gahd / trainer_state.json
jagoldz's picture
Upload model files
edc3791 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"eval_steps": 500,
"global_step": 2010,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 9.950248756218906e-06,
"loss": 0.6611,
"step": 10
},
{
"epoch": 0.05,
"learning_rate": 9.900497512437812e-06,
"loss": 0.6622,
"step": 20
},
{
"epoch": 0.07,
"learning_rate": 9.850746268656717e-06,
"loss": 0.6558,
"step": 30
},
{
"epoch": 0.1,
"learning_rate": 9.800995024875622e-06,
"loss": 0.6535,
"step": 40
},
{
"epoch": 0.12,
"learning_rate": 9.751243781094527e-06,
"loss": 0.6359,
"step": 50
},
{
"epoch": 0.15,
"learning_rate": 9.701492537313434e-06,
"loss": 0.6274,
"step": 60
},
{
"epoch": 0.17,
"learning_rate": 9.651741293532339e-06,
"loss": 0.6193,
"step": 70
},
{
"epoch": 0.2,
"learning_rate": 9.601990049751244e-06,
"loss": 0.5715,
"step": 80
},
{
"epoch": 0.22,
"learning_rate": 9.552238805970149e-06,
"loss": 0.5749,
"step": 90
},
{
"epoch": 0.25,
"learning_rate": 9.502487562189056e-06,
"loss": 0.536,
"step": 100
},
{
"epoch": 0.27,
"learning_rate": 9.45273631840796e-06,
"loss": 0.5248,
"step": 110
},
{
"epoch": 0.3,
"learning_rate": 9.402985074626867e-06,
"loss": 0.525,
"step": 120
},
{
"epoch": 0.32,
"learning_rate": 9.353233830845772e-06,
"loss": 0.5185,
"step": 130
},
{
"epoch": 0.35,
"learning_rate": 9.303482587064677e-06,
"loss": 0.4787,
"step": 140
},
{
"epoch": 0.37,
"learning_rate": 9.253731343283582e-06,
"loss": 0.5257,
"step": 150
},
{
"epoch": 0.4,
"learning_rate": 9.203980099502489e-06,
"loss": 0.5418,
"step": 160
},
{
"epoch": 0.42,
"learning_rate": 9.154228855721394e-06,
"loss": 0.4907,
"step": 170
},
{
"epoch": 0.45,
"learning_rate": 9.104477611940299e-06,
"loss": 0.4937,
"step": 180
},
{
"epoch": 0.47,
"learning_rate": 9.054726368159204e-06,
"loss": 0.5307,
"step": 190
},
{
"epoch": 0.5,
"learning_rate": 9.00497512437811e-06,
"loss": 0.5081,
"step": 200
},
{
"epoch": 0.52,
"learning_rate": 8.955223880597016e-06,
"loss": 0.4422,
"step": 210
},
{
"epoch": 0.55,
"learning_rate": 8.905472636815922e-06,
"loss": 0.4795,
"step": 220
},
{
"epoch": 0.57,
"learning_rate": 8.855721393034826e-06,
"loss": 0.5052,
"step": 230
},
{
"epoch": 0.6,
"learning_rate": 8.805970149253732e-06,
"loss": 0.4952,
"step": 240
},
{
"epoch": 0.62,
"learning_rate": 8.756218905472637e-06,
"loss": 0.4394,
"step": 250
},
{
"epoch": 0.65,
"learning_rate": 8.706467661691544e-06,
"loss": 0.5088,
"step": 260
},
{
"epoch": 0.67,
"learning_rate": 8.656716417910447e-06,
"loss": 0.4878,
"step": 270
},
{
"epoch": 0.7,
"learning_rate": 8.606965174129354e-06,
"loss": 0.4851,
"step": 280
},
{
"epoch": 0.72,
"learning_rate": 8.557213930348259e-06,
"loss": 0.4383,
"step": 290
},
{
"epoch": 0.75,
"learning_rate": 8.507462686567165e-06,
"loss": 0.4377,
"step": 300
},
{
"epoch": 0.77,
"learning_rate": 8.45771144278607e-06,
"loss": 0.4818,
"step": 310
},
{
"epoch": 0.8,
"learning_rate": 8.407960199004975e-06,
"loss": 0.4399,
"step": 320
},
{
"epoch": 0.82,
"learning_rate": 8.35820895522388e-06,
"loss": 0.5072,
"step": 330
},
{
"epoch": 0.85,
"learning_rate": 8.308457711442787e-06,
"loss": 0.4585,
"step": 340
},
{
"epoch": 0.87,
"learning_rate": 8.258706467661692e-06,
"loss": 0.4994,
"step": 350
