{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.8, "eval_steps": 179, "global_step": 2864, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.005027932960893855, "grad_norm": 28.300092697143555, "learning_rate": 6.331471135940411e-07, "loss": 1.0198, "step": 18 }, { "epoch": 0.01005586592178771, "grad_norm": 1.059788465499878, "learning_rate": 1.303538175046555e-06, "loss": 0.8564, "step": 36 }, { "epoch": 0.015083798882681564, "grad_norm": 9.368417739868164, "learning_rate": 1.9739292364990693e-06, "loss": 0.685, "step": 54 }, { "epoch": 0.02011173184357542, "grad_norm": 3.4458816051483154, "learning_rate": 2.644320297951583e-06, "loss": 0.7667, "step": 72 }, { "epoch": 0.025139664804469275, "grad_norm": 4.497154712677002, "learning_rate": 3.314711359404097e-06, "loss": 0.7907, "step": 90 }, { "epoch": 0.030167597765363128, "grad_norm": 4.394736289978027, "learning_rate": 3.985102420856611e-06, "loss": 0.9719, "step": 108 }, { "epoch": 0.03519553072625698, "grad_norm": 10.305765151977539, "learning_rate": 4.655493482309125e-06, "loss": 1.1097, "step": 126 }, { "epoch": 0.04022346368715084, "grad_norm": 5.402688980102539, "learning_rate": 5.325884543761639e-06, "loss": 1.0463, "step": 144 }, { "epoch": 0.045251396648044694, "grad_norm": 0.2811428904533386, "learning_rate": 5.996275605214154e-06, "loss": 0.8796, "step": 162 }, { "epoch": 0.05, "eval_nli-pairs_loss": 0.9358150959014893, "eval_nli-pairs_runtime": 3.7163, "eval_nli-pairs_samples_per_second": 40.363, "eval_nli-pairs_steps_per_second": 0.538, "eval_sts-test_pearson_cosine": 0.7920907827756308, "eval_sts-test_pearson_dot": 0.553905804338793, "eval_sts-test_pearson_euclidean": 0.7477067417028286, "eval_sts-test_pearson_manhattan": 0.7454637009944656, "eval_sts-test_pearson_max": 0.7920907827756308, "eval_sts-test_spearman_cosine": 0.8050902340286232, "eval_sts-test_spearman_dot": 0.5325926730168974, "eval_sts-test_spearman_euclidean": 0.73705041377373, "eval_sts-test_spearman_manhattan": 0.7373939486807765, "eval_sts-test_spearman_max": 0.8050902340286232, "step": 179 }, { "epoch": 0.05, "eval_vitaminc-pairs_loss": 4.57532262802124, "eval_vitaminc-pairs_runtime": 0.8386, "eval_vitaminc-pairs_samples_per_second": 149.066, "eval_vitaminc-pairs_steps_per_second": 2.385, "step": 179 }, { "epoch": 0.05, "eval_qnli-contrastive_loss": 0.260235071182251, "eval_qnli-contrastive_runtime": 0.1227, "eval_qnli-contrastive_samples_per_second": 1222.824, "eval_qnli-contrastive_steps_per_second": 16.304, "step": 179 }, { "epoch": 0.05, "eval_scitail-pairs-qa_loss": 0.06116430461406708, "eval_scitail-pairs-qa_runtime": 0.6972, "eval_scitail-pairs-qa_samples_per_second": 215.142, "eval_scitail-pairs-qa_steps_per_second": 2.869, "step": 179 }, { "epoch": 0.05, "eval_scitail-pairs-pos_loss": 0.3001463711261749, "eval_scitail-pairs-pos_runtime": 1.8226, "eval_scitail-pairs-pos_samples_per_second": 82.299, "eval_scitail-pairs-pos_steps_per_second": 1.097, "step": 179 }, { "epoch": 0.05, "eval_xsum-pairs_loss": 0.43996042013168335, "eval_xsum-pairs_runtime": 0.1548, "eval_xsum-pairs_samples_per_second": 529.555, "eval_xsum-pairs_steps_per_second": 6.458, "step": 179 }, { "epoch": 0.05, "eval_compression-pairs_loss": 0.10088926553726196, "eval_compression-pairs_runtime": 0.1051, "eval_compression-pairs_samples_per_second": 1427.213, "eval_compression-pairs_steps_per_second": 19.03, "step": 179 }, { "epoch": 0.05, "eval_sciq_pairs_loss": 0.15509434044361115, "eval_sciq_pairs_runtime": 4.9415, "eval_sciq_pairs_samples_per_second": 30.355, "eval_sciq_pairs_steps_per_second": 0.405, "step": 179 }, { "epoch": 0.05, "eval_qasc_pairs_loss": 0.24072103202342987, "eval_qasc_pairs_runtime": 0.6453, "eval_qasc_pairs_samples_per_second": 232.455, "eval_qasc_pairs_steps_per_second": 3.099, "step": 179 }, { "epoch": 0.05, "eval_openbookqa_pairs_loss": 1.5241280794143677, "eval_openbookqa_pairs_runtime": 0.7059, "eval_openbookqa_pairs_samples_per_second": 212.48, "eval_openbookqa_pairs_steps_per_second": 2.833, "step": 179 }, { "epoch": 0.05, "eval_msmarco_pairs_loss": 0.6149919629096985, "eval_msmarco_pairs_runtime": 1.2013, "eval_msmarco_pairs_samples_per_second": 124.863, "eval_msmarco_pairs_steps_per_second": 1.665, "step": 179 }, { "epoch": 0.05, "eval_nq_pairs_loss": 0.591522216796875, "eval_nq_pairs_runtime": 2.2917, "eval_nq_pairs_samples_per_second": 65.455, "eval_nq_pairs_steps_per_second": 0.873, "step": 179 }, { "epoch": 0.05, "eval_trivia_pairs_loss": 1.0899044275283813, "eval_trivia_pairs_runtime": 3.1583, "eval_trivia_pairs_samples_per_second": 47.494, "eval_trivia_pairs_steps_per_second": 0.633, "step": 179 }, { "epoch": 0.05, "eval_quora_pairs_loss": 0.21252885460853577, "eval_quora_pairs_runtime": 0.3861, "eval_quora_pairs_samples_per_second": 388.452, "eval_quora_pairs_steps_per_second": 5.179, "step": 179 }, { "epoch": 0.05, "eval_gooaq_pairs_loss": 0.5593109130859375, "eval_gooaq_pairs_runtime": 0.7918, "eval_gooaq_pairs_samples_per_second": 189.449, "eval_gooaq_pairs_steps_per_second": 2.526, "step": 179 }, { "epoch": 0.05, "eval_mrpc_pairs_loss": 0.055620357394218445, "eval_mrpc_pairs_runtime": 0.1048, "eval_mrpc_pairs_samples_per_second": 1431.291, "eval_mrpc_pairs_steps_per_second": 19.084, "step": 179 }, { "epoch": 0.05027932960893855, "grad_norm": 9.769979476928711, "learning_rate": 6.666666666666667e-06, "loss": 0.7777, "step": 180 }, { "epoch": 0.0553072625698324, "grad_norm": 9.294490814208984, "learning_rate": 7.337057728119181e-06, "loss": 0.9075, "step": 198 }, { "epoch": 0.060335195530726256, "grad_norm": 0.6179952621459961, "learning_rate": 8.007448789571696e-06, "loss": 0.8239, "step": 216 }, { "epoch": 0.06536312849162011, "grad_norm": 6.369142532348633, "learning_rate": 8.677839851024209e-06, "loss": 0.949, "step": 234 }, { "epoch": 0.07039106145251396, "grad_norm": 1.7066636085510254, "learning_rate": 9.348230912476724e-06, "loss": 1.2046, "step": 252 }, { "epoch": 0.07541899441340782, "grad_norm": 0.33892622590065, "learning_rate": 1.0018621973929237e-05, "loss": 0.8526, "step": 270 }, { "epoch": 0.08044692737430167, "grad_norm": 24.46626091003418, "learning_rate": 1.0689013035381753e-05, "loss": 0.8629, "step": 288 }, { "epoch": 0.08547486033519552, "grad_norm": 9.228001594543457, "learning_rate": 1.1359404096834266e-05, "loss": 1.1474, "step": 306 }, { "epoch": 0.09050279329608939, "grad_norm": 3.441566228866577, "learning_rate": 1.2029795158286779e-05, "loss": 0.9141, "step": 324 }, { "epoch": 0.09553072625698324, "grad_norm": 0.3621586859226227, "learning_rate": 1.2700186219739294e-05, "loss": 0.9756, "step": 342 }, { "epoch": 0.1, "eval_nli-pairs_loss": 0.9455356001853943, "eval_nli-pairs_runtime": 3.6115, "eval_nli-pairs_samples_per_second": 41.533, "eval_nli-pairs_steps_per_second": 0.554, "eval_sts-test_pearson_cosine": 0.7944563743777492, "eval_sts-test_pearson_dot": 0.554185693724103, "eval_sts-test_pearson_euclidean": 0.7483225064388481, "eval_sts-test_pearson_manhattan": 0.7454936340776467, "eval_sts-test_pearson_max": 0.7944563743777492, "eval_sts-test_spearman_cosine": 0.8064910811776218, "eval_sts-test_spearman_dot": 0.5333761592947532, "eval_sts-test_spearman_euclidean": 0.7382698586949932, "eval_sts-test_spearman_manhattan": 0.7378149008082869, "eval_sts-test_spearman_max": 0.8064910811776218, "step": 358 }, { "epoch": 0.1, "eval_vitaminc-pairs_loss": 4.517832279205322, "eval_vitaminc-pairs_runtime": 0.8908, "eval_vitaminc-pairs_samples_per_second": 140.33, "eval_vitaminc-pairs_steps_per_second": 2.245, "step": 358 }, { "epoch": 0.1, "eval_qnli-contrastive_loss": 0.2627922296524048, "eval_qnli-contrastive_runtime": 0.1229, "eval_qnli-contrastive_samples_per_second": 1220.886, "eval_qnli-contrastive_steps_per_second": 16.278, "step": 358 }, { "epoch": 0.1, "eval_scitail-pairs-qa_loss": 0.06381849944591522, "eval_scitail-pairs-qa_runtime": 0.7125, "eval_scitail-pairs-qa_samples_per_second": 210.516, "eval_scitail-pairs-qa_steps_per_second": 2.807, "step": 358 }, { "epoch": 0.1, "eval_scitail-pairs-pos_loss": 0.30350035429000854, "eval_scitail-pairs-pos_runtime": 1.877, "eval_scitail-pairs-pos_samples_per_second": 79.916, "eval_scitail-pairs-pos_steps_per_second": 1.066, "step": 358 }, { "epoch": 0.1, "eval_xsum-pairs_loss": 0.44631534814834595, "eval_xsum-pairs_runtime": 0.1512, "eval_xsum-pairs_samples_per_second": 542.377, "eval_xsum-pairs_steps_per_second": 6.614, "step": 358 }, { "epoch": 0.1, "eval_compression-pairs_loss": 0.10376789420843124, "eval_compression-pairs_runtime": 0.1034, "eval_compression-pairs_samples_per_second": 1450.105, "eval_compression-pairs_steps_per_second": 19.335, "step": 358 }, { "epoch": 0.1, "eval_sciq_pairs_loss": 0.15615364909172058, "eval_sciq_pairs_runtime": 5.1236, "eval_sciq_pairs_samples_per_second": 29.276, "eval_sciq_pairs_steps_per_second": 0.39, "step": 358 }, { "epoch": 0.1, "eval_qasc_pairs_loss": 0.24106402695178986, "eval_qasc_pairs_runtime": 0.6663, "eval_qasc_pairs_samples_per_second": 225.133, "eval_qasc_pairs_steps_per_second": 