4.5 KiB
4.5 KiB
| 1 | step | calame_pt | lambada_pt | enem | bluex | oab_exams | assin2_rte | assin2_sts | faquad_nli | hatebr | hatespeech_pt | tweet_br | arc_pt | hellaswag_pt | truthfulqa |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2 | 40000 | 51.25 | 26.99 | 19.45 | 24.62 | 25.51 | 41.58 | 2.55 | 43.97 | 32.09 | 40.99 | 20.73 | 25.98 | 34.81 | 41.94 |
| 3 | 80000 | 54.0 | 34.52 | 20.57 | 19.05 | 22.96 | 33.33 | 0.71 | 43.97 | 33.96 | 24.04 | 26.32 | 27.26 | 37.14 | 38.62 |
| 4 | 120000 | 54.67 | 32.95 | 18.89 | 26.56 | 26.42 | 33.33 | 24.1 | 43.82 | 42.73 | 23.55 | 22.78 | 28.46 | 38.86 | 42.54 |
| 5 | 160000 | 54.82 | 30.39 | 20.78 | 25.87 | 26.29 | 33.33 | 0.71 | 43.97 | 33.33 | 22.99 | 26.61 | 27.86 | 39.78 | 42.11 |
| 6 | 200000 | 55.15 | 32.16 | 19.66 | 24.76 | 26.83 | 36.07 | 10.35 | 43.97 | 33.36 | 41.23 | 24.04 | 30.26 | 40.46 | 40.68 |
| 7 | 240000 | 56.41 | 32.8 | 19.45 | 24.62 | 26.65 | 33.7 | 8.49 | 43.97 | 37.1 | 44.38 | 35.64 | 29.15 | 41.01 | 39.68 |
| 8 | 280000 | 55.88 | 32.06 | 20.01 | 21.14 | 24.19 | 33.33 | 6.41 | 43.97 | 33.33 | 41.23 | 26.43 | 28.8 | 41.71 | 37.56 |
| 9 | 320000 | 56.31 | 33.22 | 20.85 | 21.0 | 24.6 | 33.33 | 10.44 | 43.92 | 36.65 | 23.36 | 20.73 | 30.0 | 41.51 | 40.04 |
| 10 | 360000 | 56.36 | 32.74 | 21.13 | 22.39 | 25.28 | 33.33 | 0.57 | 46.25 | 33.33 | 22.99 | 20.85 | 29.91 | 42.62 | 38.78 |
| 11 | 400000 | 56.36 | 32.33 | 19.59 | 24.62 | 26.88 | 33.42 | 2.96 | 43.87 | 33.33 | 23.36 | 45.33 | 29.32 | 42.15 | 40.1 |
| 12 | 440000 | 55.83 | 32.06 | 20.29 | 22.11 | 24.56 | 33.33 | 4.93 | 43.77 | 33.49 | 23.85 | 39.03 | 29.83 | 43.25 | 38.62 |
| 13 | 480000 | 56.12 | 34.29 | 20.99 | 25.17 | 24.42 | 61.45 | 16.39 | 43.97 | 33.33 | 41.23 | 21.93 | 30.6 | 42.7 | 39.39 |
| 14 | 520000 | 56.5 | 32.54 | 19.52 | 21.0 | 25.51 | 33.33 | 10.2 | 45.06 | 33.74 | 23.85 | 25.38 | 29.23 | 43.28 | 39.97 |
| 15 | 560000 | 56.84 | 34.76 | 20.78 | 22.25 | 24.05 | 39.39 | 10.62 | 43.97 | 34.3 | 29.15 | 39.95 | 30.0 | 43.35 | 38.43 |
| 16 | 600000 | 56.31 | 34.37 | 20.57 | 23.37 | 25.33 | 36.05 | 0.0 | 43.97 | 30.95 | 41.98 | 41.18 | 29.06 | 43.3 | 40.38 |