},
{
"epoch": 0.9,
"learning_rate": 8.208955223880599e-06,
"loss": 0.4836,
"step": 360
},
{
"epoch": 0.92,
"learning_rate": 8.159203980099502e-06,
"loss": 0.4744,
"step": 370
},
{
"epoch": 0.95,
"learning_rate": 8.109452736318409e-06,
"loss": 0.4244,
"step": 380
},
{
"epoch": 0.97,
"learning_rate": 8.059701492537314e-06,
"loss": 0.4591,
"step": 390
},
{
"epoch": 1.0,
"learning_rate": 8.00995024875622e-06,
"loss": 0.4019,
"step": 400
},
{
"epoch": 1.0,
"eval_DeTox:accuracy": 0.7757009345794392,
"eval_DeTox:f1-macro": 0.7292154841128502,
"eval_DeTox:precision": 0.6744186046511628,
"eval_DeTox:recall": 0.5686274509803921,
"eval_HASOC2019:accuracy": 0.7575757575757576,
"eval_HASOC2019:f1-macro": 0.7380952380952381,
"eval_HASOC2019:precision": 0.6153846153846154,
"eval_HASOC2019:recall": 0.7272727272727273,
"eval_HASOC2020:accuracy": 0.6511627906976745,
"eval_HASOC2020:f1-macro": 0.5274725274725275,
"eval_HASOC2020:precision": 0.3333333333333333,
"eval_HASOC2020:recall": 0.25,
"eval_MHC:accuracy": NaN,
"eval_MHC:f1-macro": NaN,
"eval_MHC:precision": 0.0,
"eval_MHC:recall": 0.0,
"eval_RP:accuracy": 0.8848684210526315,
"eval_RP:f1-macro": 0.8664105364922722,
"eval_RP:precision": 0.9397590361445783,
"eval_RP:recall": 0.7222222222222222,
"eval_Round1:accuracy": NaN,
"eval_Round1:f1-macro": NaN,
"eval_Round1:precision": 0.0,
"eval_Round1:recall": 0.0,
"eval_Round2:accuracy": NaN,
"eval_Round2:f1-macro": NaN,
"eval_Round2:precision": 0.0,
"eval_Round2:recall": 0.0,
"eval_Round3:accuracy": NaN,
"eval_Round3:f1-macro": NaN,
"eval_Round3:precision": 0.0,
"eval_Round3:recall": 0.0,
"eval_Round4:accuracy": NaN,
"eval_Round4:f1-macro": NaN,
"eval_Round4:precision": 0.0,
"eval_Round4:recall": 0.0,
"eval_accuracy": 0.7868085106382978,
"eval_f1-macro": 0.7744064408987412,
"eval_loss": 0.46475154161453247,
"eval_precision": 0.7572928821470245,
"eval_recall": 0.6889596602972399,
"eval_runtime": 9.743,
"eval_samples_per_second": 241.2,
"eval_steps_per_second": 30.176,
"step": 402
},
{
"epoch": 1.02,
"learning_rate": 7.960199004975125e-06,
"loss": 0.4151,
"step": 410
},
{
"epoch": 1.04,
"learning_rate": 7.91044776119403e-06,
"loss": 0.4307,
"step": 420
},
{
"epoch": 1.07,
"learning_rate": 7.860696517412935e-06,
"loss": 0.4405,
"step": 430
},
{
"epoch": 1.09,
"learning_rate": 7.810945273631842e-06,
"loss": 0.4275,
"step": 440
},
{
"epoch": 1.12,
"learning_rate": 7.761194029850747e-06,
"loss": 0.4203,
"step": 450
},
{
"epoch": 1.14,
"learning_rate": 7.711442786069654e-06,
"loss": 0.4934,
"step": 460
},
{
"epoch": 1.17,
"learning_rate": 7.661691542288557e-06,
"loss": 0.4166,
"step": 470
},
{
"epoch": 1.19,
"learning_rate": 7.611940298507463e-06,
"loss": 0.3872,
"step": 480
},
{
"epoch": 1.22,
"learning_rate": 7.5621890547263685e-06,
"loss": 0.362,
"step": 490
},
{
"epoch": 1.24,
"learning_rate": 7.512437810945274e-06,
"loss": 0.3795,
"step": 500
},
{
"epoch": 1.27,
"learning_rate": 7.46268656716418e-06,
"loss": 0.3736,
"step": 510
},
{
"epoch": 1.29,
"learning_rate": 7.412935323383084e-06,
"loss": 0.3603,
"step": 520
},
{
"epoch": 1.32,
"learning_rate": 7.36318407960199e-06,
"loss": 0.3526,
"step": 530
},
{
"epoch": 1.34,
"learning_rate": 7.313432835820896e-06,
"loss": 0.3054,
"step": 540
},
{
