bobox's picture
Training in progress, step 160, checkpoint
596bcdc verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.1415929203539823,
"eval_steps": 40,
"global_step": 160,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0008849557522123894,
"grad_norm": NaN,
"learning_rate": 0.0,
"loss": 5.8564,
"step": 1
},
{
"epoch": 0.0017699115044247787,
"grad_norm": NaN,
"learning_rate": 0.0,
"loss": 7.1716,
"step": 2
},
{
"epoch": 0.002654867256637168,
"grad_norm": NaN,
"learning_rate": 0.0,
"loss": 5.9095,
"step": 3
},
{
"epoch": 0.0035398230088495575,
"grad_norm": 21.95326805114746,
"learning_rate": 3.5377358490566036e-09,
"loss": 5.0841,
"step": 4
},
{
"epoch": 0.004424778761061947,
"grad_norm": 16.607179641723633,
"learning_rate": 7.075471698113207e-09,
"loss": 4.0184,
"step": 5
},
{
"epoch": 0.005309734513274336,
"grad_norm": 33.789615631103516,
"learning_rate": 1.0613207547169811e-08,
"loss": 6.2191,
"step": 6
},
{
"epoch": 0.006194690265486726,
"grad_norm": 28.073551177978516,
"learning_rate": 1.4150943396226414e-08,
"loss": 5.6124,
"step": 7
},
{
"epoch": 0.007079646017699115,
"grad_norm": 17.365602493286133,
"learning_rate": 1.768867924528302e-08,
"loss": 3.9544,
"step": 8
},
{
"epoch": 0.007964601769911504,
"grad_norm": 19.384475708007812,
"learning_rate": 2.1226415094339622e-08,
"loss": 4.7149,
"step": 9
},
{
"epoch": 0.008849557522123894,
"grad_norm": 19.67770004272461,
"learning_rate": 2.4764150943396227e-08,
"loss": 4.9616,
"step": 10
},
{
"epoch": 0.009734513274336283,
"grad_norm": 24.233421325683594,
"learning_rate": 2.830188679245283e-08,
"loss": 5.2794,
"step": 11
},
{
"epoch": 0.010619469026548672,
"grad_norm": Infinity,
"learning_rate": 2.830188679245283e-08,
"loss": 8.8704,
"step": 12
},
{
"epoch": 0.011504424778761062,
"grad_norm": 34.37785720825195,
"learning_rate": 3.183962264150943e-08,
"loss": 6.0707,
"step": 13
},
{
"epoch": 0.012389380530973451,
"grad_norm": 25.11741065979004,
"learning_rate": 3.537735849056604e-08,
"loss": 5.4071,
"step": 14
},
{
"epoch": 0.01327433628318584,
"grad_norm": 53.84364700317383,
"learning_rate": 3.891509433962264e-08,
"loss": 6.9104,
"step": 15
},
{
"epoch": 0.01415929203539823,
"grad_norm": 32.0903434753418,
"learning_rate": 4.2452830188679244e-08,
"loss": 6.0276,
"step": 16
},
{
"epoch": 0.01504424778761062,
"grad_norm": 39.742130279541016,
"learning_rate": 4.599056603773585e-08,
"loss": 6.737,
"step": 17
},
{
"epoch": 0.01592920353982301,
"grad_norm": 45.267417907714844,
"learning_rate": 4.9528301886792454e-08,
"loss": 6.5354,
"step": 18
},
{
"epoch": 0.016814159292035398,
"grad_norm": 22.39731788635254,
"learning_rate": 5.3066037735849055e-08,
"loss": 5.206,
"step": 19
},
{
"epoch": 0.017699115044247787,
"grad_norm": 20.858232498168945,
"learning_rate": 5.660377358490566e-08,
"loss": 5.2469,
"step": 20
},
{
"epoch": 0.018584070796460177,
"grad_norm": 23.96446990966797,
"learning_rate": 6.014150943396226e-08,
"loss": 5.3771,
"step": 21
},
{
"epoch": 0.019469026548672566,
"grad_norm": 22.945741653442383,
"learning_rate": 6.367924528301887e-08,
"loss": 4.979,
"step": 22
},
{
"epoch": 0.020353982300884955,
"grad_norm": 15.497300148010254,
"learning_rate": 6.721698113207547e-08,
"loss": 4.7909,
"step": 23
},
{
"epoch": 0.021238938053097345,
"grad_norm": 20.039024353027344,
"learning_rate": 7.075471698113208e-08,
"loss": 4.9086,
"step": 24
},
{
"epoch": 0.022123893805309734,
"grad_norm": 21.30576515197754,
"learning_rate": 7.429245283018869e-08,
"loss": 4.8826,
"step": 25
},
{
"epoch": 0.023008849557522124,
"grad_norm": 64.5285873413086,
"learning_rate": 7.783018867924529e-08,
"loss": 8.2266,
"step": 26
},
{
"epoch": 0.023893805309734513,
"grad_norm": 59.894893646240234,
"learning_rate": 8.13679245283019e-08,
"loss": 8.3024,
"step": 27
},
{
"epoch": 0.024778761061946902,
"grad_norm": 25.504356384277344,
"learning_rate": 8.490566037735849e-08,
"loss": 5.8745,
"step": 28
},
{
"epoch": 0.02566371681415929,
"grad_norm": 15.169568061828613,
"learning_rate": 8.84433962264151e-08,
"loss": 4.7298,
"step": 29
},
{
"epoch": 0.02654867256637168,
"grad_norm": 24.09995460510254,
"learning_rate": 9.19811320754717e-08,
"loss": 5.4614,
"step": 30
},
{
"epoch": 0.02743362831858407,
"grad_norm": 28.669275283813477,
"learning_rate": 9.55188679245283e-08,
"loss": 5.8594,
"step": 31
},
{
"epoch": 0.02831858407079646,
"grad_norm": 23.37987518310547,
"learning_rate": 9.905660377358491e-08,
"loss": 5.2401,
"step": 32
},
{
"epoch": 0.02920353982300885,
"grad_norm": 22.815292358398438,
"learning_rate": 1.0259433962264152e-07,
"loss": 5.1579,
"step": 33
},
{
"epoch": 0.03008849557522124,
"grad_norm": 13.775344848632812,
"learning_rate": 1.0613207547169811e-07,
"loss": 5.2181,
"step": 34
},
{
"epoch": 0.030973451327433628,
"grad_norm": 18.642087936401367,
"learning_rate": 1.0966981132075472e-07,
"loss": 4.6328,
"step": 35
},
{
"epoch": 0.03185840707964602,
"grad_norm": 18.041406631469727,
"learning_rate": 1.1320754716981131e-07,
"loss": 2.121,
"step": 36
},
{
"epoch": 0.03274336283185841,
"grad_norm": 23.423933029174805,
"learning_rate": 1.1674528301886792e-07,
"loss": 5.9026,
"step": 37
},
{
"epoch": 0.033628318584070796,
"grad_norm": 46.25591278076172,
"learning_rate": 1.2028301886792452e-07,
"loss": 7.3796,
"step": 38
},
{
"epoch": 0.034513274336283185,
"grad_norm": 20.376422882080078,
"learning_rate": 1.2382075471698114e-07,
"loss": 5.5361,
"step": 39
},
{
"epoch": 0.035398230088495575,
"grad_norm": 12.82562255859375,
"learning_rate": 1.2735849056603773e-07,
"loss": 4.0243,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_Qnli-dev_cosine_accuracy": 0.5859375,
"eval_Qnli-dev_cosine_accuracy_threshold": 0.9302856922149658,
"eval_Qnli-dev_cosine_ap": 0.5480269179285036,
"eval_Qnli-dev_cosine_f1": 0.6315789473684211,
"eval_Qnli-dev_cosine_f1_threshold": 0.7634451389312744,
"eval_Qnli-dev_cosine_precision": 0.4633663366336634,
"eval_Qnli-dev_cosine_recall": 0.9915254237288136,
"eval_Qnli-dev_dot_accuracy": 0.5859375,
"eval_Qnli-dev_dot_accuracy_threshold": 714.4895629882812,
"eval_Qnli-dev_dot_ap": 0.548060663242546,
"eval_Qnli-dev_dot_f1": 0.6315789473684211,
"eval_Qnli-dev_dot_f1_threshold": 586.342529296875,
"eval_Qnli-dev_dot_precision": 0.4633663366336634,
"eval_Qnli-dev_dot_recall": 0.9915254237288136,
"eval_Qnli-dev_euclidean_accuracy": 0.5859375,
"eval_Qnli-dev_euclidean_accuracy_threshold": 10.348224639892578,
"eval_Qnli-dev_euclidean_ap": 0.5480269179285036,