3.002, "step": 358 }, { "epoch": 0.1, "eval_openbookqa_pairs_loss": 1.5405189990997314, "eval_openbookqa_pairs_runtime": 0.7365, "eval_openbookqa_pairs_samples_per_second": 203.658, "eval_openbookqa_pairs_steps_per_second": 2.715, "step": 358 }, { "epoch": 0.1, "eval_msmarco_pairs_loss": 0.6067730188369751, "eval_msmarco_pairs_runtime": 1.2272, "eval_msmarco_pairs_samples_per_second": 122.233, "eval_msmarco_pairs_steps_per_second": 1.63, "step": 358 }, { "epoch": 0.1, "eval_nq_pairs_loss": 0.5683658123016357, "eval_nq_pairs_runtime": 2.3358, "eval_nq_pairs_samples_per_second": 64.219, "eval_nq_pairs_steps_per_second": 0.856, "step": 358 }, { "epoch": 0.1, "eval_trivia_pairs_loss": 1.1103042364120483, "eval_trivia_pairs_runtime": 3.2191, "eval_trivia_pairs_samples_per_second": 46.596, "eval_trivia_pairs_steps_per_second": 0.621, "step": 358 }, { "epoch": 0.1, "eval_quora_pairs_loss": 0.2291153520345688, "eval_quora_pairs_runtime": 0.39, "eval_quora_pairs_samples_per_second": 384.605, "eval_quora_pairs_steps_per_second": 5.128, "step": 358 }, { "epoch": 0.1, "eval_gooaq_pairs_loss": 0.5594914555549622, "eval_gooaq_pairs_runtime": 0.814, "eval_gooaq_pairs_samples_per_second": 184.276, "eval_gooaq_pairs_steps_per_second": 2.457, "step": 358 }, { "epoch": 0.1, "eval_mrpc_pairs_loss": 0.056830935180187225, "eval_mrpc_pairs_runtime": 0.1099, "eval_mrpc_pairs_samples_per_second": 1364.673, "eval_mrpc_pairs_steps_per_second": 18.196, "step": 358 }, { "epoch": 0.1005586592178771, "grad_norm": 0.40709275007247925, "learning_rate": 1.3370577281191808e-05, "loss": 0.7904, "step": 360 }, { "epoch": 0.10558659217877095, "grad_norm": 8.056321144104004, "learning_rate": 1.4040968342644321e-05, "loss": 1.2543, "step": 378 }, { "epoch": 0.1106145251396648, "grad_norm": 6.643378257751465, "learning_rate": 1.4711359404096834e-05, "loss": 0.854, "step": 396 }, { "epoch": 0.11564245810055866, "grad_norm": 7.473087310791016, "learning_rate": 1.538175046554935e-05, "loss": 0.4719, "step": 414 }, { "epoch": 0.12067039106145251, "grad_norm": 8.68940258026123, "learning_rate": 1.6052141527001864e-05, "loss": 1.4229, "step": 432 }, { "epoch": 0.12569832402234637, "grad_norm": 8.564693450927734, "learning_rate": 1.6722532588454377e-05, "loss": 0.7646, "step": 450 }, { "epoch": 0.13072625698324022, "grad_norm": 0.39765194058418274, "learning_rate": 1.7392923649906893e-05, "loss": 0.8512, "step": 468 }, { "epoch": 0.13575418994413407, "grad_norm": 7.3941192626953125, "learning_rate": 1.8063314711359406e-05, "loss": 1.1333, "step": 486 }, { "epoch": 0.14078212290502792, "grad_norm": 3.183729410171509, "learning_rate": 1.873370577281192e-05, "loss": 0.4451, "step": 504 }, { "epoch": 0.1458100558659218, "grad_norm": 1.2496144771575928, "learning_rate": 1.9404096834264436e-05, "loss": 0.8859, "step": 522 }, { "epoch": 0.15, "eval_nli-pairs_loss": 0.9459429979324341, "eval_nli-pairs_runtime": 3.418, "eval_nli-pairs_samples_per_second": 43.886, "eval_nli-pairs_steps_per_second": 0.585, "eval_sts-test_pearson_cosine": 0.7949855875446147, "eval_sts-test_pearson_dot": 0.5640384325880634, "eval_sts-test_pearson_euclidean": 0.751389277231376, "eval_sts-test_pearson_manhattan": 0.7487497192657817, "eval_sts-test_pearson_max": 0.7949855875446147, "eval_sts-test_spearman_cosine": 0.8064088099185094, "eval_sts-test_spearman_dot": 0.5419498621910633, "eval_sts-test_spearman_euclidean": 0.7411803691316964, "eval_sts-test_spearman_manhattan": 0.7408925147897409, "eval_sts-test_spearman_max": 0.8064088099185094, "step": 537 }, { "epoch": 0.15, "eval_vitaminc-pairs_loss": 4.557371139526367, "eval_vitaminc-pairs_runtime": 0.8813, "eval_vitaminc-pairs_samples_per_second": 141.843, "eval_vitaminc-pairs_steps_per_second": 2.269, "step": 537 }, { "epoch": 0.15, "eval_qnli-contrastive_loss": 0.2963089942932129, "eval_qnli-contrastive_runtime": 0.1173, "eval_qnli-contrastive_samples_per_second": 1279.24, "eval_qnli-contrastive_steps_per_second": 17.057, "step": 537 }, { "epoch": 0.15, "eval_scitail-pairs-qa_loss": 0.06569412350654602, "eval_scitail-pairs-qa_runtime": 0.6754, "eval_scitail-pairs-qa_samples_per_second": 222.091, "eval_scitail-pairs-qa_steps_per_second": 2.961, "step": 537 }, { "epoch": 0.15, "eval_scitail-pairs-pos_loss": 0.3104495108127594, "eval_scitail-pairs-pos_runtime": 1.7852, "eval_scitail-pairs-pos_samples_per_second": 84.022, "eval_scitail-pairs-pos_steps_per_second": 1.12, "step": 537 }, { "epoch": 0.15, "eval_xsum-pairs_loss": 0.44305890798568726, "eval_xsum-pairs_runtime": 0.1668, "eval_xsum-pairs_samples_per_second": 491.658, "eval_xsum-pairs_steps_per_second": 5.996, "step": 537 }, { "epoch": 0.15, "eval_compression-pairs_loss": 0.10611961036920547, "eval_compression-pairs_runtime": 0.1033, "eval_compression-pairs_samples_per_second": 1451.474, "eval_compression-pairs_steps_per_second": 19.353, "step": 537 }, { "epoch": 0.15, "eval_sciq_pairs_loss": 0.15942400693893433, "eval_sciq_pairs_runtime": 4.8457, "eval_sciq_pairs_samples_per_second": 30.955, "eval_sciq_pairs_steps_per_second": 0.413, "step": 537 }, { "epoch": 0.15, "eval_qasc_pairs_loss": 0.2380143702030182, "eval_qasc_pairs_runtime": 0.6528, "eval_qasc_pairs_samples_per_second": 229.78, "eval_qasc_pairs_steps_per_second": 3.064, "step": 537 }, { "epoch": 0.15, "eval_openbookqa_pairs_loss": 1.556572437286377, "eval_openbookqa_pairs_runtime": 0.6905, "eval_openbookqa_pairs_samples_per_second": 217.246, "eval_openbookqa_pairs_steps_per_second": 2.897, "step": 537 }, { "epoch": 0.15, "eval_msmarco_pairs_loss": 0.5950079560279846, "eval_msmarco_pairs_runtime": 1.1541, "eval_msmarco_pairs_samples_per_second": 129.968, "eval_msmarco_pairs_steps_per_second": 1.733, "step": 537 }, { "epoch": 0.15, "eval_nq_pairs_loss": 0.5976797938346863, "eval_nq_pairs_runtime": 2.2683, "eval_nq_pairs_samples_per_second": 66.128, "eval_nq_pairs_steps_per_second": 0.882, "step": 537 }, { "epoch": 0.15, "eval_trivia_pairs_loss": 1.1377041339874268, "eval_trivia_pairs_runtime": 3.1791, "eval_trivia_pairs_samples_per_second": 47.183, "eval_trivia_pairs_steps_per_second": 0.629, "step": 537 }, { "epoch": 0.15, "eval_quora_pairs_loss": 0.20836791396141052, "eval_quora_pairs_runtime": 0.374, "eval_quora_pairs_samples_per_second": 401.115, "eval_quora_pairs_steps_per_second": 5.348, "step": 537 }, { "epoch": 0.15, "eval_gooaq_pairs_loss": 0.5545207262039185, "eval_gooaq_pairs_runtime": 0.7667, "eval_gooaq_pairs_samples_per_second": 195.642, "eval_gooaq_pairs_steps_per_second": 2.609, "step": 537 }, { "epoch": 0.15, "eval_mrpc_pairs_loss": 0.060290463268756866, "eval_mrpc_pairs_runtime": 0.1026, "eval_mrpc_pairs_samples_per_second": 1462.106, "eval_mrpc_pairs_steps_per_second": 19.495, "step": 537 }, { "epoch": 0.15083798882681565, "grad_norm": 1.3315377235412598, "learning_rate": 1.9999898748390674e-05, "loss": 0.9245, "step": 540 }, { "epoch": 0.1558659217877095, "grad_norm": 8.015869140625, "learning_rate": 1.9989876718445097e-05, "loss": 1.0384, "step": 558 }, { "epoch": 0.16089385474860335, "grad_norm": 1.3641886711120605, "learning_rate": 1.9963472835306562e-05, "loss": 0.8826, "step": 576 }, { "epoch": 0.1659217877094972, "grad_norm": 7.676197528839111, "learning_rate": 1.9920735213598995e-05, "loss": 0.6693, "step": 594 }, { "epoch": 0.17094972067039105, "grad_norm": 8.144271850585938, "learning_rate": 1.9861741732192866e-05, "loss": 0.8437, "step": 612 }, { "epoch": 0.17597765363128492, "grad_norm": 11.723029136657715, "learning_rate": 1.9786599892290035e-05, "loss": 1.2241, "step": 630 }, { "epoch": 0.18100558659217877, "grad_norm": 12.107389450073242, "learning_rate": 1.9695446621529053e-05, "loss": 0.7169, "step": 648 }, { "epoch": 0.18603351955307262, "grad_norm": 0.671999454498291, "learning_rate": 1.9588448024468015e-05, "loss": 0.7937, "step": 666 }, { "epoch": 0.19106145251396647, "grad_norm": 2.9247829914093018, "learning_rate": 1.9465799079899608e-05, "loss": 0.8666, "step": 684 }, { "epoch": 0.19608938547486032, "grad_norm": 9.634854316711426, "learning_rate": 1.932772328554989e-05, "loss": 0.8002, "step": 702 }, { "epoch": 0.2, "eval_nli-pairs_loss": 0.9524073004722595, "eval_nli-pairs_runtime": 3.3652, "eval_nli-pairs_samples_per_second": 44.573, "eval_nli-pairs_steps_per_second": 0.594, "eval_sts-test_pearson_cosine": 0.7960974627777679, "eval_sts-test_pearson_dot": 0.5653927619143873, "eval_sts-test_pearson_euclidean": 0.7510324476182012, "eval_sts-test_pearson_manhattan": 0.7478581635465753, "eval_sts-test_pearson_max": 0.7960974627777679, "eval_sts-test_spearman_cosine": 0.8076217698074869, "eval_sts-test_spearman_dot": 0.5437275169754562, "eval_sts-test_spearman_euclidean": 0.7394612354069316, "eval_sts-test_spearman_manhattan": 0.7383580125266631, "eval_sts-test_spearman_max": 0.8076217698074869, "step": 716 }, { "epoch": 0.2, "eval_vitaminc-pairs_loss": 4.544824600219727, "eval_vitaminc-pairs_runtime": 0.8209, "eval_vitaminc-pairs_samples_per_second": 152.27, "eval_vitaminc-pairs_steps_per_second": 2.436, "step": 716 }, { "epoch": 