"epoch": 1.37,
"learning_rate": 7.263681592039802e-06,
"loss": 0.3601,
"step": 550
},
{
"epoch": 1.39,
"learning_rate": 7.213930348258708e-06,
"loss": 0.4352,
"step": 560
},
{
"epoch": 1.42,
"learning_rate": 7.164179104477612e-06,
"loss": 0.3177,
"step": 570
},
{
"epoch": 1.44,
"learning_rate": 7.114427860696518e-06,
"loss": 0.3465,
"step": 580
},
{
"epoch": 1.47,
"learning_rate": 7.064676616915423e-06,
"loss": 0.3679,
"step": 590
},
{
"epoch": 1.49,
"learning_rate": 7.014925373134329e-06,
"loss": 0.3933,
"step": 600
},
{
"epoch": 1.52,
"learning_rate": 6.965174129353235e-06,
"loss": 0.3282,
"step": 610
},
{
"epoch": 1.54,
"learning_rate": 6.915422885572139e-06,
"loss": 0.361,
"step": 620
},
{
"epoch": 1.57,
"learning_rate": 6.865671641791045e-06,
"loss": 0.3475,
"step": 630
},
{
"epoch": 1.59,
"learning_rate": 6.815920398009951e-06,
"loss": 0.3196,
"step": 640
},
{
"epoch": 1.62,
"learning_rate": 6.766169154228857e-06,
"loss": 0.2777,
"step": 650
},
{
"epoch": 1.64,
"learning_rate": 6.7164179104477625e-06,
"loss": 0.2905,
"step": 660
},
{
"epoch": 1.67,
"learning_rate": 6.666666666666667e-06,
"loss": 0.3959,
"step": 670
},
{
"epoch": 1.69,
"learning_rate": 6.6169154228855725e-06,
"loss": 0.3215,
"step": 680
},
{
"epoch": 1.72,
"learning_rate": 6.567164179104478e-06,
"loss": 0.3283,
"step": 690
},
{
"epoch": 1.74,
"learning_rate": 6.517412935323384e-06,
"loss": 0.2946,
"step": 700
},
{
"epoch": 1.77,
"learning_rate": 6.46766169154229e-06,
"loss": 0.3383,
"step": 710
},
{
"epoch": 1.79,
"learning_rate": 6.417910447761194e-06,
"loss": 0.279,
"step": 720
},
{
"epoch": 1.82,
"learning_rate": 6.3681592039801e-06,
"loss": 0.3557,
"step": 730
},
{
"epoch": 1.84,
"learning_rate": 6.318407960199006e-06,
"loss": 0.352,
"step": 740
},
{
"epoch": 1.87,
"learning_rate": 6.2686567164179116e-06,
"loss": 0.2888,
"step": 750
},
{
"epoch": 1.89,
"learning_rate": 6.218905472636816e-06,
"loss": 0.3714,
"step": 760
},
{
"epoch": 1.92,
"learning_rate": 6.1691542288557215e-06,
"loss": 0.2962,
"step": 770
},
{
"epoch": 1.94,
"learning_rate": 6.119402985074627e-06,
"loss": 0.3303,
"step": 780
},
{
"epoch": 1.97,
"learning_rate": 6.069651741293533e-06,
"loss": 0.311,
"step": 790
},
{
"epoch": 1.99,
"learning_rate": 6.019900497512439e-06,
"loss": 0.2763,
"step": 800
},
{
"epoch": 2.0,
"eval_DeTox:accuracy": 0.778816199376947,
"eval_DeTox:f1-macro": 0.7338028662531974,
"eval_DeTox:precision": 0.6781609195402298,
"eval_DeTox:recall": 0.5784313725490197,
"eval_HASOC2019:accuracy": 0.7878787878787878,
"eval_HASOC2019:f1-macro": 0.7664307381193125,
"eval_HASOC2019:precision": 0.6666666666666666,
"eval_HASOC2019:recall": 0.7272727272727273,
"eval_HASOC2020:accuracy": 0.6511627906976745,
"eval_HASOC2020:f1-macro": 0.5274725274725275,
"eval_HASOC2020:precision": 0.3333333333333333,
"eval_HASOC2020:recall": 0.25,
"eval_MHC:accuracy": NaN,
"eval_MHC:f1-macro": NaN,
"eval_MHC:precision": 0.0,
"eval_MHC:recall": 0.0,
"eval_RP:accuracy": 0.8618421052631579,
"eval_RP:f1-macro": 0.8342679127725857,
"eval_RP:precision": 0.9583333333333334,
"eval_RP:recall": 0.6388888888888888,
"eval_Round1:accuracy": NaN,
"eval_Round1:f1-macro": NaN,
"eval_Round1:precision": 0.0,
"eval_Round1:recall": 0.0,
"eval_Round2:accuracy": NaN,