"eval_Qnli-dev_euclidean_f1": 0.6315789473684211,
"eval_Qnli-dev_euclidean_f1_threshold": 19.05518341064453,
"eval_Qnli-dev_euclidean_precision": 0.4633663366336634,
"eval_Qnli-dev_euclidean_recall": 0.9915254237288136,
"eval_Qnli-dev_manhattan_accuracy": 0.59765625,
"eval_Qnli-dev_manhattan_accuracy_threshold": 175.22628784179688,
"eval_Qnli-dev_manhattan_ap": 0.5780924813828909,
"eval_Qnli-dev_manhattan_f1": 0.6291834002677376,
"eval_Qnli-dev_manhattan_f1_threshold": 334.39178466796875,
"eval_Qnli-dev_manhattan_precision": 0.4598825831702544,
"eval_Qnli-dev_manhattan_recall": 0.9957627118644068,
"eval_Qnli-dev_max_accuracy": 0.59765625,
"eval_Qnli-dev_max_accuracy_threshold": 714.4895629882812,
"eval_Qnli-dev_max_ap": 0.5780924813828909,
"eval_Qnli-dev_max_f1": 0.6315789473684211,
"eval_Qnli-dev_max_f1_threshold": 586.342529296875,
"eval_Qnli-dev_max_precision": 0.4633663366336634,
"eval_Qnli-dev_max_recall": 0.9957627118644068,
"eval_allNLI-dev_cosine_accuracy": 0.6640625,
"eval_allNLI-dev_cosine_accuracy_threshold": 0.9888672828674316,
"eval_allNLI-dev_cosine_ap": 0.32886365768247516,
"eval_allNLI-dev_cosine_f1": 0.5095729013254787,
"eval_allNLI-dev_cosine_f1_threshold": 0.7477295398712158,
"eval_allNLI-dev_cosine_precision": 0.34189723320158105,
"eval_allNLI-dev_cosine_recall": 1.0,
"eval_allNLI-dev_dot_accuracy": 0.6640625,
"eval_allNLI-dev_dot_accuracy_threshold": 759.483154296875,
"eval_allNLI-dev_dot_ap": 0.3288581611938815,
"eval_allNLI-dev_dot_f1": 0.5095729013254787,
"eval_allNLI-dev_dot_f1_threshold": 574.2760620117188,
"eval_allNLI-dev_dot_precision": 0.34189723320158105,
"eval_allNLI-dev_dot_recall": 1.0,
"eval_allNLI-dev_euclidean_accuracy": 0.6640625,
"eval_allNLI-dev_euclidean_accuracy_threshold": 3.8085508346557617,
"eval_allNLI-dev_euclidean_ap": 0.32886365768247516,
"eval_allNLI-dev_euclidean_f1": 0.5095729013254787,
"eval_allNLI-dev_euclidean_f1_threshold": 19.684810638427734,
"eval_allNLI-dev_euclidean_precision": 0.34189723320158105,
"eval_allNLI-dev_euclidean_recall": 1.0,
"eval_allNLI-dev_manhattan_accuracy": 0.6640625,
"eval_allNLI-dev_manhattan_accuracy_threshold": 65.93238830566406,
"eval_allNLI-dev_manhattan_ap": 0.33852594919898543,
"eval_allNLI-dev_manhattan_f1": 0.5058479532163743,
"eval_allNLI-dev_manhattan_f1_threshold": 335.4263916015625,
"eval_allNLI-dev_manhattan_precision": 0.3385518590998043,
"eval_allNLI-dev_manhattan_recall": 1.0,
"eval_allNLI-dev_max_accuracy": 0.6640625,
"eval_allNLI-dev_max_accuracy_threshold": 759.483154296875,
"eval_allNLI-dev_max_ap": 0.33852594919898543,
"eval_allNLI-dev_max_f1": 0.5095729013254787,
"eval_allNLI-dev_max_f1_threshold": 574.2760620117188,
"eval_allNLI-dev_max_precision": 0.34189723320158105,
"eval_allNLI-dev_max_recall": 1.0,
"eval_sequential_score": 0.5780924813828909,
"eval_sts-test_pearson_cosine": 0.1533465318414369,
"eval_sts-test_pearson_dot": 0.15333057450060855,
"eval_sts-test_pearson_euclidean": 0.1664717893342273,
"eval_sts-test_pearson_manhattan": 0.20717970064899288,
"eval_sts-test_pearson_max": 0.20717970064899288,
"eval_sts-test_spearman_cosine": 0.18786210334203038,
"eval_sts-test_spearman_dot": 0.1878347337472397,
"eval_sts-test_spearman_euclidean": 0.18786046572196458,
"eval_sts-test_spearman_manhattan": 0.22429466463153608,
"eval_sts-test_spearman_max": 0.22429466463153608,
"eval_vitaminc-pairs_loss": 2.901831865310669,
"eval_vitaminc-pairs_runtime": 4.078,
"eval_vitaminc-pairs_samples_per_second": 31.388,
"eval_vitaminc-pairs_steps_per_second": 0.245,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_negation-triplets_loss": 5.690315246582031,
"eval_negation-triplets_runtime": 0.7141,
"eval_negation-triplets_samples_per_second": 179.254,
"eval_negation-triplets_steps_per_second": 1.4,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_scitail-pairs-pos_loss": 2.1135852336883545,
"eval_scitail-pairs-pos_runtime": 0.8282,
"eval_scitail-pairs-pos_samples_per_second": 154.543,
"eval_scitail-pairs-pos_steps_per_second": 1.207,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_scitail-pairs-qa_loss": 2.8052029609680176,
"eval_scitail-pairs-qa_runtime": 0.5471,
"eval_scitail-pairs-qa_samples_per_second": 233.943,
"eval_scitail-pairs-qa_steps_per_second": 1.828,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_xsum-pairs_loss": 6.583061695098877,
"eval_xsum-pairs_runtime": 2.8921,
"eval_xsum-pairs_samples_per_second": 44.259,
"eval_xsum-pairs_steps_per_second": 0.346,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_sciq_pairs_loss": 0.8882207870483398,
"eval_sciq_pairs_runtime": 3.7993,
"eval_sciq_pairs_samples_per_second": 33.69,
"eval_sciq_pairs_steps_per_second": 0.263,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_qasc_pairs_loss": 4.1147541999816895,
"eval_qasc_pairs_runtime": 0.6768,
"eval_qasc_pairs_samples_per_second": 189.125,
"eval_qasc_pairs_steps_per_second": 1.478,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_openbookqa_pairs_loss": 5.096628665924072,
"eval_openbookqa_pairs_runtime": 0.5776,
"eval_openbookqa_pairs_samples_per_second": 221.615,
"eval_openbookqa_pairs_steps_per_second": 1.731,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_msmarco_pairs_loss": 10.391141891479492,
"eval_msmarco_pairs_runtime": 1.2577,
"eval_msmarco_pairs_samples_per_second": 101.77,
"eval_msmarco_pairs_steps_per_second": 0.795,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_nq_pairs_loss": 10.903197288513184,
"eval_nq_pairs_runtime": 2.5051,
"eval_nq_pairs_samples_per_second": 51.095,
"eval_nq_pairs_steps_per_second": 0.399,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_trivia_pairs_loss": 7.190384387969971,
"eval_trivia_pairs_runtime": 3.6482,
"eval_trivia_pairs_samples_per_second": 35.085,
"eval_trivia_pairs_steps_per_second": 0.274,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_gooaq_pairs_loss": 8.193528175354004,
"eval_gooaq_pairs_runtime": 0.9648,
"eval_gooaq_pairs_samples_per_second": 132.67,
"eval_gooaq_pairs_steps_per_second": 1.036,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_paws-pos_loss": 1.3942564725875854,
"eval_paws-pos_runtime": 0.6718,
"eval_paws-pos_samples_per_second": 190.538,
"eval_paws-pos_steps_per_second": 1.489,
"step": 40
},
{
"epoch": 0.035398230088495575,
"eval_global_dataset_loss": 5.671571731567383,
"eval_global_dataset_runtime": 23.0452,
"eval_global_dataset_samples_per_second": 28.77,
"eval_global_dataset_steps_per_second": 0.26,