0.2, "eval_qnli-contrastive_loss": 0.2854141891002655, "eval_qnli-contrastive_runtime": 0.116, "eval_qnli-contrastive_samples_per_second": 1293.335, "eval_qnli-contrastive_steps_per_second": 17.244, "step": 716 }, { "epoch": 0.2, "eval_scitail-pairs-qa_loss": 0.06479831039905548, "eval_scitail-pairs-qa_runtime": 0.6668, "eval_scitail-pairs-qa_samples_per_second": 224.963, "eval_scitail-pairs-qa_steps_per_second": 3.0, "step": 716 }, { "epoch": 0.2, "eval_scitail-pairs-pos_loss": 0.30356064438819885, "eval_scitail-pairs-pos_runtime": 1.7895, "eval_scitail-pairs-pos_samples_per_second": 83.823, "eval_scitail-pairs-pos_steps_per_second": 1.118, "step": 716 }, { "epoch": 0.2, "eval_xsum-pairs_loss": 0.4510384202003479, "eval_xsum-pairs_runtime": 0.1513, "eval_xsum-pairs_samples_per_second": 541.869, "eval_xsum-pairs_steps_per_second": 6.608, "step": 716 }, { "epoch": 0.2, "eval_compression-pairs_loss": 0.10218393057584763, "eval_compression-pairs_runtime": 0.1063, "eval_compression-pairs_samples_per_second": 1410.495, "eval_compression-pairs_steps_per_second": 18.807, "step": 716 }, { "epoch": 0.2, "eval_sciq_pairs_loss": 0.16117030382156372, "eval_sciq_pairs_runtime": 4.8842, "eval_sciq_pairs_samples_per_second": 30.711, "eval_sciq_pairs_steps_per_second": 0.409, "step": 716 }, { "epoch": 0.2, "eval_qasc_pairs_loss": 0.25312450528144836, "eval_qasc_pairs_runtime": 0.626, "eval_qasc_pairs_samples_per_second": 239.619, "eval_qasc_pairs_steps_per_second": 3.195, "step": 716 }, { "epoch": 0.2, "eval_openbookqa_pairs_loss": 1.560429573059082, "eval_openbookqa_pairs_runtime": 0.6856, "eval_openbookqa_pairs_samples_per_second": 218.787, "eval_openbookqa_pairs_steps_per_second": 2.917, "step": 716 }, { "epoch": 0.2, "eval_msmarco_pairs_loss": 0.6264744997024536, "eval_msmarco_pairs_runtime": 1.1496, "eval_msmarco_pairs_samples_per_second": 130.479, "eval_msmarco_pairs_steps_per_second": 1.74, "step": 716 }, { "epoch": 0.2, "eval_nq_pairs_loss": 0.602968692779541, "eval_nq_pairs_runtime": 2.2714, "eval_nq_pairs_samples_per_second": 66.038, "eval_nq_pairs_steps_per_second": 0.881, "step": 716 }, { "epoch": 0.2, "eval_trivia_pairs_loss": 1.1763343811035156, "eval_trivia_pairs_runtime": 3.1779, "eval_trivia_pairs_samples_per_second": 47.201, "eval_trivia_pairs_steps_per_second": 0.629, "step": 716 }, { "epoch": 0.2, "eval_quora_pairs_loss": 0.25559327006340027, "eval_quora_pairs_runtime": 0.3926, "eval_quora_pairs_samples_per_second": 382.056, "eval_quora_pairs_steps_per_second": 5.094, "step": 716 }, { "epoch": 0.2, "eval_gooaq_pairs_loss": 0.5762031078338623, "eval_gooaq_pairs_runtime": 0.7962, "eval_gooaq_pairs_samples_per_second": 188.399, "eval_gooaq_pairs_steps_per_second": 2.512, "step": 716 }, { "epoch": 0.2, "eval_mrpc_pairs_loss": 0.05901935696601868, "eval_mrpc_pairs_runtime": 0.1022, "eval_mrpc_pairs_samples_per_second": 1467.265, "eval_mrpc_pairs_steps_per_second": 19.564, "step": 716 }, { "epoch": 0.2011173184357542, "grad_norm": 2.0602877140045166, "learning_rate": 1.9174472250808337e-05, "loss": 0.9815, "step": 720 }, { "epoch": 0.20614525139664805, "grad_norm": 10.161604881286621, "learning_rate": 1.90063252382312e-05, "loss": 0.568, "step": 738 }, { "epoch": 0.2111731843575419, "grad_norm": 18.392881393432617, "learning_rate": 1.882358865465376e-05, "loss": 0.8045, "step": 756 }, { "epoch": 0.21620111731843575, "grad_norm": 1.8240203857421875, "learning_rate": 1.8626595492838756e-05, "loss": 0.9535, "step": 774 }, { "epoch": 0.2212290502793296, "grad_norm": 0.7226653099060059, "learning_rate": 1.8415704724678457e-05, "loss": 0.8132, "step": 792 }, { "epoch": 0.22625698324022347, "grad_norm": 5.971342086791992, "learning_rate": 1.819130064705612e-05, "loss": 0.6948, "step": 810 }, { "epoch": 0.23128491620111732, "grad_norm": 8.432775497436523, "learning_rate": 1.795379218155883e-05, "loss": 1.2272, "step": 828 }, { "epoch": 0.23631284916201117, "grad_norm": 18.75815200805664, "learning_rate": 1.7703612129317886e-05, "loss": 0.9203, "step": 846 }, { "epoch": 0.24134078212290502, "grad_norm": 0.6807297468185425, "learning_rate": 1.7441216382334472e-05, "loss": 0.6491, "step": 864 }, { "epoch": 0.24636871508379887, "grad_norm": 0.27311134338378906, "learning_rate": 1.716708309272793e-05, "loss": 0.7194, "step": 882 }, { "epoch": 0.25, "eval_nli-pairs_loss": 0.9641673564910889, "eval_nli-pairs_runtime": 3.4016, "eval_nli-pairs_samples_per_second": 44.097, "eval_nli-pairs_steps_per_second": 0.588, "eval_sts-test_pearson_cosine": 0.7984203994030687, "eval_sts-test_pearson_dot": 0.5718132073426122, "eval_sts-test_pearson_euclidean": 0.751738033448039, "eval_sts-test_pearson_manhattan": 0.7480386170862788, "eval_sts-test_pearson_max": 0.7984203994030687, "eval_sts-test_spearman_cosine": 0.8106824393227386, "eval_sts-test_spearman_dot": 0.5505203809846263, "eval_sts-test_spearman_euclidean": 0.7424109996455601, "eval_sts-test_spearman_manhattan": 0.740994183877676, "eval_sts-test_spearman_max": 0.8106824393227386, "step": 895 }, { "epoch": 0.25, "eval_vitaminc-pairs_loss": 4.445808410644531, "eval_vitaminc-pairs_runtime": 0.8341, "eval_vitaminc-pairs_samples_per_second": 149.86, "eval_vitaminc-pairs_steps_per_second": 2.398, "step": 895 }, { "epoch": 0.25, "eval_qnli-contrastive_loss": 0.26446962356567383, "eval_qnli-contrastive_runtime": 0.1192, "eval_qnli-contrastive_samples_per_second": 1258.221, "eval_qnli-contrastive_steps_per_second": 16.776, "step": 895 }, { "epoch": 0.25, "eval_scitail-pairs-qa_loss": 0.06552065908908844, "eval_scitail-pairs-qa_runtime": 0.685, "eval_scitail-pairs-qa_samples_per_second": 218.99, "eval_scitail-pairs-qa_steps_per_second": 2.92, "step": 895 }, { "epoch": 0.25, "eval_scitail-pairs-pos_loss": 0.2940276265144348, "eval_scitail-pairs-pos_runtime": 1.7906, "eval_scitail-pairs-pos_samples_per_second": 83.773, "eval_scitail-pairs-pos_steps_per_second": 1.117, "step": 895 }, { "epoch": 0.25, "eval_xsum-pairs_loss": 0.46034616231918335, "eval_xsum-pairs_runtime": 0.1605, "eval_xsum-pairs_samples_per_second": 510.946, "eval_xsum-pairs_steps_per_second": 6.231, "step": 895 }, { "epoch": 0.25, "eval_compression-pairs_loss": 0.10285492241382599, "eval_compression-pairs_runtime": 0.1054, "eval_compression-pairs_samples_per_second": 1423.451, "eval_compression-pairs_steps_per_second": 18.979, "step": 895 }, { "epoch": 0.25, "eval_sciq_pairs_loss": 0.16273371875286102, "eval_sciq_pairs_runtime": 4.8468, "eval_sciq_pairs_samples_per_second": 30.948, "eval_sciq_pairs_steps_per_second": 0.413, "step": 895 }, { "epoch": 0.25, "eval_qasc_pairs_loss": 0.24400106072425842, "eval_qasc_pairs_runtime": 0.6336, "eval_qasc_pairs_samples_per_second": 236.76, "eval_qasc_pairs_steps_per_second": 3.157, "step": 895 }, { "epoch": 0.25, "eval_openbookqa_pairs_loss": 1.5389502048492432, "eval_openbookqa_pairs_runtime": 0.6864, "eval_openbookqa_pairs_samples_per_second": 218.536, "eval_openbookqa_pairs_steps_per_second": 2.914, "step": 895 }, { "epoch": 0.25, "eval_msmarco_pairs_loss": 0.6092618703842163, "eval_msmarco_pairs_runtime": 1.1561, "eval_msmarco_pairs_samples_per_second": 129.746, "eval_msmarco_pairs_steps_per_second": 1.73, "step": 895 }, { "epoch": 0.25, "eval_nq_pairs_loss": 0.5972890853881836, "eval_nq_pairs_runtime": 2.2849, "eval_nq_pairs_samples_per_second": 65.648, "eval_nq_pairs_steps_per_second": 0.875, "step": 895 }, { "epoch": 0.25, "eval_trivia_pairs_loss": 1.1646121740341187, "eval_trivia_pairs_runtime": 3.1713, "eval_trivia_pairs_samples_per_second": 47.299, "eval_trivia_pairs_steps_per_second": 0.631, "step": 895 }, { "epoch": 0.25, "eval_quora_pairs_loss": 0.2002115696668625, "eval_quora_pairs_runtime": 0.3724, "eval_quora_pairs_samples_per_second": 402.785, "eval_quora_pairs_steps_per_second": 5.37, "step": 895 }, { "epoch": 0.25, "eval_gooaq_pairs_loss": 0.5770832300186157, "eval_gooaq_pairs_runtime": 0.7637, "eval_gooaq_pairs_samples_per_second": 196.419, "eval_gooaq_pairs_steps_per_second": 2.619, "step": 895 }, { "epoch": 0.25, "eval_mrpc_pairs_loss": 0.060786984860897064, "eval_mrpc_pairs_runtime": 0.1004, "eval_mrpc_pairs_samples_per_second": 1493.964, "eval_mrpc_pairs_steps_per_second": 19.92, "step": 895 }, { "epoch": 0.25139664804469275, "grad_norm": 9.622016906738281, "learning_rate": 1.6881711801420374e-05, "loss": 0.9089, "step": 900 }, { "epoch": 0.25642458100558657, "grad_norm": 3.134366035461426, "learning_rate": 1.658562252784545e-05, "loss": 0.9682, "step": 918 }, { "epoch": 0.26145251396648045, "grad_norm": 11.332123756408691, "learning_rate": 1.627935482234001e-05, "loss": 1.0025, "step": 936 }, { "epoch": 0.2664804469273743, "grad_norm": 10.678857803344727, "learning_rate": 1.5963466782945497e-05, "loss": 1.0259, "step": 954 }, { "epoch": 0.27150837988826815, "grad_norm": 0.8130522966384888, "learning_rate": 1.5638534038410646e-05, "loss": 0.8946, "step": 972 }, { "epoch": 0.276536312849162, "grad_norm": 25.857967376708984, "learning_rate": 1.5305148699248803e-05, "loss": 0.72, "step": 990 }, { "epoch": 0.28156424581005585, "grad_norm": 