toxicity / eval /toxicity_eval.json
bogdanraduta's picture
Upload eval/toxicity_eval.json with huggingface_hub
66c5390 verified
Raw
History Blame Contribute Delete
21 kB
{
"task": "toxicity",
"threshold": 0.81,
"per_language": {
"az": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"bg": {
"support": 20,
"precision": 0.9524,
"recall": 1.0,
"f1": 0.9756,
"fpr": 0.0303,
"accuracy": 0.9811,
"n": 53
},
"cs": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"da": {
"support": 20,
"precision": 0.9524,
"recall": 1.0,
"f1": 0.9756,
"fpr": 0.0303,
"accuracy": 0.9811,
"n": 53
},
"de": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"el": {
"support": 19,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 52
},
"en": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"es": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 52
},
"et": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"fi": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9811,
"n": 53
},
"fr": {
"support": 20,
"precision": 1.0,
"recall": 0.95,
"f1": 0.9744,
"fpr": 0.0,
"accuracy": 0.9811,
"n": 53
},
"ga": {
"support": 19,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 52
},
"hr": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"hu": {
"support": 20,
"precision": 1.0,
"recall": 0.95,
"f1": 0.9744,
"fpr": 0.0,
"accuracy": 0.9811,
"n": 53
},
"it": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"lt": {
"support": 20,
"precision": 0.9524,
"recall": 1.0,
"f1": 0.9756,
"fpr": 0.0303,
"accuracy": 0.9811,
"n": 53
},
"lv": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"mt": {
"support": 20,
"precision": 0.9091,
"recall": 1.0,
"f1": 0.9524,
"fpr": 0.0606,
"accuracy": 0.8491,
"n": 53
},
"nl": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"pl": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"pt": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"ro": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"sk": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"sl": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 53
},
"sv": {
"support": 20,
"precision": 0.95,
"recall": 0.95,
"f1": 0.95,
"fpr": 0.0303,
"accuracy": 0.9434,
"n": 53
},
"tr": {
"support": 20,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9811,
"n": 53
}
},
"per_register": {
"angry_but_civil": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0385,
"accuracy": 0.9615,
"n": 104
},
"civil_disagreement": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 104
},
"group_directed": {
"support": 130,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 130
},
"harsh_criticism_of_work": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 104
},
"mild_insult": {
"support": 130,
"precision": 1.0,
"recall": 0.9846,
"f1": 0.9922,
"fpr": 0.0,
"accuracy": 0.9538,
"n": 130
},
"mundane_informational": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 78
},
"mundane_operational": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0128,
"accuracy": 0.9872,
"n": 78
},
"mundane_transactional": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 78
},
"persistent_targeting": {
"support": 128,
"precision": 1.0,
"recall": 1.0,
"f1": 1.0,
"fpr": 0.0,
"accuracy": 0.9609,
"n": 128
},
"profanity_without_target": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0097,
"accuracy": 0.9903,
"n": 103
},
"quoted_abuse_in_complaint": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 104
},
"reclaimed_ingroup": {
"support": 0,
"precision": 0.0,
"recall": 0.0,
"f1": 0.0,
"fpr": 0.0,
"accuracy": 1.0,
"n": 104
},
"veiled_threat": {
"support": 130,
"precision": 1.0,
"recall": 0.9923,
"f1": 0.9961,
"fpr": 0.0,
"accuracy": 0.9923,
"n": 130
}
},
"per_label": {
"harassment": {
"support": 128,
"precision": 0.984,
"recall": 0.9609,
"f1": 0.9723,
"fpr": 0.0016,
"accuracy": 0.0,
"n": 0
},
"identity_attack": {
"support": 130,
"precision": 0.9848,
"recall": 1.0,
"f1": 0.9924,
"fpr": 0.0016,
"accuracy": 0.0,
"n": 0
},
"insult": {
"support": 258,
"precision": 0.9844,
"recall": 0.9806,
"f1": 0.9825,
"fpr": 0.0036,
"accuracy": 0.0,
"n": 0
},
"threat": {
"support": 130,
"precision": 0.9773,
"recall": 0.9923,
"f1": 0.9847,
"fpr": 0.0024,
"accuracy": 0.0,
"n": 0
}
},
"worst_languages": [
[
"sv",
0.9500000000000001
],
[
"mt",
0.9523809523809523
],
[
"fr",
0.9743589743589743
],
[
"hu",
0.9743589743589743
],
[
"bg",
0.975609756097561
]
],
"calibration": {
"threshold": 0.81,
"objective": "macro_f1",
"value_on_val": 0.9894,
"default_threshold": 0.5,
"default_value_on_val": 0.9885,
"improvement": 0.0009,
"chosen_on": "validation split, never test",
"curve": [
{
"threshold": 0.02,
"macro_f1": 0.8994,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0128
},
{
"threshold": 0.03,
"macro_f1": 0.9465,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0105
},
{
"threshold": 0.04,
"macro_f1": 0.9643,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0093
},
{
"threshold": 0.05,
"macro_f1": 0.9705,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0093
},
{
"threshold": 0.06,
"macro_f1": 0.9746,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0093
},
{
"threshold": 0.07,
"macro_f1": 0.9754,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0093
},
{
"threshold": 0.08,
"macro_f1": 0.9764,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0093
},
{
"threshold": 0.09,
"macro_f1": 0.9773,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0082
},
{
"threshold": 0.1,
"macro_f1": 0.9773,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0082
},
{
"threshold": 0.11,
"macro_f1": 0.9773,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0082
},
{
"threshold": 0.12,
"macro_f1": 0.9782,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0082
},
{
"threshold": 0.13,
"macro_f1": 0.9772,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0082
},
{
"threshold": 0.14,
"macro_f1": 0.9772,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0082
},
{
"threshold": 0.15,
"macro_f1": 0.9781,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0082
},
{
"threshold": 0.16,
"macro_f1": 0.9781,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0082
},
{
"threshold": 0.17,
"macro_f1": 0.9791,
"any_label_recall": 0.9981,
"any_label_fpr": 0.007
},
{
"threshold": 0.18,
"macro_f1": 0.9791,
"any_label_recall": 0.9981,
"any_label_fpr": 0.007
},
{
"threshold": 0.19,
"macro_f1": 0.9791,
"any_label_recall": 0.9981,