"step": 40
},
{
"epoch": 0.036283185840707964,
"grad_norm": 18.026830673217773,
"learning_rate": 1.3089622641509433e-07,
"loss": 4.9072,
"step": 41
},
{
"epoch": 0.03716814159292035,
"grad_norm": 15.423810958862305,
"learning_rate": 1.3443396226415095e-07,
"loss": 3.4439,
"step": 42
},
{
"epoch": 0.03805309734513274,
"grad_norm": 16.31403160095215,
"learning_rate": 1.3797169811320754e-07,
"loss": 4.9787,
"step": 43
},
{
"epoch": 0.03893805309734513,
"grad_norm": 21.37955093383789,
"learning_rate": 1.4150943396226417e-07,
"loss": 5.8318,
"step": 44
},
{
"epoch": 0.03982300884955752,
"grad_norm": 18.23583984375,
"learning_rate": 1.4504716981132076e-07,
"loss": 5.3226,
"step": 45
},
{
"epoch": 0.04070796460176991,
"grad_norm": 20.878713607788086,
"learning_rate": 1.4858490566037738e-07,
"loss": 5.1181,
"step": 46
},
{
"epoch": 0.0415929203539823,
"grad_norm": 18.71149444580078,
"learning_rate": 1.5212264150943398e-07,
"loss": 4.7834,
"step": 47
},
{
"epoch": 0.04247787610619469,
"grad_norm": 38.85902786254883,
"learning_rate": 1.5566037735849057e-07,
"loss": 6.6303,
"step": 48
},
{
"epoch": 0.04336283185840708,
"grad_norm": 37.41562271118164,
"learning_rate": 1.591981132075472e-07,
"loss": 5.8171,
"step": 49
},
{
"epoch": 0.04424778761061947,
"grad_norm": 17.541080474853516,
"learning_rate": 1.627358490566038e-07,
"loss": 5.1962,
"step": 50
},
{
"epoch": 0.04513274336283186,
"grad_norm": 16.145116806030273,
"learning_rate": 1.6627358490566038e-07,
"loss": 5.2096,
"step": 51
},
{
"epoch": 0.04601769911504425,
"grad_norm": 20.175189971923828,
"learning_rate": 1.6981132075471698e-07,
"loss": 5.0943,
"step": 52
},
{
"epoch": 0.046902654867256637,
"grad_norm": 13.441214561462402,
"learning_rate": 1.733490566037736e-07,
"loss": 4.9038,
"step": 53
},
{
"epoch": 0.047787610619469026,
"grad_norm": 13.396607398986816,
"learning_rate": 1.768867924528302e-07,
"loss": 4.6479,
"step": 54
},
{
"epoch": 0.048672566371681415,
"grad_norm": 13.68046760559082,
"learning_rate": 1.804245283018868e-07,
"loss": 5.5098,
"step": 55
},
{
"epoch": 0.049557522123893805,
"grad_norm": 13.278443336486816,
"learning_rate": 1.839622641509434e-07,
"loss": 4.6979,
"step": 56
},
{
"epoch": 0.050442477876106194,
"grad_norm": 15.295453071594238,
"learning_rate": 1.875e-07,
"loss": 3.1969,
"step": 57
},
{
"epoch": 0.05132743362831858,
"grad_norm": 12.185781478881836,
"learning_rate": 1.910377358490566e-07,
"loss": 4.4127,
"step": 58
},
{
"epoch": 0.05221238938053097,
"grad_norm": 10.874494552612305,
"learning_rate": 1.9457547169811322e-07,
"loss": 3.7746,
"step": 59
},
{
"epoch": 0.05309734513274336,
"grad_norm": 9.654823303222656,
"learning_rate": 1.9811320754716982e-07,
"loss": 4.5378,
"step": 60
},
{
"epoch": 0.05398230088495575,
"grad_norm": 21.123645782470703,
"learning_rate": 2.016509433962264e-07,
"loss": 5.0209,
"step": 61
},
{
"epoch": 0.05486725663716814,
"grad_norm": 33.47934341430664,
"learning_rate": 2.0518867924528303e-07,
"loss": 6.5936,
"step": 62
},
{
"epoch": 0.05575221238938053,
"grad_norm": 10.2566556930542,
"learning_rate": 2.0872641509433963e-07,
"loss": 4.2315,
"step": 63
},
{
"epoch": 0.05663716814159292,
"grad_norm": 28.198625564575195,
"learning_rate": 2.1226415094339622e-07,
"loss": 6.4269,
"step": 64
},
{
"epoch": 0.05752212389380531,
"grad_norm": 9.386558532714844,
"learning_rate": 2.1580188679245282e-07,
"loss": 4.2644,
"step": 65
},
{
"epoch": 0.0584070796460177,
"grad_norm": 12.687555313110352,
"learning_rate": 2.1933962264150944e-07,
"loss": 5.1388,
"step": 66
},
{
"epoch": 0.05929203539823009,
"grad_norm": 14.834878921508789,
"learning_rate": 2.2287735849056603e-07,
"loss": 5.1852,
"step": 67
},
{
"epoch": 0.06017699115044248,
"grad_norm": 10.888677597045898,
"learning_rate": 2.2641509433962263e-07,
"loss": 4.8057,
"step": 68
},
{
"epoch": 0.061061946902654866,
"grad_norm": 13.97256851196289,
"learning_rate": 2.2995283018867925e-07,
"loss": 3.1725,
"step": 69
},
{
"epoch": 0.061946902654867256,
"grad_norm": 11.82534122467041,
"learning_rate": 2.3349056603773584e-07,
"loss": 3.3322,
"step": 70
},
{
"epoch": 0.06283185840707965,
"grad_norm": 16.99266242980957,
"learning_rate": 2.3702830188679244e-07,
"loss": 5.139,
"step": 71
},
{
"epoch": 0.06371681415929203,
"grad_norm": 8.74513053894043,
"learning_rate": 2.4056603773584903e-07,
"loss": 4.307,
"step": 72
},
{
"epoch": 0.06460176991150443,
"grad_norm": 11.715869903564453,
"learning_rate": 2.4410377358490563e-07,
"loss": 5.0133,
"step": 73
},
{
"epoch": 0.06548672566371681,
"grad_norm": 9.844196319580078,
"learning_rate": 2.476415094339623e-07,
"loss": 4.0507,
"step": 74
},
{
"epoch": 0.06637168141592921,
"grad_norm": 12.447444915771484,
"learning_rate": 2.5117924528301887e-07,
"loss": 3.3895,
"step": 75
},
{
"epoch": 0.06725663716814159,
"grad_norm": 23.91596794128418,
"learning_rate": 2.5471698113207547e-07,
"loss": 5.6736,
"step": 76
},
{
"epoch": 0.06814159292035399,
"grad_norm": 9.635603904724121,
"learning_rate": 2.5825471698113206e-07,
"loss": 4.2572,
"step": 77
},
{
"epoch": 0.06902654867256637,
"grad_norm": 14.971665382385254,
"learning_rate": 2.6179245283018866e-07,
"loss": 3.0796,
"step": 78
},
{
"epoch": 0.06991150442477877,
"grad_norm": 11.226128578186035,
"learning_rate": 2.6533018867924525e-07,
"loss": 5.0199,
"step": 79
},
{
"epoch": 0.07079646017699115,
"grad_norm": 11.01388931274414,
"learning_rate": 2.688679245283019e-07,
"loss": 4.1414,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_Qnli-dev_cosine_accuracy": 0.591796875,
"eval_Qnli-dev_cosine_accuracy_threshold": 0.9258557558059692,
"eval_Qnli-dev_cosine_ap": 0.5585355274462735,
"eval_Qnli-dev_cosine_f1": 0.6291834002677376,
"eval_Qnli-dev_cosine_f1_threshold": 0.750666618347168,
"eval_Qnli-dev_cosine_precision": 0.4598825831702544,
"eval_Qnli-dev_cosine_recall": 0.9957627118644068,
"eval_Qnli-dev_dot_accuracy": 0.591796875,
"eval_Qnli-dev_dot_accuracy_threshold": 711.18359375,
"eval_Qnli-dev_dot_ap": 0.5585297234749824,
"eval_Qnli-dev_dot_f1": 0.6291834002677376,
"eval_Qnli-dev_dot_f1_threshold": 576.5970458984375,
"eval_Qnli-dev_dot_precision": 0.4598825831702544,
"eval_Qnli-dev_dot_recall": 0.9957627118644068,
"eval_Qnli-dev_euclidean_accuracy": 0.591796875,
"eval_Qnli-dev_euclidean_accuracy_threshold": 10.672666549682617,
"eval_Qnli-dev_euclidean_ap": 0.5585355274462735,
"eval_Qnli-dev_euclidean_f1": 0.6291834002677376,
"eval_Qnli-dev_euclidean_f1_threshold": 19.553747177124023,