"eval_Round2:f1-macro": NaN,
"eval_Round2:precision": 0.0,
"eval_Round2:recall": 0.0,
"eval_Round3:accuracy": NaN,
"eval_Round3:f1-macro": NaN,
"eval_Round3:precision": 0.0,
"eval_Round3:recall": 0.0,
"eval_Round4:accuracy": NaN,
"eval_Round4:f1-macro": NaN,
"eval_Round4:precision": 0.0,
"eval_Round4:recall": 0.0,
"eval_accuracy": 0.7919148936170213,
"eval_f1-macro": 0.7729607214046385,
"eval_loss": 0.5181771516799927,
"eval_precision": 0.8106995884773662,
"eval_recall": 0.6273885350318471,
"eval_runtime": 9.762,
"eval_samples_per_second": 240.729,
"eval_steps_per_second": 30.117,
"step": 804
},
{
"epoch": 2.01,
"learning_rate": 5.970149253731343e-06,
"loss": 0.2548,
"step": 810
},
{
"epoch": 2.04,
"learning_rate": 5.920398009950249e-06,
"loss": 0.3048,
"step": 820
},
{
"epoch": 2.06,
"learning_rate": 5.870646766169155e-06,
"loss": 0.2669,
"step": 830
},
{
"epoch": 2.09,
"learning_rate": 5.820895522388061e-06,
"loss": 0.2954,
"step": 840
},
{
"epoch": 2.11,
"learning_rate": 5.771144278606966e-06,
"loss": 0.2801,
"step": 850
},
{
"epoch": 2.14,
"learning_rate": 5.721393034825871e-06,
"loss": 0.4024,
"step": 860
},
{
"epoch": 2.16,
"learning_rate": 5.671641791044776e-06,
"loss": 0.2838,
"step": 870
},
{
"epoch": 2.19,
"learning_rate": 5.621890547263682e-06,
"loss": 0.2733,
"step": 880
},
{
"epoch": 2.21,
"learning_rate": 5.572139303482588e-06,
"loss": 0.2462,
"step": 890
},
{
"epoch": 2.24,
"learning_rate": 5.522388059701493e-06,
"loss": 0.225,
"step": 900
},
{
"epoch": 2.26,
"learning_rate": 5.472636815920398e-06,
"loss": 0.2447,
"step": 910
},
{
"epoch": 2.29,
"learning_rate": 5.422885572139304e-06,
"loss": 0.2392,
"step": 920
},
{
"epoch": 2.31,
"learning_rate": 5.37313432835821e-06,
"loss": 0.2926,
"step": 930
},
{
"epoch": 2.34,
"learning_rate": 5.323383084577115e-06,
"loss": 0.1636,
"step": 940
},
{
"epoch": 2.36,
"learning_rate": 5.2736318407960205e-06,
"loss": 0.2151,
"step": 950
},
{
"epoch": 2.39,
"learning_rate": 5.2238805970149255e-06,
"loss": 0.26,
"step": 960
},
{
"epoch": 2.41,
"learning_rate": 5.174129353233831e-06,
"loss": 0.2943,
"step": 970
},
{
"epoch": 2.44,
"learning_rate": 5.124378109452737e-06,
"loss": 0.2122,
"step": 980
},
{
"epoch": 2.46,
"learning_rate": 5.074626865671642e-06,
"loss": 0.2709,
"step": 990
},
{
"epoch": 2.49,
"learning_rate": 5.024875621890548e-06,
"loss": 0.2713,
"step": 1000
},
{
"epoch": 2.51,
"learning_rate": 4.975124378109453e-06,
"loss": 0.2157,
"step": 1010
},
{
"epoch": 2.54,
"learning_rate": 4.925373134328359e-06,
"loss": 0.2441,
"step": 1020
},
{
"epoch": 2.56,
"learning_rate": 4.875621890547264e-06,
"loss": 0.2299,
"step": 1030
},
{
"epoch": 2.59,
"learning_rate": 4.8258706467661695e-06,
"loss": 0.2002,
"step": 1040
},
{
"epoch": 2.61,
"learning_rate": 4.7761194029850745e-06,
"loss": 0.1875,
"step": 1050
},
{
"epoch": 2.64,
"learning_rate": 4.72636815920398e-06,
"loss": 0.2272,
"step": 1060
},
{
"epoch": 2.66,
"learning_rate": 4.676616915422886e-06,
"loss": 0.2605,
"step": 1070
},
{
"epoch": 2.69,
"learning_rate": 4.626865671641791e-06,
"loss": 0.2487,
"step": 1080
},
{
"epoch": 2.71,
"learning_rate": 4.577114427860697e-06,
"loss": 0.2037,
"step": 1090
},
{
"epoch": 2.74,
"learning_rate": 4.527363184079602e-06,
"loss": 0.2093,