8.641229629516602, "learning_rate": 1.4963918278761172e-05, "loss": 0.8911, "step": 1008 }, { "epoch": 0.2865921787709497, "grad_norm": 5.093626499176025, "learning_rate": 1.4615464585992371e-05, "loss": 0.8059, "step": 1026 }, { "epoch": 0.2916201117318436, "grad_norm": 0.18065588176250458, "learning_rate": 1.4260422592635362e-05, "loss": 1.0758, "step": 1044 }, { "epoch": 0.2966480446927374, "grad_norm": 10.29207706451416, "learning_rate": 1.3899439275950759e-05, "loss": 0.8875, "step": 1062 }, { "epoch": 0.3, "eval_nli-pairs_loss": 0.9130206108093262, "eval_nli-pairs_runtime": 3.3804, "eval_nli-pairs_samples_per_second": 44.373, "eval_nli-pairs_steps_per_second": 0.592, "eval_sts-test_pearson_cosine": 0.7935564606718174, "eval_sts-test_pearson_dot": 0.5781608636588449, "eval_sts-test_pearson_euclidean": 0.7502295476446079, "eval_sts-test_pearson_manhattan": 0.746737096593109, "eval_sts-test_pearson_max": 0.7935564606718174, "eval_sts-test_spearman_cosine": 0.8086930151421495, "eval_sts-test_spearman_dot": 0.5549031403563933, "eval_sts-test_spearman_euclidean": 0.739389609130169, "eval_sts-test_spearman_manhattan": 0.7380312227930571, "eval_sts-test_spearman_max": 0.8086930151421495, "step": 1074 }, { "epoch": 0.3, "eval_vitaminc-pairs_loss": 4.481905937194824, "eval_vitaminc-pairs_runtime": 0.8115, "eval_vitaminc-pairs_samples_per_second": 154.041, "eval_vitaminc-pairs_steps_per_second": 2.465, "step": 1074 }, { "epoch": 0.3, "eval_qnli-contrastive_loss": 0.271823525428772, "eval_qnli-contrastive_runtime": 0.1166, "eval_qnli-contrastive_samples_per_second": 1286.546, "eval_qnli-contrastive_steps_per_second": 17.154, "step": 1074 }, { "epoch": 0.3, "eval_scitail-pairs-qa_loss": 0.06460323184728622, "eval_scitail-pairs-qa_runtime": 0.6655, "eval_scitail-pairs-qa_samples_per_second": 225.379, "eval_scitail-pairs-qa_steps_per_second": 3.005, "step": 1074 }, { "epoch": 0.3, "eval_scitail-pairs-pos_loss": 0.28618788719177246, "eval_scitail-pairs-pos_runtime": 1.7829, "eval_scitail-pairs-pos_samples_per_second": 84.131, "eval_scitail-pairs-pos_steps_per_second": 1.122, "step": 1074 }, { "epoch": 0.3, "eval_xsum-pairs_loss": 0.46461817622184753, "eval_xsum-pairs_runtime": 0.1506, "eval_xsum-pairs_samples_per_second": 544.581, "eval_xsum-pairs_steps_per_second": 6.641, "step": 1074 }, { "epoch": 0.3, "eval_compression-pairs_loss": 0.0993683859705925, "eval_compression-pairs_runtime": 0.1041, "eval_compression-pairs_samples_per_second": 1440.48, "eval_compression-pairs_steps_per_second": 19.206, "step": 1074 }, { "epoch": 0.3, "eval_sciq_pairs_loss": 0.16428890824317932, "eval_sciq_pairs_runtime": 4.9141, "eval_sciq_pairs_samples_per_second": 30.524, "eval_sciq_pairs_steps_per_second": 0.407, "step": 1074 }, { "epoch": 0.3, "eval_qasc_pairs_loss": 0.2515706419944763, "eval_qasc_pairs_runtime": 0.6397, "eval_qasc_pairs_samples_per_second": 234.494, "eval_qasc_pairs_steps_per_second": 3.127, "step": 1074 }, { "epoch": 0.3, "eval_openbookqa_pairs_loss": 1.5371934175491333, "eval_openbookqa_pairs_runtime": 0.6887, "eval_openbookqa_pairs_samples_per_second": 217.793, "eval_openbookqa_pairs_steps_per_second": 2.904, "step": 1074 }, { "epoch": 0.3, "eval_msmarco_pairs_loss": 0.6019803881645203, "eval_msmarco_pairs_runtime": 1.1582, "eval_msmarco_pairs_samples_per_second": 129.515, "eval_msmarco_pairs_steps_per_second": 1.727, "step": 1074 }, { "epoch": 0.3, "eval_nq_pairs_loss": 0.6118470430374146, "eval_nq_pairs_runtime": 2.2708, "eval_nq_pairs_samples_per_second": 66.056, "eval_nq_pairs_steps_per_second": 0.881, "step": 1074 }, { "epoch": 0.3, "eval_trivia_pairs_loss": 1.12074613571167, "eval_trivia_pairs_runtime": 3.1611, "eval_trivia_pairs_samples_per_second": 47.452, "eval_trivia_pairs_steps_per_second": 0.633, "step": 1074 }, { "epoch": 0.3, "eval_quora_pairs_loss": 0.22596687078475952, "eval_quora_pairs_runtime": 0.3879, "eval_quora_pairs_samples_per_second": 386.741, "eval_quora_pairs_steps_per_second": 5.157, "step": 1074 }, { "epoch": 0.3, "eval_gooaq_pairs_loss": 0.5833558440208435, "eval_gooaq_pairs_runtime": 0.7697, "eval_gooaq_pairs_samples_per_second": 194.881, "eval_gooaq_pairs_steps_per_second": 2.598, "step": 1074 }, { "epoch": 0.3, "eval_mrpc_pairs_loss": 0.058091748505830765, "eval_mrpc_pairs_runtime": 0.1053, "eval_mrpc_pairs_samples_per_second": 1424.202, "eval_mrpc_pairs_steps_per_second": 18.989, "step": 1074 }, { "epoch": 0.3016759776536313, "grad_norm": 3.0486669540405273, "learning_rate": 1.3533172439808864e-05, "loss": 1.1998, "step": 1080 }, { "epoch": 0.3067039106145251, "grad_norm": 9.376039505004883, "learning_rate": 1.3162289516002921e-05, "loss": 0.7894, "step": 1098 }, { "epoch": 0.311731843575419, "grad_norm": 2.824523448944092, "learning_rate": 1.2787466348017752e-05, "loss": 1.1598, "step": 1116 }, { "epoch": 0.3167597765363129, "grad_norm": 13.443314552307129, "learning_rate": 1.2430464185616653e-05, "loss": 1.0983, "step": 1134 }, { "epoch": 0.3217877094972067, "grad_norm": 7.764914035797119, "learning_rate": 1.2049940057022904e-05, "loss": 0.8006, "step": 1152 }, { "epoch": 0.3268156424581006, "grad_norm": 7.54190731048584, "learning_rate": 1.1667502669280646e-05, "loss": 0.6122, "step": 1170 }, { "epoch": 0.3318435754189944, "grad_norm": 9.039295196533203, "learning_rate": 1.1283848921061318e-05, "loss": 0.5932, "step": 1188 }, { "epoch": 0.3368715083798883, "grad_norm": 7.766181468963623, "learning_rate": 1.0899677927556045e-05, "loss": 0.6522, "step": 1206 }, { "epoch": 0.3418994413407821, "grad_norm": 17.672466278076172, "learning_rate": 1.0515689746509086e-05, "loss": 1.0307, "step": 1224 }, { "epoch": 0.346927374301676, "grad_norm": 8.456610679626465, "learning_rate": 1.0132584102533657e-05, "loss": 1.0892, "step": 1242 }, { "epoch": 0.35, "eval_nli-pairs_loss": 0.9248345494270325, "eval_nli-pairs_runtime": 3.3923, "eval_nli-pairs_samples_per_second": 44.218, "eval_nli-pairs_steps_per_second": 0.59, "eval_sts-test_pearson_cosine": 0.7948484764216447, "eval_sts-test_pearson_dot": 0.5671123561184989, "eval_sts-test_pearson_euclidean": 0.7505417118786587, "eval_sts-test_pearson_manhattan": 0.7471600903913781, "eval_sts-test_pearson_max": 0.7948484764216447, "eval_sts-test_spearman_cosine": 0.8098066598796416, "eval_sts-test_spearman_dot": 0.5445125288737649, "eval_sts-test_spearman_euclidean": 0.7404259515118012, "eval_sts-test_spearman_manhattan": 0.7397832260792029, "eval_sts-test_spearman_max": 0.8098066598796416, "step": 1253 }, { "epoch": 0.35, "eval_vitaminc-pairs_loss": 4.364264965057373, "eval_vitaminc-pairs_runtime": 0.8126, "eval_vitaminc-pairs_samples_per_second": 153.831, "eval_vitaminc-pairs_steps_per_second": 2.461, "step": 1253 }, { "epoch": 0.35, "eval_qnli-contrastive_loss": 0.2623118460178375, "eval_qnli-contrastive_runtime": 0.1187, "eval_qnli-contrastive_samples_per_second": 1264.134, "eval_qnli-contrastive_steps_per_second": 16.855, "step": 1253 }, { "epoch": 0.35, "eval_scitail-pairs-qa_loss": 0.0632135197520256, "eval_scitail-pairs-qa_runtime": 0.6764, "eval_scitail-pairs-qa_samples_per_second": 221.754, "eval_scitail-pairs-qa_steps_per_second": 2.957, "step": 1253 }, { "epoch": 0.35, "eval_scitail-pairs-pos_loss": 0.2840258777141571, "eval_scitail-pairs-pos_runtime": 1.7876, "eval_scitail-pairs-pos_samples_per_second": 83.913, "eval_scitail-pairs-pos_steps_per_second": 1.119, "step": 1253 }, { "epoch": 0.35, "eval_xsum-pairs_loss": 0.45192036032676697, "eval_xsum-pairs_runtime": 0.1525, "eval_xsum-pairs_samples_per_second": 537.578, "eval_xsum-pairs_steps_per_second": 6.556, "step": 1253 }, { "epoch": 0.35, "eval_compression-pairs_loss": 0.09640318900346756, "eval_compression-pairs_runtime": 0.1083, "eval_compression-pairs_samples_per_second": 1385.341, "eval_compression-pairs_steps_per_second": 18.471, "step": 1253 }, { "epoch": 0.35, "eval_sciq_pairs_loss": 0.16294126212596893, "eval_sciq_pairs_runtime": 4.9672, "eval_sciq_pairs_samples_per_second": 30.198, "eval_sciq_pairs_steps_per_second": 0.403, "step": 1253 }, { "epoch": 0.35, "eval_qasc_pairs_loss": 0.24691826105117798, "eval_qasc_pairs_runtime": 0.6342, "eval_qasc_pairs_samples_per_second": 236.512, "eval_qasc_pairs_steps_per_second": 3.153, "step": 1253 }, { "epoch": 0.35, "eval_openbookqa_pairs_loss": 1.5197923183441162, "eval_openbookqa_pairs_runtime": 0.6915, "eval_openbookqa_pairs_samples_per_second": 216.924, "eval_openbookqa_pairs_steps_per_second": 2.892, "step": 1253 }, { "epoch": 0.35, "eval_msmarco_pairs_loss": 0.5905637145042419, "eval_msmarco_pairs_runtime": 1.1595, "eval_msmarco_pairs_samples_per_second": 129.367, "eval_msmarco_pairs_steps_per_second": 1.725, "step": 1253 }, { "epoch": 0.35, "eval_nq_pairs_loss": 0.5736532211303711, "eval_nq_pairs_runtime": 2.2864, "eval_nq_pairs_samples_per_second": 65.606, "eval_nq_pairs_steps_per_second": 0.875, "step": 1253 }, { "epoch": 0.35, "eval_trivia_pairs_loss": 1.1131163835525513, "eval_trivia_pairs_runtime": 3.1686, "eval_trivia_pairs_samples_per_second": 