"eval_Qnli-dev_euclidean_precision": 0.4598825831702544,
"eval_Qnli-dev_euclidean_recall": 0.9957627118644068,
"eval_Qnli-dev_manhattan_accuracy": 0.619140625,
"eval_Qnli-dev_manhattan_accuracy_threshold": 188.09068298339844,
"eval_Qnli-dev_manhattan_ap": 0.5898283705050701,
"eval_Qnli-dev_manhattan_f1": 0.6301775147928994,
"eval_Qnli-dev_manhattan_f1_threshold": 237.80462646484375,
"eval_Qnli-dev_manhattan_precision": 0.48409090909090907,
"eval_Qnli-dev_manhattan_recall": 0.902542372881356,
"eval_Qnli-dev_max_accuracy": 0.619140625,
"eval_Qnli-dev_max_accuracy_threshold": 711.18359375,
"eval_Qnli-dev_max_ap": 0.5898283705050701,
"eval_Qnli-dev_max_f1": 0.6301775147928994,
"eval_Qnli-dev_max_f1_threshold": 576.5970458984375,
"eval_Qnli-dev_max_precision": 0.48409090909090907,
"eval_Qnli-dev_max_recall": 0.9957627118644068,
"eval_allNLI-dev_cosine_accuracy": 0.666015625,
"eval_allNLI-dev_cosine_accuracy_threshold": 0.983686089515686,
"eval_allNLI-dev_cosine_ap": 0.34411819659341086,
"eval_allNLI-dev_cosine_f1": 0.5065885797950219,
"eval_allNLI-dev_cosine_f1_threshold": 0.7642872333526611,
"eval_allNLI-dev_cosine_precision": 0.3392156862745098,
"eval_allNLI-dev_cosine_recall": 1.0,
"eval_allNLI-dev_dot_accuracy": 0.666015625,
"eval_allNLI-dev_dot_accuracy_threshold": 755.60302734375,
"eval_allNLI-dev_dot_ap": 0.344109544232086,
"eval_allNLI-dev_dot_f1": 0.5065885797950219,
"eval_allNLI-dev_dot_f1_threshold": 587.0625,
"eval_allNLI-dev_dot_precision": 0.3392156862745098,
"eval_allNLI-dev_dot_recall": 1.0,
"eval_allNLI-dev_euclidean_accuracy": 0.666015625,
"eval_allNLI-dev_euclidean_accuracy_threshold": 5.00581693649292,
"eval_allNLI-dev_euclidean_ap": 0.3441246898925644,
"eval_allNLI-dev_euclidean_f1": 0.5065885797950219,
"eval_allNLI-dev_euclidean_f1_threshold": 19.022436141967773,
"eval_allNLI-dev_euclidean_precision": 0.3392156862745098,
"eval_allNLI-dev_euclidean_recall": 1.0,
"eval_allNLI-dev_manhattan_accuracy": 0.6640625,
"eval_allNLI-dev_manhattan_accuracy_threshold": 62.69102096557617,
"eval_allNLI-dev_manhattan_ap": 0.35131239981425566,
"eval_allNLI-dev_manhattan_f1": 0.5058479532163743,
"eval_allNLI-dev_manhattan_f1_threshold": 337.6861877441406,
"eval_allNLI-dev_manhattan_precision": 0.3385518590998043,
"eval_allNLI-dev_manhattan_recall": 1.0,
"eval_allNLI-dev_max_accuracy": 0.666015625,
"eval_allNLI-dev_max_accuracy_threshold": 755.60302734375,
"eval_allNLI-dev_max_ap": 0.35131239981425566,
"eval_allNLI-dev_max_f1": 0.5065885797950219,
"eval_allNLI-dev_max_f1_threshold": 587.0625,
"eval_allNLI-dev_max_precision": 0.3392156862745098,
"eval_allNLI-dev_max_recall": 1.0,
"eval_sequential_score": 0.5898283705050701,
"eval_sts-test_pearson_cosine": 0.22248205020578934,
"eval_sts-test_pearson_dot": 0.22239084967931927,
"eval_sts-test_pearson_euclidean": 0.2323160413842197,
"eval_sts-test_pearson_manhattan": 0.26632593273308647,
"eval_sts-test_pearson_max": 0.26632593273308647,
"eval_sts-test_spearman_cosine": 0.24802235964390085,
"eval_sts-test_spearman_dot": 0.24791612015173234,
"eval_sts-test_spearman_euclidean": 0.24799036249272113,
"eval_sts-test_spearman_manhattan": 0.2843623073856928,
"eval_sts-test_spearman_max": 0.2843623073856928,
"eval_vitaminc-pairs_loss": 2.7793872356414795,
"eval_vitaminc-pairs_runtime": 3.7649,
"eval_vitaminc-pairs_samples_per_second": 33.998,
"eval_vitaminc-pairs_steps_per_second": 0.266,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_negation-triplets_loss": 4.888970851898193,
"eval_negation-triplets_runtime": 0.7134,
"eval_negation-triplets_samples_per_second": 179.432,
"eval_negation-triplets_steps_per_second": 1.402,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_scitail-pairs-pos_loss": 1.8996644020080566,
"eval_scitail-pairs-pos_runtime": 0.8506,
"eval_scitail-pairs-pos_samples_per_second": 150.477,
"eval_scitail-pairs-pos_steps_per_second": 1.176,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_scitail-pairs-qa_loss": 2.6760551929473877,
"eval_scitail-pairs-qa_runtime": 0.5685,
"eval_scitail-pairs-qa_samples_per_second": 225.171,
"eval_scitail-pairs-qa_steps_per_second": 1.759,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_xsum-pairs_loss": 6.209648609161377,
"eval_xsum-pairs_runtime": 2.9221,
"eval_xsum-pairs_samples_per_second": 43.804,
"eval_xsum-pairs_steps_per_second": 0.342,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_sciq_pairs_loss": 0.7622462511062622,
"eval_sciq_pairs_runtime": 3.7816,
"eval_sciq_pairs_samples_per_second": 33.848,
"eval_sciq_pairs_steps_per_second": 0.264,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_qasc_pairs_loss": 3.3129472732543945,
"eval_qasc_pairs_runtime": 0.6761,
"eval_qasc_pairs_samples_per_second": 189.334,
"eval_qasc_pairs_steps_per_second": 1.479,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_openbookqa_pairs_loss": 4.549765586853027,
"eval_openbookqa_pairs_runtime": 0.5767,
"eval_openbookqa_pairs_samples_per_second": 221.954,
"eval_openbookqa_pairs_steps_per_second": 1.734,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_msmarco_pairs_loss": 7.205582141876221,
"eval_msmarco_pairs_runtime": 1.2621,
"eval_msmarco_pairs_samples_per_second": 101.416,
"eval_msmarco_pairs_steps_per_second": 0.792,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_nq_pairs_loss": 7.680945873260498,
"eval_nq_pairs_runtime": 2.5052,
"eval_nq_pairs_samples_per_second": 51.095,
"eval_nq_pairs_steps_per_second": 0.399,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_trivia_pairs_loss": 6.37924861907959,
"eval_trivia_pairs_runtime": 3.6293,
"eval_trivia_pairs_samples_per_second": 35.268,
"eval_trivia_pairs_steps_per_second": 0.276,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_gooaq_pairs_loss": 6.656675338745117,
"eval_gooaq_pairs_runtime": 0.9698,
"eval_gooaq_pairs_samples_per_second": 131.988,
"eval_gooaq_pairs_steps_per_second": 1.031,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_paws-pos_loss": 1.3848179578781128,
"eval_paws-pos_runtime": 0.6727,
"eval_paws-pos_samples_per_second": 190.278,
"eval_paws-pos_steps_per_second": 1.487,
"step": 80
},
{
"epoch": 0.07079646017699115,
"eval_global_dataset_loss": 5.002967834472656,
"eval_global_dataset_runtime": 23.048,
"eval_global_dataset_samples_per_second": 28.766,
"eval_global_dataset_steps_per_second": 0.26,
"step": 80
},
{
"epoch": 0.07168141592920355,
"grad_norm": 18.9890193939209,
"learning_rate": 2.724056603773585e-07,