| 17 | 640000 | 57.37 | 35.13 | 20.64 | 21.0 | 24.65 | 46.15 | 11.39 | 43.97 | 38.45 | 45.03 | 30.52 | 28.8 | 43.93 | 39.86 |
| 18 | 680000 | 57.47 | 34.83 | 19.66 | 22.95 | 24.37 | 38.89 | 0.8 | 43.97 | 38.12 | 29.39 | 44.23 | 29.15 | 44.23 | 40.42 |
| 19 | 720000 | 58.29 | 34.14 | 21.97 | 21.56 | 24.33 | 65.11 | 9.34 | 43.97 | 43.04 | 53.88 | 48.77 | 30.94 | 43.93 | 38.04 |
| 20 | 760000 | 57.42 | 33.86 | 20.5 | 23.78 | 24.65 | 44.83 | 0.81 | 43.97 | 36.19 | 46.06 | 44.94 | 29.91 | 44.37 | 39.33 |
| 21 | 800000 | 57.47 | 33.05 | 19.52 | 25.59 | 25.38 | 66.81 | 0.31 | 43.97 | 30.91 | 41.61 | 44.72 | 29.4 | 44.67 | 39.43 |
| 22 | 840000 | 57.95 | 35.38 | 19.94 | 22.53 | 25.01 | 64.94 | 11.51 | 43.97 | 37.69 | 45.47 | 35.24 | 30.77 | 45.02 | 37.67 |
| 23 | 880000 | 56.36 | 36.46 | 19.87 | 22.25 | 23.01 | 51.8 | 1.04 | 43.97 | 44.43 | 52.45 | 24.5 | 32.22 | 45.23 | 39.63 |
| 24 | 920000 | 58.24 | 36.35 | 19.31 | 20.17 | 23.55 | 49.5 | 5.82 | 43.97 | 54.25 | 42.43 | 38.16 | 30.6 | 45.63 | 39.89 |
| 25 | 960000 | 57.37 | 36.87 | 20.5 | 22.81 | 23.51 | 59.18 | 4.36 | 43.97 | 31.81 | 41.23 | 43.85 | 30.77 | 45.28 | 39.48 |
| 26 | 1000000 | 57.95 | 34.35 | 20.64 | 23.37 | 25.15 | 56.51 | 1.5 | 43.97 | 49.05 | 41.67 | 47.91 | 30.17 | 45.94 | 39.35 |
| 27 | 1040000 | 58.53 | 36.91 | 20.29 | 22.81 | 25.19 | 60.72 | 0.56 | 43.97 | 44.82 | 42.29 | 35.01 | 30.17 | 45.67 | 40.66 |
| 28 | 1080000 | 58.33 | 36.44 | 20.15 | 23.64 | 25.92 | 33.97 | 0.6 | 43.97 | 46.24 | 41.85 | 46.96 | 30.51 | 46.62 | 39.58 |
| 29 | 1120000 | 57.85 | 37.45 | 20.71 | 25.03 | 26.24 | 65.35 | 4.29 | 43.97 | 43.54 | 50.93 | 38.41 | 29.32 | 46.79 | 40.06 |
| 30 | 1160000 | 58.0 | 35.82 | 20.85 | 25.45 | 25.56 | 35.74 | 0.21 | 43.97 | 35.77 | 41.4 | 47.88 | 29.83 | 46.53 | 40.43 |
| 31 | 1200000 | 58.53 | 37.51 | 20.57 | 24.48 | 25.56 | 33.41 | 2.38 | 43.97 | 27.68 | 41.5 | 52.41 | 31.2 | 46.45 | 40.27 |
| 32 | 1240000 | 59.25 | 36.1 | 21.48 | 22.81 | 26.24 | 62.87 | 2.89 | 43.97 | 42.05 | 48.4 | 46.03 | 30.17 | 46.54 | 39.49 |
| 33 | 1280000 | 58.38 | 37.34 | 20.64 | 23.37 | 26.2 | 37.31 | 0.04 | 43.97 | 27.84 | 41.07 | 52.37 | 30.77 | 46.87 | 38.76 |
