on-multinli | T5-Large 738M | #38 | Matched: 72.4Mismatched: 72 |
| natural-language-inference-on-multinli | LaMini-GPT 1.5B | #45 | Matched: 67.5Mismatched: 69.3 |
| natural-language-inference-on-multinli | LaMini-F-T5 783M | #46 | Matched: 61.4Mismatched: 61 |
| natural-language-inference-on-multinli | LaMini-T5 738M | #47 | Matched: 54.7Mismatched: 55.8 |
| natural-language-inference-on-multinli | GPT-2-XL 1.5B | #48 | Matched: 36.5Mismatched: 37 |
| common-sense-reasoning-on-winogrande | FLAN-T5-Large 783M | #119 | Accuracy: 59.9 |
| common-sense-reasoning-on-winogrande | GPT-2-XL 1.5B | #134 | Accuracy: 58.3 |
| common-sense-reasoning-on-winogrande | LaMini-F-T5 783M | #144 | Accuracy: 56 |
| common-sense-reasoning-on-winogrande | LaMini-GPT 1.5B | #145 | Accuracy: 56 |
| common-sense-reasoning-on-winogrande | T5-Large 738M | #151 | Accuracy: 55.2 |
| common-sense-reasoning-on-winogrande | LaMini-T5 738M | #152 | Accuracy: 54.9 |
| question-answering-on-piqa | FLAN-T5-Large 783M | #140 | Accuracy: 72.2 |
| question-answering-on-piqa | LaMini-GPT 1.5B | #149 | Accuracy: 71.3 |
| question-answering-on-piqa | LaMini-F-T5 783M | #152 | Accuracy: 70.6 |
| question-answering-on-piqa | GPT-2-XL 1.5B | #153 | Accuracy: 70.5 |
| question-answering-on-piqa | LaMini-T5 738M | #160 | Accuracy: 67.2 |
| question-answering-on-piqa | T5-Large 738M | #176 | Accuracy: 55.9 |
| sentence-completion-on-hellaswag | GPT-2-XL 1.5B | #178 | Accuracy: 50.9 |
| sentence-completion-on-hellaswag | FLAN-T5-Large 783M | #184 | Accuracy: 48.7 |
| sentence-completion-on-hellaswag | LaMini-GPT 1.5B | #186 | Accuracy: 48.3 |
| sentence-completion-on-hellaswag | LaMini-F-T5 783M | #192 | Accuracy: 43.7 |
| sentence-completion-on-hellaswag | LaMini-T5 738M | #202 | Accuracy: 40.6 |
| sentence-completion-on-hellaswag | T5-Large 738M | #204 | Accuracy: 38.9 |