"loss": 5.8604,
"step": 81
},
{
"epoch": 0.07256637168141593,
"grad_norm": 8.206193923950195,
"learning_rate": 2.759433962264151e-07,
"loss": 4.3003,
"step": 82
},
{
"epoch": 0.07345132743362832,
"grad_norm": 10.03178882598877,
"learning_rate": 2.794811320754717e-07,
"loss": 4.4568,
"step": 83
},
{
"epoch": 0.0743362831858407,
"grad_norm": 14.74673080444336,
"learning_rate": 2.8301886792452833e-07,
"loss": 4.2747,
"step": 84
},
{
"epoch": 0.0752212389380531,
"grad_norm": 19.097232818603516,
"learning_rate": 2.865566037735849e-07,
"loss": 5.52,
"step": 85
},
{
"epoch": 0.07610619469026549,
"grad_norm": 14.828218460083008,
"learning_rate": 2.900943396226415e-07,
"loss": 2.7767,
"step": 86
},
{
"epoch": 0.07699115044247788,
"grad_norm": 9.30789566040039,
"learning_rate": 2.936320754716981e-07,
"loss": 4.397,
"step": 87
},
{
"epoch": 0.07787610619469026,
"grad_norm": 15.119461059570312,
"learning_rate": 2.9716981132075476e-07,
"loss": 5.4449,
"step": 88
},
{
"epoch": 0.07876106194690266,
"grad_norm": 8.459301948547363,
"learning_rate": 3.0070754716981136e-07,
"loss": 4.2706,
"step": 89
},
{
"epoch": 0.07964601769911504,
"grad_norm": 23.59125518798828,
"learning_rate": 3.0424528301886795e-07,
"loss": 6.4759,
"step": 90
},
{
"epoch": 0.08053097345132744,
"grad_norm": 8.729449272155762,
"learning_rate": 3.0778301886792455e-07,
"loss": 4.1951,
"step": 91
},
{
"epoch": 0.08141592920353982,
"grad_norm": 8.37271785736084,
"learning_rate": 3.1132075471698114e-07,
"loss": 4.6328,
"step": 92
},
{
"epoch": 0.08230088495575222,
"grad_norm": 10.029474258422852,
"learning_rate": 3.1485849056603774e-07,
"loss": 4.1278,
"step": 93
},
{
"epoch": 0.0831858407079646,
"grad_norm": 8.706567764282227,
"learning_rate": 3.183962264150944e-07,
"loss": 4.1787,
"step": 94
},
{
"epoch": 0.084070796460177,
"grad_norm": 13.88837718963623,
"learning_rate": 3.21933962264151e-07,
"loss": 5.2156,
"step": 95
},
{
"epoch": 0.08495575221238938,
"grad_norm": 12.01068115234375,
"learning_rate": 3.254716981132076e-07,
"loss": 3.1403,
"step": 96
},
{
"epoch": 0.08584070796460178,
"grad_norm": 8.432968139648438,
"learning_rate": 3.2900943396226417e-07,
"loss": 4.0273,
"step": 97
},
{
"epoch": 0.08672566371681416,
"grad_norm": 12.645098686218262,
"learning_rate": 3.3254716981132077e-07,
"loss": 3.0624,
"step": 98
},
{
"epoch": 0.08761061946902655,
"grad_norm": 11.483688354492188,
"learning_rate": 3.3608490566037736e-07,
"loss": 4.6786,
"step": 99
},
{
"epoch": 0.08849557522123894,
"grad_norm": 8.645537376403809,
"learning_rate": 3.3962264150943395e-07,
"loss": 4.1505,
"step": 100
},
{
"epoch": 0.08938053097345133,
"grad_norm": 13.053335189819336,
"learning_rate": 3.431603773584906e-07,
"loss": 2.9529,
"step": 101
},
{
"epoch": 0.09026548672566372,
"grad_norm": 14.494400978088379,
"learning_rate": 3.466981132075472e-07,
"loss": 4.7048,
"step": 102
},
{
"epoch": 0.09115044247787611,
"grad_norm": 9.513616561889648,
"learning_rate": 3.502358490566038e-07,
"loss": 4.7388,
"step": 103
},
{
"epoch": 0.0920353982300885,
"grad_norm": 9.751347541809082,
"learning_rate": 3.537735849056604e-07,
"loss": 3.7879,
"step": 104
},
{
"epoch": 0.09292035398230089,
"grad_norm": 9.06558895111084,
"learning_rate": 3.57311320754717e-07,
"loss": 4.0311,
"step": 105
},
{
"epoch": 0.09380530973451327,
"grad_norm": 9.53257942199707,
"learning_rate": 3.608490566037736e-07,
"loss": 4.1314,
"step": 106
},
{
"epoch": 0.09469026548672567,
"grad_norm": 11.554676055908203,
"learning_rate": 3.643867924528302e-07,
"loss": 4.9411,
"step": 107
},
{
"epoch": 0.09557522123893805,
"grad_norm": 8.559597969055176,
"learning_rate": 3.679245283018868e-07,
"loss": 4.1118,
"step": 108
},
{
"epoch": 0.09646017699115045,
"grad_norm": 10.008039474487305,
"learning_rate": 3.714622641509434e-07,
"loss": 3.6971,
"step": 109
},
{
"epoch": 0.09734513274336283,
"grad_norm": 16.543254852294922,
"learning_rate": 3.75e-07,
"loss": 5.605,
"step": 110
},
{
"epoch": 0.09823008849557523,
"grad_norm": 11.816540718078613,
"learning_rate": 3.7853773584905666e-07,
"loss": 3.4563,
"step": 111
},
{
"epoch": 0.09911504424778761,
"grad_norm": 10.638028144836426,
"learning_rate": 3.820754716981132e-07,
"loss": 3.7422,
"step": 112
},
{
"epoch": 0.1,
"grad_norm": 8.5276460647583,
"learning_rate": 3.8561320754716985e-07,
"loss": 3.8055,
"step": 113
},
{
"epoch": 0.10088495575221239,
"grad_norm": 13.437420845031738,
"learning_rate": 3.8915094339622644e-07,
"loss": 5.2369,
"step": 114
},
{
"epoch": 0.10176991150442478,
"grad_norm": 21.039424896240234,
"learning_rate": 3.926886792452831e-07,
"loss": 5.6518,
"step": 115
},
{
"epoch": 0.10265486725663717,
"grad_norm": 13.487382888793945,
"learning_rate": 3.9622641509433963e-07,
"loss": 3.2906,
"step": 116
},
{
"epoch": 0.10353982300884956,
"grad_norm": 11.895822525024414,
"learning_rate": 3.997641509433963e-07,
"loss": 3.4996,
"step": 117
},
{
"epoch": 0.10442477876106195,
"grad_norm": 10.83902359008789,
"learning_rate": 4.033018867924528e-07,
"loss": 3.6283,
"step": 118
},
{
"epoch": 0.10530973451327434,
"grad_norm": 10.552660942077637,
"learning_rate": 4.0683962264150947e-07,
"loss": 4.1487,
"step": 119
},
{
"epoch": 0.10619469026548672,
"grad_norm": 9.924088478088379,
"learning_rate": 4.1037735849056606e-07,
"loss": 4.3996,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_Qnli-dev_cosine_accuracy": 0.595703125,
"eval_Qnli-dev_cosine_accuracy_threshold": 0.9275249242782593,
"eval_Qnli-dev_cosine_ap": 0.5645920090286662,
"eval_Qnli-dev_cosine_f1": 0.6327077747989276,
"eval_Qnli-dev_cosine_f1_threshold": 0.7267085313796997,
"eval_Qnli-dev_cosine_precision": 0.4627450980392157,
"eval_Qnli-dev_cosine_recall": 1.0,
"eval_Qnli-dev_dot_accuracy": 0.595703125,
"eval_Qnli-dev_dot_accuracy_threshold": 712.4608154296875,
"eval_Qnli-dev_dot_ap": 0.5646837736357366,
"eval_Qnli-dev_dot_f1": 0.6327077747989276,
"eval_Qnli-dev_dot_f1_threshold": 558.2177734375,
"eval_Qnli-dev_dot_precision": 0.4627450980392157,
"eval_Qnli-dev_dot_recall": 1.0,
"eval_Qnli-dev_euclidean_accuracy": 0.595703125,
"eval_Qnli-dev_euclidean_accuracy_threshold": 10.551876068115234,
"eval_Qnli-dev_euclidean_ap": 0.5645997569733668,
"eval_Qnli-dev_euclidean_f1": 0.6327077747989276,
"eval_Qnli-dev_euclidean_f1_threshold": 20.490163803100586,
"eval_Qnli-dev_euclidean_precision": 0.4627450980392157,
"eval_Qnli-dev_euclidean_recall": 1.0,
"eval_Qnli-dev_manhattan_accuracy": 0.626953125,