47.34, "eval_trivia_pairs_steps_per_second": 0.631, "step": 1253 }, { "epoch": 0.35, "eval_quora_pairs_loss": 0.2874162197113037, "eval_quora_pairs_runtime": 0.4087, "eval_quora_pairs_samples_per_second": 366.989, "eval_quora_pairs_steps_per_second": 4.893, "step": 1253 }, { "epoch": 0.35, "eval_gooaq_pairs_loss": 0.5454840660095215, "eval_gooaq_pairs_runtime": 0.7785, "eval_gooaq_pairs_samples_per_second": 192.683, "eval_gooaq_pairs_steps_per_second": 2.569, "step": 1253 }, { "epoch": 0.35, "eval_mrpc_pairs_loss": 0.05724556744098663, "eval_mrpc_pairs_runtime": 0.1564, "eval_mrpc_pairs_samples_per_second": 958.775, "eval_mrpc_pairs_steps_per_second": 12.784, "step": 1253 }, { "epoch": 0.35195530726256985, "grad_norm": 9.2592191696167, "learning_rate": 9.75105911203486e-06, "loss": 1.0845, "step": 1260 }, { "epoch": 0.35698324022346367, "grad_norm": 8.712442398071289, "learning_rate": 9.371810011063151e-06, "loss": 0.931, "step": 1278 }, { "epoch": 0.36201117318435755, "grad_norm": 12.440937995910645, "learning_rate": 8.995527888416549e-06, "loss": 0.6159, "step": 1296 }, { "epoch": 0.36703910614525137, "grad_norm": 7.335140705108643, "learning_rate": 8.6228984263002e-06, "loss": 0.6498, "step": 1314 }, { "epoch": 0.37206703910614525, "grad_norm": 9.586339950561523, "learning_rate": 8.254600650838152e-06, "loss": 0.9762, "step": 1332 }, { "epoch": 0.3770949720670391, "grad_norm": 5.655767917633057, "learning_rate": 7.891305694714182e-06, "loss": 0.7534, "step": 1350 }, { "epoch": 0.38212290502793295, "grad_norm": 11.050688743591309, "learning_rate": 7.533675574196519e-06, "loss": 0.6792, "step": 1368 }, { "epoch": 0.3871508379888268, "grad_norm": 8.111286163330078, "learning_rate": 7.182361982774984e-06, "loss": 0.6524, "step": 1386 }, { "epoch": 0.39217877094972065, "grad_norm": 0.9727733135223389, "learning_rate": 6.8380051036088735e-06, "loss": 0.8079, "step": 1404 }, { "epoch": 0.3972067039106145, "grad_norm": 7.61720609664917, "learning_rate": 6.501232442949635e-06, "loss": 1.2497, "step": 1422 }, { "epoch": 0.4, "eval_nli-pairs_loss": 0.9316923022270203, "eval_nli-pairs_runtime": 3.6665, "eval_nli-pairs_samples_per_second": 40.911, "eval_nli-pairs_steps_per_second": 0.545, "eval_sts-test_pearson_cosine": 0.7951706194225081, "eval_sts-test_pearson_dot": 0.5655762874531658, "eval_sts-test_pearson_euclidean": 0.748417242742893, "eval_sts-test_pearson_manhattan": 0.745117038641416, "eval_sts-test_pearson_max": 0.7951706194225081, "eval_sts-test_spearman_cosine": 0.808265389218048, "eval_sts-test_spearman_dot": 0.5439374661732892, "eval_sts-test_spearman_euclidean": 0.737979067648202, "eval_sts-test_spearman_manhattan": 0.7367152789420606, "eval_sts-test_spearman_max": 0.808265389218048, "step": 1432 }, { "epoch": 0.4, "eval_vitaminc-pairs_loss": 4.391602993011475, "eval_vitaminc-pairs_runtime": 0.8561, "eval_vitaminc-pairs_samples_per_second": 146.012, "eval_vitaminc-pairs_steps_per_second": 2.336, "step": 1432 }, { "epoch": 0.4, "eval_qnli-contrastive_loss": 0.2666786313056946, "eval_qnli-contrastive_runtime": 0.1247, "eval_qnli-contrastive_samples_per_second": 1202.967, "eval_qnli-contrastive_steps_per_second": 16.04, "step": 1432 }, { "epoch": 0.4, "eval_scitail-pairs-qa_loss": 0.06357826292514801, "eval_scitail-pairs-qa_runtime": 0.7031, "eval_scitail-pairs-qa_samples_per_second": 213.329, "eval_scitail-pairs-qa_steps_per_second": 2.844, "step": 1432 }, { "epoch": 0.4, "eval_scitail-pairs-pos_loss": 0.2862562835216522, "eval_scitail-pairs-pos_runtime": 1.9248, "eval_scitail-pairs-pos_samples_per_second": 77.929, "eval_scitail-pairs-pos_steps_per_second": 1.039, "step": 1432 }, { "epoch": 0.4, "eval_xsum-pairs_loss": 0.4477728009223938, "eval_xsum-pairs_runtime": 0.1563, "eval_xsum-pairs_samples_per_second": 524.667, "eval_xsum-pairs_steps_per_second": 6.398, "step": 1432 }, { "epoch": 0.4, "eval_compression-pairs_loss": 0.09554792195558548, "eval_compression-pairs_runtime": 0.1113, "eval_compression-pairs_samples_per_second": 1348.03, "eval_compression-pairs_steps_per_second": 17.974, "step": 1432 }, { "epoch": 0.4, "eval_sciq_pairs_loss": 0.16296182572841644, "eval_sciq_pairs_runtime": 4.8695, "eval_sciq_pairs_samples_per_second": 30.804, "eval_sciq_pairs_steps_per_second": 0.411, "step": 1432 }, { "epoch": 0.4, "eval_qasc_pairs_loss": 0.2451217621564865, "eval_qasc_pairs_runtime": 0.6585, "eval_qasc_pairs_samples_per_second": 227.783, "eval_qasc_pairs_steps_per_second": 3.037, "step": 1432 }, { "epoch": 0.4, "eval_openbookqa_pairs_loss": 1.5293627977371216, "eval_openbookqa_pairs_runtime": 0.6922, "eval_openbookqa_pairs_samples_per_second": 216.714, "eval_openbookqa_pairs_steps_per_second": 2.89, "step": 1432 }, { "epoch": 0.4, "eval_msmarco_pairs_loss": 0.5807374119758606, "eval_msmarco_pairs_runtime": 1.1603, "eval_msmarco_pairs_samples_per_second": 129.282, "eval_msmarco_pairs_steps_per_second": 1.724, "step": 1432 }, { "epoch": 0.4, "eval_nq_pairs_loss": 0.5674509406089783, "eval_nq_pairs_runtime": 2.2725, "eval_nq_pairs_samples_per_second": 66.007, "eval_nq_pairs_steps_per_second": 0.88, "step": 1432 }, { "epoch": 0.4, "eval_trivia_pairs_loss": 1.0616999864578247, "eval_trivia_pairs_runtime": 3.1673, "eval_trivia_pairs_samples_per_second": 47.359, "eval_trivia_pairs_steps_per_second": 0.631, "step": 1432 }, { "epoch": 0.4, "eval_quora_pairs_loss": 0.18273381888866425, "eval_quora_pairs_runtime": 0.384, "eval_quora_pairs_samples_per_second": 390.615, "eval_quora_pairs_steps_per_second": 5.208, "step": 1432 }, { "epoch": 0.4, "eval_gooaq_pairs_loss": 0.5485998392105103, "eval_gooaq_pairs_runtime": 0.7712, "eval_gooaq_pairs_samples_per_second": 194.508, "eval_gooaq_pairs_steps_per_second": 2.593, "step": 1432 }, { "epoch": 0.4, "eval_mrpc_pairs_loss": 0.05663560330867767, "eval_mrpc_pairs_runtime": 0.1016, "eval_mrpc_pairs_samples_per_second": 1475.752, "eval_mrpc_pairs_steps_per_second": 19.677, "step": 1432 }, { "epoch": 0.4022346368715084, "grad_norm": 9.786046028137207, "learning_rate": 6.172657686664055e-06, "loss": 0.8498, "step": 1440 }, { "epoch": 0.4072625698324022, "grad_norm": 0.5503944158554077, "learning_rate": 5.852879581941775e-06, "loss": 0.5437, "step": 1458 }, { "epoch": 0.4122905027932961, "grad_norm": 7.578904151916504, "learning_rate": 5.542480846224858e-06, "loss": 1.0774, "step": 1476 }, { "epoch": 0.4173184357541899, "grad_norm": 10.365352630615234, "learning_rate": 5.2420271053476904e-06, "loss": 0.9975, "step": 1494 }, { "epoch": 0.4223463687150838, "grad_norm": 0.6473840475082397, "learning_rate": 4.9520658628221e-06, "loss": 0.9214, "step": 1512 }, { "epoch": 0.4273743016759777, "grad_norm": 10.450447082519531, "learning_rate": 4.6731255021460435e-06, "loss": 0.8964, "step": 1530 }, { "epoch": 0.4324022346368715, "grad_norm": 9.311734199523926, "learning_rate": 4.405714323953814e-06, "loss": 0.669, "step": 1548 }, { "epoch": 0.4374301675977654, "grad_norm": 36.57959747314453, "learning_rate": 4.150319619762361e-06, "loss": 1.1688, "step": 1566 }, { "epoch": 0.4424581005586592, "grad_norm": 15.4668607711792, "learning_rate": 3.907406784001587e-06, "loss": 0.5588, "step": 1584 }, { "epoch": 0.4474860335195531, "grad_norm": 11.20753288269043, "learning_rate": 3.6774184659467597e-06, "loss": 0.8192, "step": 1602 }, { "epoch": 0.45, "eval_nli-pairs_loss": 0.9224842190742493, "eval_nli-pairs_runtime": 3.4345, "eval_nli-pairs_samples_per_second": 43.675, "eval_nli-pairs_steps_per_second": 0.582, "eval_sts-test_pearson_cosine": 0.7947548033548021, "eval_sts-test_pearson_dot": 0.5642970712925082, "eval_sts-test_pearson_euclidean": 0.7480023824232991, "eval_sts-test_pearson_manhattan": 0.7449444544911363, "eval_sts-test_pearson_max": 0.7947548033548021, "eval_sts-test_spearman_cosine": 0.8077938075661354, "eval_sts-test_spearman_dot": 0.5418922459663886, "eval_sts-test_spearman_euclidean": 0.7372467308885063, "eval_sts-test_spearman_manhattan": 0.736073760123024, "eval_sts-test_spearman_max": 0.8077938075661354, "step": 1611 }, { "epoch": 0.45, "eval_vitaminc-pairs_loss": 4.448488235473633, "eval_vitaminc-pairs_runtime": 0.8352, "eval_vitaminc-pairs_samples_per_second": 149.657, "eval_vitaminc-pairs_steps_per_second": 2.395, "step": 1611 }, { "epoch": 0.45, "eval_qnli-contrastive_loss": 0.27063658833503723, "eval_qnli-contrastive_runtime": 0.1172, "eval_qnli-contrastive_samples_per_second": 1280.373, "eval_qnli-contrastive_steps_per_second": 17.072, "step": 1611 }, { "epoch": 0.45, "eval_scitail-pairs-qa_loss": 0.06333789229393005, "eval_scitail-pairs-qa_runtime": 0.6731, "eval_scitail-pairs-qa_samples_per_second": 222.838, "eval_scitail-pairs-qa_steps_per_second": 2.971, "step": 1611 }, { "epoch": 0.45, "eval_scitail-pairs-pos_loss": 0.2783344089984894, "eval_scitail-pairs-pos_runtime": 1.773, "eval_scitail-pairs-pos_samples_per_second": 84.601, "eval_scitail-pairs-pos_steps_per_second": 1.128, "step": 1611 }, { "epoch": 0.45, "eval_xsum-pairs_loss": 0.44794514775276184, "eval_xsum-pairs_runtime": 0.1515, "eval_xsum-pairs_samples_per_second": 