"step": 1100
},
{
"epoch": 2.76,
"learning_rate": 4.477611940298508e-06,
"loss": 0.1975,
"step": 1110
},
{
"epoch": 2.79,
"learning_rate": 4.427860696517413e-06,
"loss": 0.1983,
"step": 1120
},
{
"epoch": 2.81,
"learning_rate": 4.378109452736319e-06,
"loss": 0.2215,
"step": 1130
},
{
"epoch": 2.84,
"learning_rate": 4.3283582089552236e-06,
"loss": 0.2411,
"step": 1140
},
{
"epoch": 2.86,
"learning_rate": 4.278606965174129e-06,
"loss": 0.2356,
"step": 1150
},
{
"epoch": 2.89,
"learning_rate": 4.228855721393035e-06,
"loss": 0.2352,
"step": 1160
},
{
"epoch": 2.91,
"learning_rate": 4.17910447761194e-06,
"loss": 0.1669,
"step": 1170
},
{
"epoch": 2.94,
"learning_rate": 4.129353233830846e-06,
"loss": 0.2586,
"step": 1180
},
{
"epoch": 2.96,
"learning_rate": 4.079601990049751e-06,
"loss": 0.2224,
"step": 1190
},
{
"epoch": 2.99,
"learning_rate": 4.029850746268657e-06,
"loss": 0.2033,
"step": 1200
},
{
"epoch": 3.0,
"eval_DeTox:accuracy": 0.8037383177570093,
"eval_DeTox:f1-macro": 0.774259083551934,
"eval_DeTox:precision": 0.6893203883495146,
"eval_DeTox:recall": 0.696078431372549,
"eval_HASOC2019:accuracy": 0.7575757575757576,
"eval_HASOC2019:f1-macro": 0.7518796992481203,
"eval_HASOC2019:precision": 0.5882352941176471,
"eval_HASOC2019:recall": 0.9090909090909091,
"eval_HASOC2020:accuracy": 0.7209302325581395,
"eval_HASOC2020:f1-macro": 0.6692307692307692,
"eval_HASOC2020:precision": 0.5,
"eval_HASOC2020:recall": 0.5833333333333334,
"eval_MHC:accuracy": NaN,
"eval_MHC:f1-macro": NaN,
"eval_MHC:precision": 0.0,
"eval_MHC:recall": 0.0,
"eval_RP:accuracy": 0.8782894736842105,
"eval_RP:f1-macro": 0.8595667644671952,
"eval_RP:precision": 0.9176470588235294,
"eval_RP:recall": 0.7222222222222222,
"eval_Round1:accuracy": NaN,
"eval_Round1:f1-macro": NaN,
"eval_Round1:precision": 0.0,
"eval_Round1:recall": 0.0,
"eval_Round2:accuracy": NaN,
"eval_Round2:f1-macro": NaN,
"eval_Round2:precision": 0.0,
"eval_Round2:recall": 0.0,
"eval_Round3:accuracy": NaN,
"eval_Round3:f1-macro": NaN,
"eval_Round3:precision": 0.0,
"eval_Round3:recall": 0.0,
"eval_Round4:accuracy": NaN,
"eval_Round4:f1-macro": NaN,
"eval_Round4:precision": 0.0,
"eval_Round4:recall": 0.0,
"eval_accuracy": 0.8246808510638298,
"eval_f1-macro": 0.8160155062328975,
"eval_loss": 0.4982839524745941,
"eval_precision": 0.7951002227171492,
"eval_recall": 0.7579617834394905,
"eval_runtime": 9.7615,
"eval_samples_per_second": 240.741,
"eval_steps_per_second": 30.118,
"step": 1206
},
{
"epoch": 3.01,
"learning_rate": 3.980099502487563e-06,
"loss": 0.15,
"step": 1210
},
{
"epoch": 3.03,
"learning_rate": 3.930348258706468e-06,
"loss": 0.2823,
"step": 1220
},
{
"epoch": 3.06,
"learning_rate": 3.8805970149253735e-06,
"loss": 0.2278,
"step": 1230
},
{
"epoch": 3.08,
"learning_rate": 3.8308457711442784e-06,
"loss": 0.1614,
"step": 1240
},
{
"epoch": 3.11,
"learning_rate": 3.7810945273631843e-06,
"loss": 0.1793,
"step": 1250
},
{
"epoch": 3.13,
"learning_rate": 3.73134328358209e-06,
"loss": 0.3277,
"step": 1260
},
{
"epoch": 3.16,
"learning_rate": 3.681592039800995e-06,
"loss": 0.185,
"step": 1270
},
{
"epoch": 3.18,
"learning_rate": 3.631840796019901e-06,
"loss": 0.1947,
"step": 1280
},
{
"epoch": 3.21,
"learning_rate": 3.582089552238806e-06,