"eval_Qnli-dev_manhattan_accuracy_threshold": 195.12744140625,
"eval_Qnli-dev_manhattan_ap": 0.5975206086733145,
"eval_Qnli-dev_manhattan_f1": 0.6322008862629247,
"eval_Qnli-dev_manhattan_f1_threshold": 256.6172180175781,
"eval_Qnli-dev_manhattan_precision": 0.4852607709750567,
"eval_Qnli-dev_manhattan_recall": 0.9067796610169492,
"eval_Qnli-dev_max_accuracy": 0.626953125,
"eval_Qnli-dev_max_accuracy_threshold": 712.4608154296875,
"eval_Qnli-dev_max_ap": 0.5975206086733145,
"eval_Qnli-dev_max_f1": 0.6327077747989276,
"eval_Qnli-dev_max_f1_threshold": 558.2177734375,
"eval_Qnli-dev_max_precision": 0.4852607709750567,
"eval_Qnli-dev_max_recall": 1.0,
"eval_allNLI-dev_cosine_accuracy": 0.666015625,
"eval_allNLI-dev_cosine_accuracy_threshold": 0.983871340751648,
"eval_allNLI-dev_cosine_ap": 0.36035507065342104,
"eval_allNLI-dev_cosine_f1": 0.5051395007342143,
"eval_allNLI-dev_cosine_f1_threshold": 0.7787582874298096,
"eval_allNLI-dev_cosine_precision": 0.33858267716535434,
"eval_allNLI-dev_cosine_recall": 0.9942196531791907,
"eval_allNLI-dev_dot_accuracy": 0.666015625,
"eval_allNLI-dev_dot_accuracy_threshold": 755.7670288085938,
"eval_allNLI-dev_dot_ap": 0.36031241443166284,
"eval_allNLI-dev_dot_f1": 0.5051395007342143,
"eval_allNLI-dev_dot_f1_threshold": 598.2041625976562,
"eval_allNLI-dev_dot_precision": 0.33858267716535434,
"eval_allNLI-dev_dot_recall": 0.9942196531791907,
"eval_allNLI-dev_euclidean_accuracy": 0.666015625,
"eval_allNLI-dev_euclidean_accuracy_threshold": 4.964720249176025,
"eval_allNLI-dev_euclidean_ap": 0.36035507065342104,
"eval_allNLI-dev_euclidean_f1": 0.5051395007342143,
"eval_allNLI-dev_euclidean_f1_threshold": 18.434789657592773,
"eval_allNLI-dev_euclidean_precision": 0.33858267716535434,
"eval_allNLI-dev_euclidean_recall": 0.9942196531791907,
"eval_allNLI-dev_manhattan_accuracy": 0.6640625,
"eval_allNLI-dev_manhattan_accuracy_threshold": 66.59053039550781,
"eval_allNLI-dev_manhattan_ap": 0.3692975841596879,
"eval_allNLI-dev_manhattan_f1": 0.5029239766081871,
"eval_allNLI-dev_manhattan_f1_threshold": 380.123779296875,
"eval_allNLI-dev_manhattan_precision": 0.33659491193737767,
"eval_allNLI-dev_manhattan_recall": 0.9942196531791907,
"eval_allNLI-dev_max_accuracy": 0.666015625,
"eval_allNLI-dev_max_accuracy_threshold": 755.7670288085938,
"eval_allNLI-dev_max_ap": 0.3692975841596879,
"eval_allNLI-dev_max_f1": 0.5051395007342143,
"eval_allNLI-dev_max_f1_threshold": 598.2041625976562,
"eval_allNLI-dev_max_precision": 0.33858267716535434,
"eval_allNLI-dev_max_recall": 0.9942196531791907,
"eval_sequential_score": 0.5975206086733145,
"eval_sts-test_pearson_cosine": 0.2980667522290251,
"eval_sts-test_pearson_dot": 0.29795063801865274,
"eval_sts-test_pearson_euclidean": 0.30279956330153407,
"eval_sts-test_pearson_manhattan": 0.32939035635624725,
"eval_sts-test_pearson_max": 0.32939035635624725,
"eval_sts-test_spearman_cosine": 0.3148821747085771,
"eval_sts-test_spearman_dot": 0.3149517475826025,
"eval_sts-test_spearman_euclidean": 0.31489636085812106,
"eval_sts-test_spearman_manhattan": 0.34558301612848313,
"eval_sts-test_spearman_max": 0.34558301612848313,
"eval_vitaminc-pairs_loss": 2.727938652038574,
"eval_vitaminc-pairs_runtime": 3.7459,
"eval_vitaminc-pairs_samples_per_second": 34.17,
"eval_vitaminc-pairs_steps_per_second": 0.267,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_negation-triplets_loss": 4.394620418548584,
"eval_negation-triplets_runtime": 0.7078,
"eval_negation-triplets_samples_per_second": 180.852,
"eval_negation-triplets_steps_per_second": 1.413,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_scitail-pairs-pos_loss": 1.4130322933197021,
"eval_scitail-pairs-pos_runtime": 0.8587,
"eval_scitail-pairs-pos_samples_per_second": 149.07,
"eval_scitail-pairs-pos_steps_per_second": 1.165,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_scitail-pairs-qa_loss": 2.1150403022766113,
"eval_scitail-pairs-qa_runtime": 0.549,
"eval_scitail-pairs-qa_samples_per_second": 233.163,
"eval_scitail-pairs-qa_steps_per_second": 1.822,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_xsum-pairs_loss": 6.048598289489746,
"eval_xsum-pairs_runtime": 2.9142,
"eval_xsum-pairs_samples_per_second": 43.923,
"eval_xsum-pairs_steps_per_second": 0.343,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_sciq_pairs_loss": 0.7171850800514221,
"eval_sciq_pairs_runtime": 3.7786,
"eval_sciq_pairs_samples_per_second": 33.875,
"eval_sciq_pairs_steps_per_second": 0.265,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_qasc_pairs_loss": 2.96693754196167,
"eval_qasc_pairs_runtime": 0.6718,
"eval_qasc_pairs_samples_per_second": 190.538,
"eval_qasc_pairs_steps_per_second": 1.489,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_openbookqa_pairs_loss": 4.418018341064453,
"eval_openbookqa_pairs_runtime": 0.577,
"eval_openbookqa_pairs_samples_per_second": 221.852,
"eval_openbookqa_pairs_steps_per_second": 1.733,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_msmarco_pairs_loss": 6.302182197570801,
"eval_msmarco_pairs_runtime": 1.2547,
"eval_msmarco_pairs_samples_per_second": 102.016,
"eval_msmarco_pairs_steps_per_second": 0.797,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_nq_pairs_loss": 6.841231822967529,
"eval_nq_pairs_runtime": 2.5052,
"eval_nq_pairs_samples_per_second": 51.094,
"eval_nq_pairs_steps_per_second": 0.399,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_trivia_pairs_loss": 6.201311111450195,
"eval_trivia_pairs_runtime": 3.6311,
"eval_trivia_pairs_samples_per_second": 35.251,
"eval_trivia_pairs_steps_per_second": 0.275,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_gooaq_pairs_loss": 6.098212718963623,
"eval_gooaq_pairs_runtime": 0.9643,
"eval_gooaq_pairs_samples_per_second": 132.741,
"eval_gooaq_pairs_steps_per_second": 1.037,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_paws-pos_loss": 0.9473956823348999,
"eval_paws-pos_runtime": 0.6684,
"eval_paws-pos_samples_per_second": 191.51,
"eval_paws-pos_steps_per_second": 1.496,
"step": 120
},
{
"epoch": 0.10619469026548672,
"eval_global_dataset_loss": 4.385201454162598,
"eval_global_dataset_runtime": 23.0455,
"eval_global_dataset_samples_per_second": 28.769,
"eval_global_dataset_steps_per_second": 0.26,
"step": 120
},
{
"epoch": 0.10707964601769912,
"grad_norm": 12.284002304077148,
"learning_rate": 4.1391509433962266e-07,
"loss": 3.5291,
"step": 121
},
{
"epoch": 0.1079646017699115,
"grad_norm": 10.567977905273438,