"any_label_fpr": 0.007
},
{
"threshold": 0.2,
"macro_f1": 0.9805,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0058
},
{
"threshold": 0.21,
"macro_f1": 0.981,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0058
},
{
"threshold": 0.22,
"macro_f1": 0.981,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0058
},
{
"threshold": 0.23,
"macro_f1": 0.981,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0058
},
{
"threshold": 0.24,
"macro_f1": 0.9819,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0058
},
{
"threshold": 0.25,
"macro_f1": 0.9829,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0058
},
{
"threshold": 0.26,
"macro_f1": 0.9829,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0058
},
{
"threshold": 0.27,
"macro_f1": 0.9838,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0047
},
{
"threshold": 0.28,
"macro_f1": 0.9838,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0047
},
{
"threshold": 0.29,
"macro_f1": 0.9838,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0047
},
{
"threshold": 0.3,
"macro_f1": 0.9838,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0047
},
{
"threshold": 0.31,
"macro_f1": 0.9838,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0047
},
{
"threshold": 0.32,
"macro_f1": 0.9838,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0047
},
{
"threshold": 0.33,
"macro_f1": 0.9838,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0047
},
{
"threshold": 0.34,
"macro_f1": 0.9838,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0047
},
{
"threshold": 0.35,
"macro_f1": 0.9838,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0047
},
{
"threshold": 0.36,
"macro_f1": 0.9848,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.37,
"macro_f1": 0.9848,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.38,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.39,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.4,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.41,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.42,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.43,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.44,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.45,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.46,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.47,
"macro_f1": 0.9857,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.48,
"macro_f1": 0.9866,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.49,
"macro_f1": 0.9885,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.5,
"macro_f1": 0.9885,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.51,
"macro_f1": 0.9885,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.52,
"macro_f1": 0.9885,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.53,
"macro_f1": 0.9885,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.54,
"macro_f1": 0.9885,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.55,
"macro_f1": 0.9885,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.56,
"macro_f1": 0.9885,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.57,
"macro_f1": 0.9885,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.58,
"macro_f1": 0.989,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.59,
"macro_f1": 0.989,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.6,
"macro_f1": 0.989,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.61,
"macro_f1": 0.989,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.62,
"macro_f1": 0.989,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.63,
"macro_f1": 0.989,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.64,
"macro_f1": 0.988,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.65,
"macro_f1": 0.988,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.66,
"macro_f1": 0.988,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.67,
"macro_f1": 0.9875,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.68,
"macro_f1": 0.9875,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.69,
"macro_f1": 0.9875,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.7,
"macro_f1": 0.9875,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.71,
"macro_f1": 0.9875,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.72,
"macro_f1": 0.9875,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.73,
"macro_f1": 0.9884,
"any_label_recall": 0.9981,
"any_label_fpr": 0.0035
},
{
"threshold": 0.74,
"macro_f1": 0.9879,
"any_label_recall": 0.9961,
"any_label_fpr": 0.0035
},
{
"threshold": 0.75,
"macro_f1": 0.9879,
"any_label_recall": 0.9961,
"any_label_fpr": 0.0035
},
{
"threshold": 0.76,
"macro_f1": 0.9879,
"any_label_recall": 0.9961,
"any_label_fpr": 0.0035
},
{
"threshold": 0.77,
"macro_f1": 0.9879,
"any_label_recall": 0.9961,
"any_label_fpr": 0.0035
},
{
"threshold": 0.78,
"macro_f1": 0.9879,
"any_label_recall": 0.9961,
"any_label_fpr": 0.0035
},
{
"threshold": 0.79,
"macro_f1": 0.9884,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.8,
"macro_f1": 0.9884,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.81,
"macro_f1": 0.9894,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.82,
"macro_f1": 0.9894,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.83,
"macro_f1": 0.9894,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.84,
"macro_f1": 0.9894,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.85,
"macro_f1": 0.9894,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.86,
"macro_f1": 0.9894,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.87,
"macro_f1": 0.9894,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.88,
"macro_f1": 0.9894,
"any_label_recall": 0.9942,
"any_label_fpr": 0.0035
},
{
"threshold": 0.89,
"macro_f1": 0.9879,
"any_label_recall": 0.9923,
"any_label_fpr": 0.0035
},
{
"threshold": 0.9,
"macro_f1": 0.9879,
"any_label_recall": 0.9923,
"any_label_fpr": 0.0035
},
{
"threshold": 0.91,
"macro_f1": 0.9864,
"any_label_recall": 0.9884,
"any_label_fpr": 0.0035
},
{
"threshold": 0.92,
"macro_f1": 0.9864,
"any_label_recall": 0.9884,
"any_label_fpr": 0.0035
},
{
"threshold": 0.93,
"macro_f1": 0.9868,
"any_label_recall": 0.9846,
"any_label_fpr": 0.0035
},
{
"threshold": 0.94,
"macro_f1": 0.9863,
"any_label_recall": 0.9826,
"any_label_fpr": 0.0035
},
{
"threshold": 0.95,
"macro_f1": 0.9854,
"any_label_recall": 0.9807,
"any_label_fpr": 0.0035
}
],
"notes": []
}
}