"loss": 0.1548,
"step": 1290
},
{
"epoch": 3.23,
"learning_rate": 3.5323383084577117e-06,
"loss": 0.1412,
"step": 1300
},
{
"epoch": 3.26,
"learning_rate": 3.4825870646766175e-06,
"loss": 0.1608,
"step": 1310
},
{
"epoch": 3.28,
"learning_rate": 3.4328358208955225e-06,
"loss": 0.1994,
"step": 1320
},
{
"epoch": 3.31,
"learning_rate": 3.3830845771144283e-06,
"loss": 0.2287,
"step": 1330
},
{
"epoch": 3.33,
"learning_rate": 3.3333333333333333e-06,
"loss": 0.2052,
"step": 1340
},
{
"epoch": 3.36,
"learning_rate": 3.283582089552239e-06,
"loss": 0.1324,
"step": 1350
},
{
"epoch": 3.38,
"learning_rate": 3.233830845771145e-06,
"loss": 0.1727,
"step": 1360
},
{
"epoch": 3.41,
"learning_rate": 3.18407960199005e-06,
"loss": 0.2653,
"step": 1370
},
{
"epoch": 3.43,
"learning_rate": 3.1343283582089558e-06,
"loss": 0.1253,
"step": 1380
},
{
"epoch": 3.46,
"learning_rate": 3.0845771144278608e-06,
"loss": 0.1939,
"step": 1390
},
{
"epoch": 3.48,
"learning_rate": 3.0348258706467666e-06,
"loss": 0.1671,
"step": 1400
},
{
"epoch": 3.51,
"learning_rate": 2.9850746268656716e-06,
"loss": 0.1425,
"step": 1410
},
{
"epoch": 3.53,
"learning_rate": 2.9353233830845774e-06,
"loss": 0.2261,
"step": 1420
},
{
"epoch": 3.56,
"learning_rate": 2.885572139303483e-06,
"loss": 0.152,
"step": 1430
},
{
"epoch": 3.58,
"learning_rate": 2.835820895522388e-06,
"loss": 0.15,
"step": 1440
},
{
"epoch": 3.61,
"learning_rate": 2.786069651741294e-06,
"loss": 0.1616,
"step": 1450
},
{
"epoch": 3.63,
"learning_rate": 2.736318407960199e-06,
"loss": 0.147,
"step": 1460
},
{
"epoch": 3.66,
"learning_rate": 2.686567164179105e-06,
"loss": 0.2288,
"step": 1470
},
{
"epoch": 3.68,
"learning_rate": 2.6368159203980102e-06,
"loss": 0.1681,
"step": 1480
},
{
"epoch": 3.71,
"learning_rate": 2.5870646766169156e-06,
"loss": 0.1668,
"step": 1490
},
{
"epoch": 3.73,
"learning_rate": 2.537313432835821e-06,
"loss": 0.159,
"step": 1500
},
{
"epoch": 3.76,
"learning_rate": 2.4875621890547264e-06,
"loss": 0.1464,
"step": 1510
},
{
"epoch": 3.78,
"learning_rate": 2.437810945273632e-06,
"loss": 0.1449,
"step": 1520
},
{
"epoch": 3.81,
"learning_rate": 2.3880597014925373e-06,
"loss": 0.1459,
"step": 1530
},
{
"epoch": 3.83,
"learning_rate": 2.338308457711443e-06,
"loss": 0.2006,
"step": 1540
},
{
"epoch": 3.86,
"learning_rate": 2.2885572139303485e-06,
"loss": 0.1621,
"step": 1550
},
{
"epoch": 3.88,
"learning_rate": 2.238805970149254e-06,
"loss": 0.2158,
"step": 1560
},
{
"epoch": 3.91,
"learning_rate": 2.1890547263681593e-06,
"loss": 0.1351,
"step": 1570
},
{
"epoch": 3.93,
"learning_rate": 2.1393034825870647e-06,
"loss": 0.1395,
"step": 1580
},
{
"epoch": 3.96,
"learning_rate": 2.08955223880597e-06,
"loss": 0.1569,
"step": 1590
},
{
"epoch": 3.98,
"learning_rate": 2.0398009950248755e-06,
"loss": 0.1868,
"step": 1600
},
{
"epoch": 4.0,
"eval_DeTox:accuracy": 0.8130841121495327,
"eval_DeTox:f1-macro": 0.7855615925892937,
"eval_DeTox:precision": 0.7019230769230769,
"eval_DeTox:recall": 0.7156862745098039,
"eval_HASOC2019:accuracy": 0.7575757575757576,
"eval_HASOC2019:f1-macro": 0.7518796992481203,
"eval_HASOC2019:precision": 0.5882352941176471,
"eval_HASOC2019:recall": 0.9090909090909091,
"eval_HASOC2020:accuracy": 0.7441860465116279,