"learning_rate": 4.1745283018867925e-07,
"loss": 3.8232,
"step": 122
},
{
"epoch": 0.1088495575221239,
"grad_norm": 11.508279800415039,
"learning_rate": 4.209905660377359e-07,
"loss": 4.6035,
"step": 123
},
{
"epoch": 0.10973451327433628,
"grad_norm": 10.180809020996094,
"learning_rate": 4.2452830188679244e-07,
"loss": 3.7607,
"step": 124
},
{
"epoch": 0.11061946902654868,
"grad_norm": 9.519749641418457,
"learning_rate": 4.280660377358491e-07,
"loss": 3.8461,
"step": 125
},
{
"epoch": 0.11150442477876106,
"grad_norm": 11.971588134765625,
"learning_rate": 4.3160377358490563e-07,
"loss": 3.3413,
"step": 126
},
{
"epoch": 0.11238938053097346,
"grad_norm": 9.211153984069824,
"learning_rate": 4.351415094339623e-07,
"loss": 4.2777,
"step": 127
},
{
"epoch": 0.11327433628318584,
"grad_norm": 12.393014907836914,
"learning_rate": 4.386792452830189e-07,
"loss": 4.3597,
"step": 128
},
{
"epoch": 0.11415929203539824,
"grad_norm": 14.332024574279785,
"learning_rate": 4.422169811320755e-07,
"loss": 3.9046,
"step": 129
},
{
"epoch": 0.11504424778761062,
"grad_norm": 10.091246604919434,
"learning_rate": 4.4575471698113207e-07,
"loss": 4.0527,
"step": 130
},
{
"epoch": 0.11592920353982301,
"grad_norm": 15.043377876281738,
"learning_rate": 4.492924528301887e-07,
"loss": 5.0883,
"step": 131
},
{
"epoch": 0.1168141592920354,
"grad_norm": 12.942100524902344,
"learning_rate": 4.5283018867924526e-07,
"loss": 3.8308,
"step": 132
},
{
"epoch": 0.11769911504424779,
"grad_norm": 11.961737632751465,
"learning_rate": 4.563679245283019e-07,
"loss": 3.572,
"step": 133
},
{
"epoch": 0.11858407079646018,
"grad_norm": 12.325026512145996,
"learning_rate": 4.599056603773585e-07,
"loss": 3.4299,
"step": 134
},
{
"epoch": 0.11946902654867257,
"grad_norm": 12.118773460388184,
"learning_rate": 4.6344339622641515e-07,
"loss": 4.1541,
"step": 135
},
{
"epoch": 0.12035398230088495,
"grad_norm": 11.99026107788086,
"learning_rate": 4.669811320754717e-07,
"loss": 3.584,
"step": 136
},
{
"epoch": 0.12123893805309735,
"grad_norm": 15.083515167236328,
"learning_rate": 4.7051886792452834e-07,
"loss": 5.0977,
"step": 137
},
{
"epoch": 0.12212389380530973,
"grad_norm": 15.059394836425781,
"learning_rate": 4.740566037735849e-07,
"loss": 4.6769,
"step": 138
},
{
"epoch": 0.12300884955752213,
"grad_norm": 8.864882469177246,
"learning_rate": 4.775943396226415e-07,
"loss": 3.8396,
"step": 139
},
{
"epoch": 0.12389380530973451,
"grad_norm": 12.116555213928223,
"learning_rate": 4.811320754716981e-07,
"loss": 3.2875,
"step": 140
},
{
"epoch": 0.12477876106194691,
"grad_norm": 14.214646339416504,
"learning_rate": 4.846698113207547e-07,
"loss": 4.1946,
"step": 141
},
{
"epoch": 0.1256637168141593,
"grad_norm": 16.207908630371094,
"learning_rate": 4.882075471698113e-07,
"loss": 4.9602,
"step": 142
},
{
"epoch": 0.12654867256637167,
"grad_norm": 11.662668228149414,
"learning_rate": 4.917452830188679e-07,
"loss": 4.1531,
"step": 143
},
{
"epoch": 0.12743362831858407,
"grad_norm": 12.429448127746582,
"learning_rate": 4.952830188679246e-07,
"loss": 3.8351,
"step": 144
},
{
"epoch": 0.12831858407079647,
"grad_norm": 11.522616386413574,
"learning_rate": 4.988207547169812e-07,
"loss": 3.112,
"step": 145
},
{
"epoch": 0.12920353982300886,
"grad_norm": 14.556803703308105,
"learning_rate": 5.023584905660377e-07,
"loss": 2.3145,
"step": 146
},
{
"epoch": 0.13008849557522123,
"grad_norm": 12.348714828491211,
"learning_rate": 5.058962264150944e-07,
"loss": 4.0989,
"step": 147
},
{
"epoch": 0.13097345132743363,
"grad_norm": 13.150403022766113,
"learning_rate": 5.094339622641509e-07,
"loss": 3.2173,
"step": 148
},
{
"epoch": 0.13185840707964602,
"grad_norm": 12.066205978393555,
"learning_rate": 5.129716981132076e-07,
"loss": 2.7913,
"step": 149
},
{
"epoch": 0.13274336283185842,
"grad_norm": 11.519116401672363,
"learning_rate": 5.165094339622641e-07,
"loss": 3.7627,
"step": 150
},
{
"epoch": 0.1336283185840708,
"grad_norm": 12.59196662902832,
"learning_rate": 5.200471698113208e-07,
"loss": 3.3669,
"step": 151
},
{
"epoch": 0.13451327433628318,
"grad_norm": 13.791536331176758,
"learning_rate": 5.235849056603773e-07,
"loss": 2.6775,
"step": 152
},
{
"epoch": 0.13539823008849558,
"grad_norm": 11.906597137451172,
"learning_rate": 5.27122641509434e-07,
"loss": 3.2804,
"step": 153
},
{
"epoch": 0.13628318584070798,
"grad_norm": 11.267363548278809,
"learning_rate": 5.306603773584905e-07,
"loss": 3.0676,
"step": 154
},
{
"epoch": 0.13716814159292035,
"grad_norm": 12.373686790466309,
"learning_rate": 5.341981132075471e-07,
"loss": 3.1559,
"step": 155
},
{
"epoch": 0.13805309734513274,
"grad_norm": 13.258451461791992,
"learning_rate": 5.377358490566038e-07,
"loss": 2.6638,
"step": 156
},
{
"epoch": 0.13893805309734514,
"grad_norm": 12.79727554321289,
"learning_rate": 5.412735849056604e-07,
"loss": 2.8045,
"step": 157
},
{
"epoch": 0.13982300884955753,
"grad_norm": 13.88683032989502,
"learning_rate": 5.44811320754717e-07,
"loss": 4.0568,
"step": 158
},
{
"epoch": 0.1407079646017699,
"grad_norm": 12.57358169555664,
"learning_rate": 5.483490566037736e-07,
"loss": 2.7554,
"step": 159
},
{
"epoch": 0.1415929203539823,
"grad_norm": 14.520818710327148,
"learning_rate": 5.518867924528302e-07,
"loss": 3.7407,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_Qnli-dev_cosine_accuracy": 0.62890625,
"eval_Qnli-dev_cosine_accuracy_threshold": 0.9045097827911377,
"eval_Qnli-dev_cosine_ap": 0.6193527955003784,
"eval_Qnli-dev_cosine_f1": 0.6397415185783522,
"eval_Qnli-dev_cosine_f1_threshold": 0.8351442813873291,
"eval_Qnli-dev_cosine_precision": 0.5169712793733682,
"eval_Qnli-dev_cosine_recall": 0.8389830508474576,
"eval_Qnli-dev_dot_accuracy": 0.62890625,
"eval_Qnli-dev_dot_accuracy_threshold": 694.7778930664062,
"eval_Qnli-dev_dot_ap": 0.6194150916988216,
"eval_Qnli-dev_dot_f1": 0.6397415185783522,
"eval_Qnli-dev_dot_f1_threshold": 641.4969482421875,
"eval_Qnli-dev_dot_precision": 0.5169712793733682,
"eval_Qnli-dev_dot_recall": 0.8389830508474576,
"eval_Qnli-dev_euclidean_accuracy": 0.62890625,
"eval_Qnli-dev_euclidean_accuracy_threshold": 12.111844062805176,
"eval_Qnli-dev_euclidean_ap": 0.6193576186776235,
"eval_Qnli-dev_euclidean_f1": 0.6397415185783522,
"eval_Qnli-dev_euclidean_f1_threshold": 15.914146423339844,
"eval_Qnli-dev_euclidean_precision": 0.5169712793733682,
"eval_Qnli-dev_euclidean_recall": 0.8389830508474576,
"eval_Qnli-dev_manhattan_accuracy": 0.646484375,