541.356, "eval_xsum-pairs_steps_per_second": 6.602, "step": 1611 }, { "epoch": 0.45, "eval_compression-pairs_loss": 0.0947626531124115, "eval_compression-pairs_runtime": 0.1026, "eval_compression-pairs_samples_per_second": 1461.376, "eval_compression-pairs_steps_per_second": 19.485, "step": 1611 }, { "epoch": 0.45, "eval_sciq_pairs_loss": 0.16194498538970947, "eval_sciq_pairs_runtime": 4.8935, "eval_sciq_pairs_samples_per_second": 30.653, "eval_sciq_pairs_steps_per_second": 0.409, "step": 1611 }, { "epoch": 0.45, "eval_qasc_pairs_loss": 0.24168159067630768, "eval_qasc_pairs_runtime": 0.633, "eval_qasc_pairs_samples_per_second": 236.951, "eval_qasc_pairs_steps_per_second": 3.159, "step": 1611 }, { "epoch": 0.45, "eval_openbookqa_pairs_loss": 1.5171433687210083, "eval_openbookqa_pairs_runtime": 0.6889, "eval_openbookqa_pairs_samples_per_second": 217.741, "eval_openbookqa_pairs_steps_per_second": 2.903, "step": 1611 }, { "epoch": 0.45, "eval_msmarco_pairs_loss": 0.5660761594772339, "eval_msmarco_pairs_runtime": 1.1539, "eval_msmarco_pairs_samples_per_second": 129.998, "eval_msmarco_pairs_steps_per_second": 1.733, "step": 1611 }, { "epoch": 0.45, "eval_nq_pairs_loss": 0.5651562213897705, "eval_nq_pairs_runtime": 2.2681, "eval_nq_pairs_samples_per_second": 66.133, "eval_nq_pairs_steps_per_second": 0.882, "step": 1611 }, { "epoch": 0.45, "eval_trivia_pairs_loss": 1.0472347736358643, "eval_trivia_pairs_runtime": 3.1346, "eval_trivia_pairs_samples_per_second": 47.854, "eval_trivia_pairs_steps_per_second": 0.638, "step": 1611 }, { "epoch": 0.45, "eval_quora_pairs_loss": 0.2106107771396637, "eval_quora_pairs_runtime": 0.3754, "eval_quora_pairs_samples_per_second": 399.527, "eval_quora_pairs_steps_per_second": 5.327, "step": 1611 }, { "epoch": 0.45, "eval_gooaq_pairs_loss": 0.5418450832366943, "eval_gooaq_pairs_runtime": 0.7637, "eval_gooaq_pairs_samples_per_second": 196.414, "eval_gooaq_pairs_steps_per_second": 2.619, "step": 1611 }, { "epoch": 0.45, "eval_mrpc_pairs_loss": 0.055642638355493546, "eval_mrpc_pairs_runtime": 0.1033, "eval_mrpc_pairs_samples_per_second": 1452.724, "eval_mrpc_pairs_steps_per_second": 19.37, "step": 1611 }, { "epoch": 0.45251396648044695, "grad_norm": 9.676753044128418, "learning_rate": 3.4607737630983708e-06, "loss": 0.8948, "step": 1620 }, { "epoch": 0.45754189944134077, "grad_norm": 7.678776264190674, "learning_rate": 3.2578674574793844e-06, "loss": 1.1183, "step": 1638 }, { "epoch": 0.46256983240223465, "grad_norm": 23.967897415161133, "learning_rate": 3.0690692962414793e-06, "loss": 1.0605, "step": 1656 }, { "epoch": 0.46759776536312847, "grad_norm": 8.221102714538574, "learning_rate": 2.894723317891229e-06, "loss": 0.972, "step": 1674 }, { "epoch": 0.47262569832402235, "grad_norm": 27.992626190185547, "learning_rate": 2.7351472253640043e-06, "loss": 0.7679, "step": 1692 }, { "epoch": 0.4776536312849162, "grad_norm": 8.492770195007324, "learning_rate": 2.590631807088036e-06, "loss": 0.919, "step": 1710 }, { "epoch": 0.48268156424581005, "grad_norm": 3.369513511657715, "learning_rate": 2.4614404070935527e-06, "loss": 0.9308, "step": 1728 }, { "epoch": 0.4877094972067039, "grad_norm": 7.596469879150391, "learning_rate": 2.347808445132657e-06, "loss": 1.0704, "step": 1746 }, { "epoch": 0.49273743016759775, "grad_norm": 11.658712387084961, "learning_rate": 2.249942987684365e-06, "loss": 1.1985, "step": 1764 }, { "epoch": 0.4977653631284916, "grad_norm": 6.525745868682861, "learning_rate": 2.168022370626553e-06, "loss": 1.0564, "step": 1782 }, { "epoch": 0.5, "eval_nli-pairs_loss": 0.9110333919525146, "eval_nli-pairs_runtime": 3.3929, "eval_nli-pairs_samples_per_second": 44.21, "eval_nli-pairs_steps_per_second": 0.589, "eval_sts-test_pearson_cosine": 0.7946860505353394, "eval_sts-test_pearson_dot": 0.5621595297878732, "eval_sts-test_pearson_euclidean": 0.7470211624755623, "eval_sts-test_pearson_manhattan": 0.7437137423401929, "eval_sts-test_pearson_max": 0.7946860505353394, "eval_sts-test_spearman_cosine": 0.8076167403072096, "eval_sts-test_spearman_dot": 0.5400005132562322, "eval_sts-test_spearman_euclidean": 0.7362746334826669, "eval_sts-test_spearman_manhattan": 0.7347256866136587, "eval_sts-test_spearman_max": 0.8076167403072096, "step": 1790 }, { "epoch": 0.5, "eval_vitaminc-pairs_loss": 4.431720733642578, "eval_vitaminc-pairs_runtime": 0.8235, "eval_vitaminc-pairs_samples_per_second": 151.785, "eval_vitaminc-pairs_steps_per_second": 2.429, "step": 1790 }, { "epoch": 0.5, "eval_qnli-contrastive_loss": 0.26112574338912964, "eval_qnli-contrastive_runtime": 0.1181, "eval_qnli-contrastive_samples_per_second": 1270.319, "eval_qnli-contrastive_steps_per_second": 16.938, "step": 1790 }, { "epoch": 0.5, "eval_scitail-pairs-qa_loss": 0.06175601854920387, "eval_scitail-pairs-qa_runtime": 0.6691, "eval_scitail-pairs-qa_samples_per_second": 224.193, "eval_scitail-pairs-qa_steps_per_second": 2.989, "step": 1790 }, { "epoch": 0.5, "eval_scitail-pairs-pos_loss": 0.2788408100605011, "eval_scitail-pairs-pos_runtime": 1.7858, "eval_scitail-pairs-pos_samples_per_second": 83.996, "eval_scitail-pairs-pos_steps_per_second": 1.12, "step": 1790 }, { "epoch": 0.5, "eval_xsum-pairs_loss": 0.4458795487880707, "eval_xsum-pairs_runtime": 0.1537, "eval_xsum-pairs_samples_per_second": 533.68, "eval_xsum-pairs_steps_per_second": 6.508, "step": 1790 }, { "epoch": 0.5, "eval_compression-pairs_loss": 0.09369674324989319, "eval_compression-pairs_runtime": 0.1042, "eval_compression-pairs_samples_per_second": 1439.162, "eval_compression-pairs_steps_per_second": 19.189, "step": 1790 }, { "epoch": 0.5, "eval_sciq_pairs_loss": 0.16267743706703186, "eval_sciq_pairs_runtime": 4.8926, "eval_sciq_pairs_samples_per_second": 30.659, "eval_sciq_pairs_steps_per_second": 0.409, "step": 1790 }, { "epoch": 0.5, "eval_qasc_pairs_loss": 0.24173419177532196, "eval_qasc_pairs_runtime": 0.6349, "eval_qasc_pairs_samples_per_second": 236.24, "eval_qasc_pairs_steps_per_second": 3.15, "step": 1790 }, { "epoch": 0.5, "eval_openbookqa_pairs_loss": 1.512660026550293, "eval_openbookqa_pairs_runtime": 0.6859, "eval_openbookqa_pairs_samples_per_second": 218.688, "eval_openbookqa_pairs_steps_per_second": 2.916, "step": 1790 }, { "epoch": 0.5, "eval_msmarco_pairs_loss": 0.5678529739379883, "eval_msmarco_pairs_runtime": 1.1574, "eval_msmarco_pairs_samples_per_second": 129.597, "eval_msmarco_pairs_steps_per_second": 1.728, "step": 1790 }, { "epoch": 0.5, "eval_nq_pairs_loss": 0.5613307952880859, "eval_nq_pairs_runtime": 2.2694, "eval_nq_pairs_samples_per_second": 66.097, "eval_nq_pairs_steps_per_second": 0.881, "step": 1790 }, { "epoch": 0.5, "eval_trivia_pairs_loss": 1.050702452659607, "eval_trivia_pairs_runtime": 3.169, "eval_trivia_pairs_samples_per_second": 47.334, "eval_trivia_pairs_steps_per_second": 0.631, "step": 1790 }, { "epoch": 0.5, "eval_quora_pairs_loss": 0.20358483493328094, "eval_quora_pairs_runtime": 0.3794, "eval_quora_pairs_samples_per_second": 395.355, "eval_quora_pairs_steps_per_second": 5.271, "step": 1790 }, { "epoch": 0.5, "eval_gooaq_pairs_loss": 0.5364322066307068, "eval_gooaq_pairs_runtime": 0.7677, "eval_gooaq_pairs_samples_per_second": 195.377, "eval_gooaq_pairs_steps_per_second": 2.605, "step": 1790 }, { "epoch": 0.5, "eval_mrpc_pairs_loss": 0.05436507612466812, "eval_mrpc_pairs_runtime": 0.1022, "eval_mrpc_pairs_samples_per_second": 1467.535, "eval_mrpc_pairs_steps_per_second": 19.567, "step": 1790 }, { "epoch": 0.5027932960893855, "grad_norm": 12.61409854888916, "learning_rate": 2.102195874262405e-06, "loss": 0.9128, "step": 1800 }, { "epoch": 0.5078212290502794, "grad_norm": 3.934692621231079, "learning_rate": 2.0525834512935438e-06, "loss": 0.9643, "step": 1818 }, { "epoch": 0.5128491620111731, "grad_norm": 0.5022761225700378, "learning_rate": 2.019275508235552e-06, "loss": 0.8366, "step": 1836 }, { "epoch": 0.517877094972067, "grad_norm": 1.4803472757339478, "learning_rate": 2.00233274067419e-06, "loss": 1.045, "step": 1854 }, { "epoch": 0.5229050279329609, "grad_norm": 3.873300552368164, "learning_rate": 2.0017860226625237e-06, "loss": 0.7239, "step": 1872 }, { "epoch": 0.5279329608938548, "grad_norm": 1.603557825088501, "learning_rate": 2.0176363504605122e-06, "loss": 0.9027, "step": 1890 }, { "epoch": 0.5329608938547487, "grad_norm": 7.767194747924805, "learning_rate": 2.0498548407195652e-06, "loss": 1.0256, "step": 1908 }, { "epoch": 0.5379888268156424, "grad_norm": 0.2919946014881134, "learning_rate": 2.0983827831153845e-06, "loss": 0.6783, "step": 1926 }, { "epoch": 0.5430167597765363, "grad_norm": 4.737668514251709, "learning_rate": 2.163131747333189e-06, "loss": 1.0205, "step": 1944 }, { "epoch": 0.5480446927374302, "grad_norm": 12.112689018249512, "learning_rate": 2.24398374421034e-06, "loss": 1.0501, "step": 1962 }, { "epoch": 0.55, "eval_nli-pairs_loss": 0.9151772260665894, "eval_nli-pairs_runtime": 3.3845, "eval_nli-pairs_samples_per_second": 44.32, "eval_nli-pairs_steps_per_second": 0.591, "eval_sts-test_pearson_cosine": 0.7950386994851146, "eval_sts-test_pearson_dot": 