| 34 | 1320000 | 58.33 | 36.39 | 20.78 | 24.76 | 25.97 | 45.17 | 4.14 | 43.97 | 26.01 | 41.14 | 51.19 | 30.85 | 46.81 | 38.41 |
| 35 | 1360000 | 58.67 | 36.21 | 20.01 | 23.92 | 25.33 | 37.87 | 2.17 | 43.97 | 36.3 | 41.52 | 52.13 | 30.68 | 46.83 | 38.76 |
| 36 | 1400000 | 59.01 | 37.09 | 20.57 | 21.84 | 25.69 | 37.56 | 0.65 | 43.97 | 32.24 | 43.42 | 56.68 | 31.28 | 46.81 | 38.93 |
| 37 | 1440000 | 58.29 | 38.02 | 21.13 | 24.2 | 24.97 | 64.55 | 1.89 | 43.97 | 29.05 | 43.4 | 58.14 | 30.68 | 46.98 | 38.75 |
| 38 | 1480000 | 58.62 | 37.16 | 21.06 | 23.64 | 25.65 | 53.01 | 1.8 | 43.97 | 29.91 | 41.94 | 55.11 | 30.68 | 46.94 | 39.38 |
| 39 | 1520000 | 58.72 | 36.74 | 21.34 | 23.23 | 25.24 | 54.99 | 0.9 | 43.97 | 26.7 | 42.34 | 50.29 | 30.26 | 46.83 | 39.2 |
| 40 | 1560000 | 58.96 | 36.79 | 21.27 | 21.7 | 25.97 | 35.61 | 0.87 | 43.97 | 28.15 | 42.72 | 54.79 | 30.85 | 46.92 | 39.68 |
| 41 | 1600000 | 58.96 | 37.03 | 21.06 | 23.92 | 26.06 | 46.66 | 1.79 | 43.97 | 29.71 | 42.73 | 57.39 | 30.77 | 47.0 | 39.49 |
| 42 | 1640000 | 58.77 | 37.53 | 21.13 | 23.5 | 25.69 | 33.33 | 0.9 | 43.97 | 28.44 | 40.98 | 58.06 | 30.17 | 47.1 | 39.12 |
| 43 | 1680000 | 59.15 | 37.3 | 20.85 | 23.37 | 25.65 | 49.68 | 1.69 | 43.97 | 27.57 | 40.98 | 54.84 | 30.26 | 46.81 | 39.32 |
| 44 | 1720000 | 58.91 | 37.09 | 20.99 | 23.09 | 26.01 | 39.83 | 0.87 | 43.97 | 27.35 | 41.31 | 57.77 | 30.43 | 47.07 | 39.34 |
| 45 | 1760000 | 58.82 | 37.14 | 20.99 | 23.09 | 26.06 | 49.37 | 0.0 | 43.97 | 28.75 | 40.94 | 56.0 | 30.43 | 47.13 | 39.54 |
| 46 | 1800000 | 59.06 | 37.24 | 20.78 | 22.53 | 25.79 | 39.94 | 1.32 | 43.97 | 27.66 | 40.94 | 56.27 | 30.26 | 47.13 | 39.15 |
| 47 | 1840000 | 58.67 | 38.19 | 20.57 | 22.67 | 25.33 | 54.47 | 1.89 | 43.97 | 30.52 | 41.89 | 57.44 | 30.43 | 47.05 | 39.8 |
| 48 | 1880000 | 58.96 | 37.63 | 21.06 | 22.67 | 25.88 | 43.65 | 0.0 | 43.97 | 32.04 | 42.29 | 56.1 | 30.51 | 47.07 | 39.15 |
| 49 | 1920000 | 58.62 | 37.26 | 20.57 | 23.23 | 25.92 | 57.38 | 1.89 | 43.97 | 31.28 | 42.33 | 57.53 | 30.26 | 47.01 | 39.11 |
| 50 | 1960000 | 59.06 | 37.67 | 20.5 | 23.23 | 25.47 | 56.27 | 1.93 | 43.97 | 29.49 | 41.98 | 58.0 | 30.43 | 47.17 | 39.3 |