"eval_Qnli-dev_manhattan_accuracy_threshold": 245.2164306640625,
"eval_Qnli-dev_manhattan_ap": 0.6417015148414534,
"eval_Qnli-dev_manhattan_f1": 0.6521060842433698,
"eval_Qnli-dev_manhattan_f1_threshold": 303.317626953125,
"eval_Qnli-dev_manhattan_precision": 0.5160493827160494,
"eval_Qnli-dev_manhattan_recall": 0.885593220338983,
"eval_Qnli-dev_max_accuracy": 0.646484375,
"eval_Qnli-dev_max_accuracy_threshold": 694.7778930664062,
"eval_Qnli-dev_max_ap": 0.6417015148414534,
"eval_Qnli-dev_max_f1": 0.6521060842433698,
"eval_Qnli-dev_max_f1_threshold": 641.4969482421875,
"eval_Qnli-dev_max_precision": 0.5169712793733682,
"eval_Qnli-dev_max_recall": 0.885593220338983,
"eval_allNLI-dev_cosine_accuracy": 0.66796875,
"eval_allNLI-dev_cosine_accuracy_threshold": 0.9767438173294067,
"eval_allNLI-dev_cosine_ap": 0.38624833037583434,
"eval_allNLI-dev_cosine_f1": 0.5100182149362477,
"eval_allNLI-dev_cosine_f1_threshold": 0.8540960550308228,
"eval_allNLI-dev_cosine_precision": 0.3723404255319149,
"eval_allNLI-dev_cosine_recall": 0.8092485549132948,
"eval_allNLI-dev_dot_accuracy": 0.66796875,
"eval_allNLI-dev_dot_accuracy_threshold": 750.345458984375,
"eval_allNLI-dev_dot_ap": 0.3862261253421553,
"eval_allNLI-dev_dot_f1": 0.5100182149362477,
"eval_allNLI-dev_dot_f1_threshold": 656.0940551757812,
"eval_allNLI-dev_dot_precision": 0.3723404255319149,
"eval_allNLI-dev_dot_recall": 0.8092485549132948,
"eval_allNLI-dev_euclidean_accuracy": 0.66796875,
"eval_allNLI-dev_euclidean_accuracy_threshold": 5.977196216583252,
"eval_allNLI-dev_euclidean_ap": 0.38624380046547035,
"eval_allNLI-dev_euclidean_f1": 0.5100182149362477,
"eval_allNLI-dev_euclidean_f1_threshold": 14.971920013427734,
"eval_allNLI-dev_euclidean_precision": 0.3723404255319149,
"eval_allNLI-dev_euclidean_recall": 0.8092485549132948,
"eval_allNLI-dev_manhattan_accuracy": 0.6640625,
"eval_allNLI-dev_manhattan_accuracy_threshold": 78.52637481689453,
"eval_allNLI-dev_manhattan_ap": 0.3898187083180651,
"eval_allNLI-dev_manhattan_f1": 0.5062388591800357,
"eval_allNLI-dev_manhattan_f1_threshold": 285.7745361328125,
"eval_allNLI-dev_manhattan_precision": 0.36597938144329895,
"eval_allNLI-dev_manhattan_recall": 0.8208092485549133,
"eval_allNLI-dev_max_accuracy": 0.66796875,
"eval_allNLI-dev_max_accuracy_threshold": 750.345458984375,
"eval_allNLI-dev_max_ap": 0.3898187083180651,
"eval_allNLI-dev_max_f1": 0.5100182149362477,
"eval_allNLI-dev_max_f1_threshold": 656.0940551757812,
"eval_allNLI-dev_max_precision": 0.3723404255319149,
"eval_allNLI-dev_max_recall": 0.8208092485549133,
"eval_sequential_score": 0.6417015148414534,
"eval_sts-test_pearson_cosine": 0.2853943019391156,
"eval_sts-test_pearson_dot": 0.28526334639473966,
"eval_sts-test_pearson_euclidean": 0.29405773952219494,
"eval_sts-test_pearson_manhattan": 0.3110310476615048,
"eval_sts-test_pearson_max": 0.3110310476615048,
"eval_sts-test_spearman_cosine": 0.31414239162305135,
"eval_sts-test_spearman_dot": 0.31380407209449446,
"eval_sts-test_spearman_euclidean": 0.3141516551339523,
"eval_sts-test_spearman_manhattan": 0.3366243060620438,
"eval_sts-test_spearman_max": 0.3366243060620438,
"eval_vitaminc-pairs_loss": 2.7439002990722656,
"eval_vitaminc-pairs_runtime": 3.7639,
"eval_vitaminc-pairs_samples_per_second": 34.007,
"eval_vitaminc-pairs_steps_per_second": 0.266,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_negation-triplets_loss": 4.63640022277832,
"eval_negation-triplets_runtime": 0.7072,
"eval_negation-triplets_samples_per_second": 180.999,
"eval_negation-triplets_steps_per_second": 1.414,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_scitail-pairs-pos_loss": 1.0088545083999634,
"eval_scitail-pairs-pos_runtime": 0.8123,
"eval_scitail-pairs-pos_samples_per_second": 157.577,
"eval_scitail-pairs-pos_steps_per_second": 1.231,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_scitail-pairs-qa_loss": 1.1228678226470947,
"eval_scitail-pairs-qa_runtime": 0.5444,
"eval_scitail-pairs-qa_samples_per_second": 235.115,
"eval_scitail-pairs-qa_steps_per_second": 1.837,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_xsum-pairs_loss": 5.4869818687438965,
"eval_xsum-pairs_runtime": 2.8888,
"eval_xsum-pairs_samples_per_second": 44.308,
"eval_xsum-pairs_steps_per_second": 0.346,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_sciq_pairs_loss": 0.628353476524353,
"eval_sciq_pairs_runtime": 3.8061,
"eval_sciq_pairs_samples_per_second": 33.631,
"eval_sciq_pairs_steps_per_second": 0.263,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_qasc_pairs_loss": 2.593322277069092,
"eval_qasc_pairs_runtime": 0.6728,
"eval_qasc_pairs_samples_per_second": 190.241,
"eval_qasc_pairs_steps_per_second": 1.486,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_openbookqa_pairs_loss": 4.394308090209961,
"eval_openbookqa_pairs_runtime": 0.5852,
"eval_openbookqa_pairs_samples_per_second": 218.729,
"eval_openbookqa_pairs_steps_per_second": 1.709,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_msmarco_pairs_loss": 5.656517505645752,
"eval_msmarco_pairs_runtime": 1.2571,
"eval_msmarco_pairs_samples_per_second": 101.822,
"eval_msmarco_pairs_steps_per_second": 0.795,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_nq_pairs_loss": 5.986983776092529,
"eval_nq_pairs_runtime": 2.5075,
"eval_nq_pairs_samples_per_second": 51.047,
"eval_nq_pairs_steps_per_second": 0.399,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_trivia_pairs_loss": 5.694415092468262,
"eval_trivia_pairs_runtime": 3.6302,
"eval_trivia_pairs_samples_per_second": 35.26,
"eval_trivia_pairs_steps_per_second": 0.275,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_gooaq_pairs_loss": 5.3856658935546875,
"eval_gooaq_pairs_runtime": 0.9618,
"eval_gooaq_pairs_samples_per_second": 133.082,
"eval_gooaq_pairs_steps_per_second": 1.04,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_paws-pos_loss": 0.3622308671474457,
"eval_paws-pos_runtime": 0.6678,
"eval_paws-pos_samples_per_second": 191.674,
"eval_paws-pos_steps_per_second": 1.497,
"step": 160
},
{
"epoch": 0.1415929203539823,
"eval_global_dataset_loss": 3.401135206222534,
"eval_global_dataset_runtime": 23.0422,
"eval_global_dataset_samples_per_second": 28.773,
"eval_global_dataset_steps_per_second": 0.26,
"step": 160
}
],
"logging_steps": 1,
"max_steps": 3390,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 80,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 42,
"trial_name": null,
"trial_params": null
}