"eval_HASOC2020:f1-macro": 0.7030759573132455,
"eval_HASOC2020:precision": 0.5333333333333333,
"eval_HASOC2020:recall": 0.6666666666666666,
"eval_MHC:accuracy": NaN,
"eval_MHC:f1-macro": NaN,
"eval_MHC:precision": 0.0,
"eval_MHC:recall": 0.0,
"eval_RP:accuracy": 0.881578947368421,
"eval_RP:f1-macro": 0.8651884700665189,
"eval_RP:precision": 0.9,
"eval_RP:recall": 0.75,
"eval_Round1:accuracy": NaN,
"eval_Round1:f1-macro": NaN,
"eval_Round1:precision": 0.0,
"eval_Round1:recall": 0.0,
"eval_Round2:accuracy": NaN,
"eval_Round2:f1-macro": NaN,
"eval_Round2:precision": 0.0,
"eval_Round2:recall": 0.0,
"eval_Round3:accuracy": NaN,
"eval_Round3:f1-macro": NaN,
"eval_Round3:precision": 0.0,
"eval_Round3:recall": 0.0,
"eval_Round4:accuracy": NaN,
"eval_Round4:f1-macro": NaN,
"eval_Round4:precision": 0.0,
"eval_Round4:recall": 0.0,
"eval_accuracy": 0.828936170212766,
"eval_f1-macro": 0.8213584698436758,
"eval_loss": 0.5340345501899719,
"eval_precision": 0.7922077922077922,
"eval_recall": 0.7770700636942676,
"eval_runtime": 10.0181,
"eval_samples_per_second": 234.576,
"eval_steps_per_second": 29.347,
"step": 1608
},
{
"epoch": 4.0,
"learning_rate": 1.9900497512437813e-06,
"loss": 0.0959,
"step": 1610
},
{
"epoch": 4.03,
"learning_rate": 1.9402985074626867e-06,
"loss": 0.2036,
"step": 1620
},
{
"epoch": 4.05,
"learning_rate": 1.8905472636815921e-06,
"loss": 0.1506,
"step": 1630
},
{
"epoch": 4.08,
"learning_rate": 1.8407960199004975e-06,
"loss": 0.1104,
"step": 1640
},
{
"epoch": 4.1,
"learning_rate": 1.791044776119403e-06,
"loss": 0.1229,
"step": 1650
},
{
"epoch": 4.13,
"learning_rate": 1.7412935323383088e-06,
"loss": 0.1676,
"step": 1660
},
{
"epoch": 4.15,
"learning_rate": 1.6915422885572142e-06,
"loss": 0.1602,
"step": 1670
},
{
"epoch": 4.18,
"learning_rate": 1.6417910447761196e-06,
"loss": 0.1397,
"step": 1680
},
{
"epoch": 4.2,
"learning_rate": 1.592039800995025e-06,
"loss": 0.1205,
"step": 1690
},
{
"epoch": 4.23,
"learning_rate": 1.5422885572139304e-06,
"loss": 0.1497,
"step": 1700
},
{
"epoch": 4.25,
"learning_rate": 1.4925373134328358e-06,
"loss": 0.085,
"step": 1710
},
{
"epoch": 4.28,
"learning_rate": 1.4427860696517414e-06,
"loss": 0.1142,
"step": 1720
},
{
"epoch": 4.3,
"learning_rate": 1.393034825870647e-06,
"loss": 0.1601,
"step": 1730
},
{
"epoch": 4.33,
"learning_rate": 1.3432835820895524e-06,
"loss": 0.1047,
"step": 1740
},
{
"epoch": 4.35,
"learning_rate": 1.2935323383084578e-06,
"loss": 0.0812,
"step": 1750
},
{
"epoch": 4.38,
"learning_rate": 1.2437810945273632e-06,
"loss": 0.0614,
"step": 1760
},
{
"epoch": 4.4,
"learning_rate": 1.1940298507462686e-06,
"loss": 0.169,
"step": 1770
},
{
"epoch": 4.43,
"learning_rate": 1.1442786069651742e-06,
"loss": 0.0794,
"step": 1780
},
{
"epoch": 4.45,
"learning_rate": 1.0945273631840796e-06,
"loss": 0.1185,
"step": 1790
},
{
"epoch": 4.48,
"learning_rate": 1.044776119402985e-06,
"loss": 0.1428,
"step": 1800
},
{
"epoch": 4.5,
"learning_rate": 9.950248756218907e-07,
"loss": 0.1351,
"step": 1810
},
{
"epoch": 4.53,
"learning_rate": 9.452736318407961e-07,
"loss": 0.1341,
"step": 1820
},
{
"epoch": 4.55,
"learning_rate": 8.955223880597015e-07,
"loss": 0.0735,
"step": 1830
},
{
"epoch": 4.58,
"learning_rate": 8.457711442786071e-07,
"loss": 0.1164,
