sol-pro / evaluation /standard_float32 /arithmark.json
j0no12's picture
Publish full float32 standard-harness evaluation
d75adc4 verified
Raw History Blame Contribute Delete
5.22 kB
{
"model": "solintellegence/sol-pro",
"tokenizer": "solintellegence/sol-pro",
"params": 138037594,
"dataset": {
"path": "/home/marimo/.cache/huggingface/hub/datasets--AxiomicLabs--ArithMark-3.0/blobs/237840149650455c4c54ff18f02203e2a76c1d75",
"sha256": "bf8ab1a5193d52cdf0e05ff0b3ca226bdfcf416cb6e75562dcbe72e7e4559435",
"items": 1000
},
"runtime": {
"device": "cuda",
"dtype": "torch.float32",
"batch_size": 32,
"max_context": 1024,
"attention_implementation": "auto",
"compiled": false,
"torch_version": "2.11.0+cu130"
},
"primary_metric": "acc_norm",
"results": {
"arithmark-3": {
"acc": 37.2,
"acc_norm": 36.0,
"raw_correct": 372,
"norm_correct": 360,
"total": 1000,
"categories": {
"elementary_school_math_continuation::addition::grades_1_2::easy": {
"acc": 23.4375,
"acc_norm": 24.21875,
"raw_correct": 30,
"norm_correct": 31,
"total": 128
},
"elementary_school_math_continuation::comparison::grades_2_3::medium": {
"acc": 20.454545454545457,
"acc_norm": 18.181818181818183,
"raw_correct": 9,
"norm_correct": 8,
"total": 44
},
"elementary_school_math_continuation::comparison_difference::grades_2_3::medium": {
"acc": 37.5,
"acc_norm": 35.41666666666667,
"raw_correct": 18,
"norm_correct": 17,
"total": 48
},
"elementary_school_math_continuation::data::grades_2_3::easy": {
"acc": 30.23255813953488,
"acc_norm": 37.2093023255814,
"raw_correct": 13,
"norm_correct": 16,
"total": 43
},
"elementary_school_math_continuation::division::grades_3_4::medium": {
"acc": 25.925925925925924,
"acc_norm": 25.925925925925924,
"raw_correct": 14,
"norm_correct": 14,
"total": 54
},
"elementary_school_math_continuation::fractions_counting::grades_3_4::medium": {
"acc": 20.0,
"acc_norm": 20.0,
"raw_correct": 10,
"norm_correct": 10,
"total": 50
},
"elementary_school_math_continuation::geometry_area::grades_4_5::medium": {
"acc": 57.692307692307686,
"acc_norm": 55.769230769230774,
"raw_correct": 30,
"norm_correct": 29,
"total": 52
},
"elementary_school_math_continuation::geometry_perimeter::grades_4_5::medium": {
"acc": 62.22222222222222,
"acc_norm": 66.66666666666666,
"raw_correct": 28,
"norm_correct": 30,
"total": 45
},
"elementary_school_math_continuation::measurement::grades_2_3::easy": {
"acc": 32.89473684210527,
"acc_norm": 31.57894736842105,
"raw_correct": 25,
"norm_correct": 24,
"total": 76
},
"elementary_school_math_continuation::money::grades_3_4::medium": {
"acc": 28.125,
"acc_norm": 29.6875,
"raw_correct": 18,
"norm_correct": 19,
"total": 64
},
"elementary_school_math_continuation::multiplication::grades_3_4::medium": {
"acc": 67.56756756756756,
"acc_norm": 59.45945945945946,
"raw_correct": 50,
"norm_correct": 44,
"total": 74
},
"elementary_school_math_continuation::patterns::grades_3_4::medium": {
"acc": 33.9622641509434,
"acc_norm": 30.18867924528302,
"raw_correct": 18,
"norm_correct": 16,
"total": 53
},
"elementary_school_math_continuation::subtraction::grades_1_2::easy": {
"acc": 28.205128205128204,
"acc_norm": 27.350427350427353,
"raw_correct": 33,
"norm_correct": 32,
"total": 117
},
"elementary_school_math_continuation::time::grades_2_3::easy": {
"acc": 100.0,
"acc_norm": 87.27272727272727,
"raw_correct": 55,
"norm_correct": 48,
"total": 55
},
"elementary_school_math_continuation::two_step_add_subtract::grades_2_3::medium": {
"acc": 17.391304347826086,
"acc_norm": 19.565217391304348,
"raw_correct": 8,
"norm_correct": 9,
"total": 46
},
"elementary_school_math_continuation::two_step_addition::grades_2_3::medium": {
"acc": 21.052631578947366,
"acc_norm": 21.052631578947366,
"raw_correct": 4,
"norm_correct": 4,
"total": 19
},
"elementary_school_math_continuation::two_step_subtraction::grades_2_3::medium": {
"acc": 28.125,
"acc_norm": 28.125,
"raw_correct": 9,
"norm_correct": 9,
"total": 32
}
},
"timing": {
"tokenization_seconds": 0.2399261619993922,
"evaluation_seconds": 49.24065730300026,
"examples_per_second": 20.308421023840992
},
"primary_metric": "acc_norm",
"primary_acc": 36.0
}
}
}