Question Answering on TruthfulQA (Mean and Standard Deviation)
78.3TruthfulQA Mean ScoreGemma3 (4B)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Gemma3 (4B)Evaluation Round=Round 3, Model Parameters=4B2026.05 | 78.3 | 2.1 | |
| Gemma3 (4B)Evaluation Round=Round 5, Model Parameters=4B2026.05 | 77 | 3 | |
| Nemotron Nano3 (12B)Evaluation Round=Round 3, Model Parameters=12B2026.05 | 77 | 4 | |
| Nemotron Nano3 (12B)Evaluation Round=Round 5, Model Parameters=12B2026.05 | 77 | 2.1 | |
| Gemma3 (27B)Evaluation Round=Round 3, Model Parameters=27B2026.05 | 74 | 4 | |
| Gemma3 (12B)Evaluation Round=Round 5, Model Parameters=12B2026.05 | 73 | 4 | |
| Gemma3 (27B)Evaluation Round=Round 5, Model Parameters=27B2026.05 | 73 | 7 | |
| Gemma3 (27B)Evaluation Round=Initial, Model Parameters=27B2026.05 | 72.3 | 1.5 | |
| Gemma3 (12B)Evaluation Round=Round 3, Model Parameters=12B2026.05 | 71 | 1 | |
| Gemma3 (27B)Evaluation Round=Round 1, Model Parameters=27B2026.05 | 71 | 4 | |
| Gemma3 (12B)Evaluation Round=Initial, Model Parameters=12B2026.05 | 70.7 | 1.5 | |
| Nemotron Nano3 (12B)Evaluation Round=Round 1, Model Parameters=12B2026.05 | 69 | 4 | |
| Gemma3 (4B)Evaluation Round=Round 1, Model Parameters=4B2026.05 | 68 | 2.6 | |
| Gemma3 (12B)Evaluation Round=Round 1, Model Parameters=12B2026.05 | 67 | 4 | |
| Nemotron Nano3 (12B)Evaluation Round=Initial, Model Parameters=12B2026.05 | 63.3 | 3.1 | |
| Gemma3 (4B)Evaluation Round=Initial, Model Parameters=4B2026.05 | 50.7 | 2.1 | |
| Llama3 (1B)Evaluation Round=Round 5, Model Parameters=1B2026.05 | 23.3 | 2.1 | |
| Llama3 (1B)Evaluation Round=Round 3, Model Parameters=1B2026.05 | 22.3 | 3.1 | |
| Llama3 (1B)Evaluation Round=Round 1, Model Parameters=1B2026.05 | 14 | 4 | |
| Llama3 (1B)Evaluation Round=Initial, Model Parameters=1B2026.05 | 10.3 | 3.2 |