"step": 1840
},
{
"epoch": 4.6,
"learning_rate": 7.960199004975125e-07,
"loss": 0.0471,
"step": 1850
},
{
"epoch": 4.63,
"learning_rate": 7.462686567164179e-07,
"loss": 0.1176,
"step": 1860
},
{
"epoch": 4.65,
"learning_rate": 6.965174129353235e-07,
"loss": 0.1874,
"step": 1870
},
{
"epoch": 4.68,
"learning_rate": 6.467661691542289e-07,
"loss": 0.159,
"step": 1880
},
{
"epoch": 4.7,
"learning_rate": 5.970149253731343e-07,
"loss": 0.0975,
"step": 1890
},
{
"epoch": 4.73,
"learning_rate": 5.472636815920398e-07,
"loss": 0.0947,
"step": 1900
},
{
"epoch": 4.75,
"learning_rate": 4.975124378109453e-07,
"loss": 0.0872,
"step": 1910
},
{
"epoch": 4.78,
"learning_rate": 4.4776119402985074e-07,
"loss": 0.0993,
"step": 1920
},
{
"epoch": 4.8,
"learning_rate": 3.9800995024875624e-07,
"loss": 0.1159,
"step": 1930
},
{
"epoch": 4.83,
"learning_rate": 3.4825870646766175e-07,
"loss": 0.1865,
"step": 1940
},
{
"epoch": 4.85,
"learning_rate": 2.9850746268656716e-07,
"loss": 0.1229,
"step": 1950
},
{
"epoch": 4.88,
"learning_rate": 2.4875621890547267e-07,
"loss": 0.1395,
"step": 1960
},
{
"epoch": 4.9,
"learning_rate": 1.9900497512437812e-07,
"loss": 0.1113,
"step": 1970
},
{
"epoch": 4.93,
"learning_rate": 1.4925373134328358e-07,
"loss": 0.1167,
"step": 1980
},
{
"epoch": 4.95,
"learning_rate": 9.950248756218906e-08,
"loss": 0.093,
"step": 1990
},
{
"epoch": 4.98,
"learning_rate": 4.975124378109453e-08,
"loss": 0.1402,
"step": 2000
},
{
"epoch": 5.0,
"learning_rate": 0.0,
"loss": 0.0734,
"step": 2010
},
{
"epoch": 5.0,
"eval_DeTox:accuracy": 0.8193146417445483,
"eval_DeTox:f1-macro": 0.7985762506491259,
"eval_DeTox:precision": 0.6896551724137931,
"eval_DeTox:recall": 0.7843137254901961,
"eval_HASOC2019:accuracy": 0.7272727272727273,
"eval_HASOC2019:f1-macro": 0.723205964585275,
"eval_HASOC2019:precision": 0.5555555555555556,
"eval_HASOC2019:recall": 0.9090909090909091,
"eval_HASOC2020:accuracy": 0.6976744186046512,
"eval_HASOC2020:f1-macro": 0.661826981246219,
"eval_HASOC2020:precision": 0.47058823529411764,
"eval_HASOC2020:recall": 0.6666666666666666,
"eval_MHC:accuracy": NaN,
"eval_MHC:f1-macro": NaN,
"eval_MHC:precision": 0.0,
"eval_MHC:recall": 0.0,
"eval_RP:accuracy": 0.881578947368421,
"eval_RP:f1-macro": 0.8672102504368084,
"eval_RP:precision": 0.875,
"eval_RP:recall": 0.7777777777777778,
"eval_Round1:accuracy": NaN,
"eval_Round1:f1-macro": NaN,
"eval_Round1:precision": 0.0,
"eval_Round1:recall": 0.0,
"eval_Round2:accuracy": NaN,
"eval_Round2:f1-macro": NaN,
"eval_Round2:precision": 0.0,
"eval_Round2:recall": 0.0,
"eval_Round3:accuracy": NaN,
"eval_Round3:f1-macro": NaN,
"eval_Round3:precision": 0.0,
"eval_Round3:recall": 0.0,
"eval_Round4:accuracy": NaN,
"eval_Round4:f1-macro": NaN,
"eval_Round4:precision": 0.0,
"eval_Round4:recall": 0.0,
"eval_accuracy": 0.8336170212765958,
"eval_f1-macro": 0.8283671540104293,
"eval_loss": 0.6073299050331116,
"eval_precision": 0.7763289869608826,
"eval_recall": 0.821656050955414,
"eval_runtime": 9.7194,
"eval_samples_per_second": 241.784,
"eval_steps_per_second": 30.249,
"step": 2010
}
],
"logging_steps": 10,
"max_steps": 2010,
"num_train_epochs": 5,
"save_steps": 500,
"total_flos": 9308565479568180.0,
"trial_name": null,
"trial_params": null
}