0.561799407632955, "eval_sts-test_pearson_euclidean": 0.7471927000576978, "eval_sts-test_pearson_manhattan": 0.7439585200721782, "eval_sts-test_pearson_max": 0.7950386994851146, "eval_sts-test_spearman_cosine": 0.8079323733036744, "eval_sts-test_spearman_dot": 0.5396753842956418, "eval_sts-test_spearman_euclidean": 0.7365066331647045, "eval_sts-test_spearman_manhattan": 0.7349643316784805, "eval_sts-test_spearman_max": 0.8079323733036744, "step": 1969 }, { "epoch": 0.55, "eval_vitaminc-pairs_loss": 4.449384689331055, "eval_vitaminc-pairs_runtime": 0.8196, "eval_vitaminc-pairs_samples_per_second": 152.52, "eval_vitaminc-pairs_steps_per_second": 2.44, "step": 1969 }, { "epoch": 0.55, "eval_qnli-contrastive_loss": 0.2493373602628708, "eval_qnli-contrastive_runtime": 0.1168, "eval_qnli-contrastive_samples_per_second": 1283.727, "eval_qnli-contrastive_steps_per_second": 17.116, "step": 1969 }, { "epoch": 0.55, "eval_scitail-pairs-qa_loss": 0.06184713542461395, "eval_scitail-pairs-qa_runtime": 0.665, "eval_scitail-pairs-qa_samples_per_second": 225.568, "eval_scitail-pairs-qa_steps_per_second": 3.008, "step": 1969 }, { "epoch": 0.55, "eval_scitail-pairs-pos_loss": 0.2792353332042694, "eval_scitail-pairs-pos_runtime": 1.8207, "eval_scitail-pairs-pos_samples_per_second": 82.388, "eval_scitail-pairs-pos_steps_per_second": 1.099, "step": 1969 }, { "epoch": 0.55, "eval_xsum-pairs_loss": 0.44985902309417725, "eval_xsum-pairs_runtime": 0.153, "eval_xsum-pairs_samples_per_second": 535.847, "eval_xsum-pairs_steps_per_second": 6.535, "step": 1969 }, { "epoch": 0.55, "eval_compression-pairs_loss": 0.09320634603500366, "eval_compression-pairs_runtime": 0.1054, "eval_compression-pairs_samples_per_second": 1423.332, "eval_compression-pairs_steps_per_second": 18.978, "step": 1969 }, { "epoch": 0.55, "eval_sciq_pairs_loss": 0.16240650415420532, "eval_sciq_pairs_runtime": 4.8632, "eval_sciq_pairs_samples_per_second": 30.844, "eval_sciq_pairs_steps_per_second": 0.411, "step": 1969 }, { "epoch": 0.55, "eval_qasc_pairs_loss": 0.23881591856479645, "eval_qasc_pairs_runtime": 0.6406, "eval_qasc_pairs_samples_per_second": 234.166, "eval_qasc_pairs_steps_per_second": 3.122, "step": 1969 }, { "epoch": 0.55, "eval_openbookqa_pairs_loss": 1.5175997018814087, "eval_openbookqa_pairs_runtime": 0.6872, "eval_openbookqa_pairs_samples_per_second": 218.27, "eval_openbookqa_pairs_steps_per_second": 2.91, "step": 1969 }, { "epoch": 0.55, "eval_msmarco_pairs_loss": 0.5688791275024414, "eval_msmarco_pairs_runtime": 1.1543, "eval_msmarco_pairs_samples_per_second": 129.949, "eval_msmarco_pairs_steps_per_second": 1.733, "step": 1969 }, { "epoch": 0.55, "eval_nq_pairs_loss": 0.559186577796936, "eval_nq_pairs_runtime": 2.2663, "eval_nq_pairs_samples_per_second": 66.187, "eval_nq_pairs_steps_per_second": 0.882, "step": 1969 }, { "epoch": 0.55, "eval_trivia_pairs_loss": 1.0515549182891846, "eval_trivia_pairs_runtime": 3.1623, "eval_trivia_pairs_samples_per_second": 47.434, "eval_trivia_pairs_steps_per_second": 0.632, "step": 1969 }, { "epoch": 0.55, "eval_quora_pairs_loss": 0.19930994510650635, "eval_quora_pairs_runtime": 0.3755, "eval_quora_pairs_samples_per_second": 399.469, "eval_quora_pairs_steps_per_second": 5.326, "step": 1969 }, { "epoch": 0.55, "eval_gooaq_pairs_loss": 0.5293665528297424, "eval_gooaq_pairs_runtime": 0.7647, "eval_gooaq_pairs_samples_per_second": 196.156, "eval_gooaq_pairs_steps_per_second": 2.615, "step": 1969 }, { "epoch": 0.55, "eval_mrpc_pairs_loss": 0.05420010909438133, "eval_mrpc_pairs_runtime": 0.1031, "eval_mrpc_pairs_samples_per_second": 1455.466, "eval_mrpc_pairs_steps_per_second": 19.406, "step": 1969 }, { "epoch": 0.553072625698324, "grad_norm": 7.374291896820068, "learning_rate": 2.340791440742762e-06, "loss": 1.0035, "step": 1980 }, { "epoch": 0.5581005586592179, "grad_norm": 10.105379104614258, "learning_rate": 2.453378428563332e-06, "loss": 0.8237, "step": 1998 }, { "epoch": 0.5631284916201117, "grad_norm": 0.47507917881011963, "learning_rate": 2.581539545403014e-06, "loss": 0.7804, "step": 2016 }, { "epoch": 0.5681564245810056, "grad_norm": 8.750273704528809, "learning_rate": 2.725041248948953e-06, "loss": 0.9602, "step": 2034 }, { "epoch": 0.5731843575418994, "grad_norm": 6.673968315124512, "learning_rate": 2.8836220424182614e-06, "loss": 0.6921, "step": 2052 }, { "epoch": 0.5782122905027933, "grad_norm": 7.76973295211792, "learning_rate": 3.056992951071994e-06, "loss": 1.1073, "step": 2070 }, { "epoch": 0.5832402234636872, "grad_norm": 11.25104808807373, "learning_rate": 3.2448380488009927e-06, "loss": 1.0952, "step": 2088 }, { "epoch": 0.588268156424581, "grad_norm": 10.827056884765625, "learning_rate": 3.446815033824003e-06, "loss": 0.986, "step": 2106 }, { "epoch": 0.5932960893854748, "grad_norm": 2.704712152481079, "learning_rate": 3.662555852449e-06, "loss": 0.9821, "step": 2124 }, { "epoch": 0.5983240223463687, "grad_norm": 12.812453269958496, "learning_rate": 3.891667369761114e-06, "loss": 1.4121, "step": 2142 }, { "epoch": 0.6, "eval_nli-pairs_loss": 0.9104093909263611, "eval_nli-pairs_runtime": 3.4531, "eval_nli-pairs_samples_per_second": 43.439, "eval_nli-pairs_steps_per_second": 0.579, "eval_sts-test_pearson_cosine": 0.7935960995769732, "eval_sts-test_pearson_dot": 0.5580627112302, "eval_sts-test_pearson_euclidean": 0.7478812796510826, "eval_sts-test_pearson_manhattan": 0.7447776866041059, "eval_sts-test_pearson_max": 0.7935960995769732, "eval_sts-test_spearman_cosine": 0.8077076901857909, "eval_sts-test_spearman_dot": 0.5352305509553539, "eval_sts-test_spearman_euclidean": 0.7369380974127236, "eval_sts-test_spearman_manhattan": 0.7355092851368631, "eval_sts-test_spearman_max": 0.8077076901857909, "step": 2148 }, { "epoch": 0.6, "eval_vitaminc-pairs_loss": 4.4026408195495605, "eval_vitaminc-pairs_runtime": 0.8318, "eval_vitaminc-pairs_samples_per_second": 150.285, "eval_vitaminc-pairs_steps_per_second": 2.405, "step": 2148 }, { "epoch": 0.6, "eval_qnli-contrastive_loss": 0.25988292694091797, "eval_qnli-contrastive_runtime": 0.1194, "eval_qnli-contrastive_samples_per_second": 1255.948, "eval_qnli-contrastive_steps_per_second": 16.746, "step": 2148 }, { "epoch": 0.6, "eval_scitail-pairs-qa_loss": 0.06044253334403038, "eval_scitail-pairs-qa_runtime": 0.674, "eval_scitail-pairs-qa_samples_per_second": 222.537, "eval_scitail-pairs-qa_steps_per_second": 2.967, "step": 2148 }, { "epoch": 0.6, "eval_scitail-pairs-pos_loss": 0.28237131237983704, "eval_scitail-pairs-pos_runtime": 1.8129, "eval_scitail-pairs-pos_samples_per_second": 82.738, "eval_scitail-pairs-pos_steps_per_second": 1.103, "step": 2148 }, { "epoch": 0.6, "eval_xsum-pairs_loss": 0.4531600773334503, "eval_xsum-pairs_runtime": 0.1542, "eval_xsum-pairs_samples_per_second": 531.931, "eval_xsum-pairs_steps_per_second": 6.487, "step": 2148 }, { "epoch": 0.6, "eval_compression-pairs_loss": 0.0922146737575531, "eval_compression-pairs_runtime": 0.1039, "eval_compression-pairs_samples_per_second": 1444.015, "eval_compression-pairs_steps_per_second": 19.254, "step": 2148 }, { "epoch": 0.6, "eval_sciq_pairs_loss": 0.16167433559894562, "eval_sciq_pairs_runtime": 4.8709, "eval_sciq_pairs_samples_per_second": 30.795, "eval_sciq_pairs_steps_per_second": 0.411, "step": 2148 }, { "epoch": 0.6, "eval_qasc_pairs_loss": 0.2368052899837494, "eval_qasc_pairs_runtime": 0.6338, "eval_qasc_pairs_samples_per_second": 236.682, "eval_qasc_pairs_steps_per_second": 3.156, "step": 2148 }, { "epoch": 0.6, "eval_openbookqa_pairs_loss": 1.515324354171753, "eval_openbookqa_pairs_runtime": 0.6973, "eval_openbookqa_pairs_samples_per_second": 215.117, "eval_openbookqa_pairs_steps_per_second": 2.868, "step": 2148 }, { "epoch": 0.6, "eval_msmarco_pairs_loss": 0.5706735253334045, "eval_msmarco_pairs_runtime": 1.2143, "eval_msmarco_pairs_samples_per_second": 123.528, "eval_msmarco_pairs_steps_per_second": 1.647, "step": 2148 }, { "epoch": 0.6, "eval_nq_pairs_loss": 0.5522983074188232, "eval_nq_pairs_runtime": 2.288, "eval_nq_pairs_samples_per_second": 65.56, "eval_nq_pairs_steps_per_second": 0.874, "step": 2148 }, { "epoch": 0.6, "eval_trivia_pairs_loss": 1.0437382459640503, "eval_trivia_pairs_runtime": 3.1388, "eval_trivia_pairs_samples_per_second": 47.789, "eval_trivia_pairs_steps_per_second": 0.637, "step": 2148 }, { "epoch": 0.6, "eval_quora_pairs_loss": 0.2811141908168793, "eval_quora_pairs_runtime": 0.3726, "eval_quora_pairs_samples_per_second": 402.612, "eval_quora_pairs_steps_per_second": 5.368, "step": 2148 }, { "epoch": 0.6, "eval_gooaq_pairs_loss": 0.5305163860321045, "eval_gooaq_pairs_runtime": 0.7714, "eval_gooaq_pairs_samples_per_second": 194.459, "eval_gooaq_pairs_steps_per_second": 2.593, "step": 2148 }, { "epoch": 0.6, "eval_mrpc_pairs_loss": 0.05323876813054085, "eval_mrpc_pairs_runtime": 0.1045, "eval_mrpc_pairs_samples_per_second": 1435.059, "eval_mrpc_pairs_steps_per_second": 19.134, "step": 2148 }, { "epoch": 0.6033519553072626, "grad_norm": 11.035157203674316, "learning_rate": 4.133732086014913e-06, "loss": 0.8797, "step": 2160 }, { "epoch": 0.6083798882681565, "grad_norm": 7.317018985748291, "learning_rate": 4.388308897425655e-06, "loss": 0.9994, "step": 2178 }, { "epoch": 0.6134078212290502, "grad_norm": 4.9388532638549805, "learning_rate": 4.654933899973131e-06, "loss": 1.0541, "step": 2196 }, { "epoch": 0.6184357541899441, "grad_norm": 1.5112260580062866, "learning_rate": 4.93312123475336e-06, "loss": 0.8337, "step": 2214 }, { "epoch": 0.623463687150838, "grad_norm": 4.268815994262695, "learning_rate": 5.222363973337719e-06, "loss": 1.1439, "step": 2232 }, { "epoch": 0.6284916201117319, "grad_norm": 14.816475868225098, "learning_rate": 5.522135041526076e-06, "loss": 0.9475, "step": 2250 }, { "epoch": 0.6335195530726258, "grad_norm": 9.077942848205566, "learning_rate": 5.831888179810709e-06, "loss": 0.9405, "step": 2268 }, { "epoch": 0.6385474860335195, "grad_norm": 0.8216752409934998, "learning_rate": 6.151058938800809e-06, "loss": 0.9019, "step": 2286 }, { "epoch": 0.6435754189944134, "grad_norm": 10.906030654907227, "learning_rate": 6.479065707793453e-06, "loss": 1.1584, "step": 2304 }, { "epoch": 0.6486033519553073, "grad_norm": 0.814568817615509, "learning_rate": 6.8153107746170145e-06, "loss": 0.7085, "step": 2322 }, { "epoch": 0.65, "eval_nli-pairs_loss": 0.9121571183204651, "eval_nli-pairs_runtime": 3.4046, "eval_nli-pairs_samples_per_second": 44.058, "eval_nli-pairs_steps_per_second": 0.587, "eval_sts-test_pearson_cosine": 0.7970926005009543, "eval_sts-test_pearson_dot": 0.5555037334963182, "eval_sts-test_pearson_euclidean": 0.7475629768769588, "eval_sts-test_pearson_manhattan": 0.7444142460571747, "eval_sts-test_pearson_max": 0.7970926005009543, "eval_sts-test_spearman_cosine": 0.8103647735324723, "eval_sts-test_spearman_dot": 0.5330195901829358, "eval_sts-test_spearman_euclidean": 0.7373475776074586, "eval_sts-test_spearman_manhattan": 0.7363774296055311, "eval_sts-test_spearman_max": 0.8103647735324723, "step": 2327 }, { "epoch": 0.65, "eval_vitaminc-pairs_loss": 4.408538818359375, "eval_vitaminc-pairs_runtime": 0.8044, "eval_vitaminc-pairs_samples_per_second": 155.391, "eval_vitaminc-pairs_steps_per_second": 2.486, "step": 2327 }, { "epoch": 0.65, "eval_qnli-contrastive_loss": 0.21634440124034882, "eval_qnli-contrastive_runtime": 0.1181, "eval_qnli-contrastive_samples_per_second": 1270.224, "eval_qnli-contrastive_steps_per_second": 16.936, "step": 2327 }, { "epoch": 0.65, "eval_scitail-pairs-qa_loss": 0.061703551560640335, "eval_scitail-pairs-qa_runtime": 0.6703, "eval_scitail-pairs-qa_samples_per_second": 223.786, "eval_scitail-pairs-qa_steps_per_second": 2.984, "step": 2327 }, { "epoch": 0.65, "eval_scitail-pairs-pos_loss": 0.28146570920944214, "eval_scitail-pairs-pos_runtime": 1.7775, "eval_scitail-pairs-pos_samples_per_second": 84.389, "eval_scitail-pairs-pos_steps_per_second": 1.125, "step": 2327 }, { "epoch": 0.65, "eval_xsum-pairs_loss": 0.45509228110313416, "eval_xsum-pairs_runtime": 0.1516, "eval_xsum-pairs_samples_per_second": 540.918, "eval_xsum-pairs_steps_per_second": 6.597, "step": 2327 }, { "epoch": 0.65, "eval_compression-pairs_loss": 0.09231606870889664, "eval_compression-pairs_runtime": 0.1055, "eval_compression-pairs_samples_per_second": 1422.184, "eval_compression-pairs_steps_per_second": 18.962, "step": 2327 }, { "epoch": 0.65, "eval_sciq_pairs_loss": 0.16050980985164642, "eval_sciq_pairs_runtime": 4.8877, "eval_sciq_pairs_samples_per_second": 30.689, "eval_sciq_pairs_steps_per_second": 0.409, "step": 2327 }, { "epoch": 0.65, "eval_qasc_pairs_loss": 0.23437899351119995, "eval_qasc_pairs_runtime": 0.6417, "eval_qasc_pairs_samples_per_second": 233.769, "eval_qasc_pairs_steps_per_second": 3.117, "step": 2327 }, { "epoch": 0.65, "eval_openbookqa_pairs_loss": 1.520011067390442, "eval_openbookqa_pairs_runtime": 0.6877, "eval_openbookqa_pairs_samples_per_second": 218.105, "eval_openbookqa_pairs_steps_per_second": 2.908, "step": 2327 }, { "epoch": 0.65, "eval_msmarco_pairs_loss": 0.5828911066055298, "eval_msmarco_pairs_runtime": 1.1629, "eval_msmarco_pairs_samples_per_second": 128.992, "eval_msmarco_pairs_steps_per_second": 1.72, "step": 2327 }, { "epoch": 0.65, "eval_nq_pairs_loss": 0.5630797147750854, "eval_nq_pairs_runtime": 2.2651, "eval_nq_pairs_samples_per_second": 66.221, "eval_nq_pairs_steps_per_second": 0.883, "step": 2327 }, { "epoch": 0.65, "eval_trivia_pairs_loss": 1.0719337463378906, "eval_trivia_pairs_runtime": 3.1435, "eval_trivia_pairs_samples_per_second": 47.718, "eval_trivia_pairs_steps_per_second": 0.636, "step": 2327 }, { "epoch": 0.65, "eval_quora_pairs_loss": 0.1920953392982483, "eval_quora_pairs_runtime": 0.3849, "eval_quora_pairs_samples_per_second": 389.684, "eval_quora_pairs_steps_per_second": 5.196, "step": 2327 }, { "epoch": 0.65, "eval_gooaq_pairs_loss": 0.5226368308067322, "eval_gooaq_pairs_runtime": 0.7853, "eval_gooaq_pairs_samples_per_second": 191.015, "eval_gooaq_pairs_steps_per_second": 2.547, "step": 2327 }, { "epoch": 0.65, "eval_mrpc_pairs_loss": 0.05354393646121025, "eval_mrpc_pairs_runtime": 0.1099, "eval_mrpc_pairs_samples_per_second": 1364.877, "eval_mrpc_pairs_steps_per_second": 18.198, "step": 2327 }, { "epoch": 0.6536312849162011, "grad_norm": 10.176216125488281, "learning_rate": 7.15918141481555e-06, "loss": 0.8162, "step": 2340 }, { "epoch": 0.658659217877095, "grad_norm": 0.32444873452186584, "learning_rate": 7.510051008189288e-06, "loss": 1.1129, "step": 2358 }, { "epoch": 0.6636871508379888, "grad_norm": 1.4217174053192139, "learning_rate": 7.86728018065684e-06, "loss": 0.7191, "step": 2376 }, { "epoch": 0.6687150837988827, "grad_norm": 2.12044095993042, "learning_rate": 8.230217969358248e-06, "loss": 1.0294, "step": 2394 }, { "epoch": 0.6737430167597765, "grad_norm": 23.4191837310791, "learning_rate": 8.598203008875594e-06, "loss": 0.8037, "step": 2412 }, { "epoch": 0.6787709497206704, "grad_norm": 13.322118759155273, "learning_rate": 8.970564736409934e-06, "loss": 1.0468, "step": 2430 }, { "epoch": 0.6837988826815642, "grad_norm": 0.4028688669204712, "learning_rate": 9.346624613718075e-06, "loss": 1.1034, "step": 2448 }, { "epoch": 0.6888268156424581, "grad_norm": 6.083842754364014, "learning_rate": 9.72569736358266e-06, "loss": 1.0727, "step": 2466 }, { "epoch": 0.693854748603352, "grad_norm": 0.2788317799568176, "learning_rate": 1.0107092218562473e-05, "loss": 1.0214, "step": 2484 }, { "epoch": 0.6988826815642458, "grad_norm": 11.104997634887695, "learning_rate": 1.0490114179747249e-05, "loss": 1.2315, "step": 2502 }, { "epoch": 0.7, "eval_nli-pairs_loss": 0.9116527438163757, "eval_nli-pairs_runtime": 3.3805, "eval_nli-pairs_samples_per_second": 44.372, "eval_nli-pairs_steps_per_second": 0.592, "eval_sts-test_pearson_cosine": 0.7968033558311296, "eval_sts-test_pearson_dot": 0.5579943563931122, "eval_sts-test_pearson_euclidean": 0.7488391717544687, "eval_sts-test_pearson_manhattan": 0.745738651526943, "eval_sts-test_pearson_max": 0.7968033558311296, "eval_sts-test_spearman_cosine": 0.8110346629360586, "eval_sts-test_spearman_dot": 0.5361696642637424, "eval_sts-test_spearman_euclidean": 0.739312659216871, "eval_sts-test_spearman_manhattan": 0.7382491849343503, "eval_sts-test_spearman_max": 0.8110346629360586, "step": 2506 }, { "epoch": 0.7, "eval_vitaminc-pairs_loss": 4.264810562133789, "eval_vitaminc-pairs_runtime": 0.8098, "eval_vitaminc-pairs_samples_per_second": 154.364, "eval_vitaminc-pairs_steps_per_second": 2.47, "step": 2506 }, { "epoch": 0.7, "eval_qnli-contrastive_loss": 0.2539684474468231, "eval_qnli-contrastive_runtime": 0.1181, "eval_qnli-contrastive_samples_per_second": 1269.624, "eval_qnli-contrastive_steps_per_second": 16.928, "step": 2506 }, { "epoch": 0.7, "eval_scitail-pairs-qa_loss": 0.06219513714313507, "eval_scitail-pairs-qa_runtime": 0.6687, "eval_scitail-pairs-qa_samples_per_second": 224.322, "eval_scitail-pairs-qa_steps_per_second": 2.991, "step": 2506 }, { "epoch": 0.7, "eval_scitail-pairs-pos_loss": 0.279880166053772, "eval_scitail-pairs-pos_runtime": 1.7838, "eval_scitail-pairs-pos_samples_per_second": 84.091, "eval_scitail-pairs-pos_steps_per_second": 1.121, "step": 2506 }, { "epoch": 0.7, "eval_xsum-pairs_loss": 0.4540919363498688, "eval_xsum-pairs_runtime": 0.1547, "eval_xsum-pairs_samples_per_second": 530.132, "eval_xsum-pairs_steps_per_second": 6.465, "step": 2506 }, { "epoch": 0.7, "eval_compression-pairs_loss": 0.09507277607917786, "eval_compression-pairs_runtime": 0.1065, "eval_compression-pairs_samples_per_second": 1408.111, "eval_compression-pairs_steps_per_second": 18.775, "step": 2506 }, { "epoch": 0.7, "eval_sciq_pairs_loss": 0.1653744876384735, "eval_sciq_pairs_runtime": 4.8569, "eval_sciq_pairs_samples_per_second": 30.884, "eval_sciq_pairs_steps_per_second": 0.412, "step": 2506 }, { "epoch": 0.7, "eval_qasc_pairs_loss": 0.23888275027275085, "eval_qasc_pairs_runtime": 0.648, "eval_qasc_pairs_samples_per_second": 231.489, "eval_qasc_pairs_steps_per_second": 3.087, "step": 2506 }, { "epoch": 0.7, "eval_openbookqa_pairs_loss": 1.5343717336654663, "eval_openbookqa_pairs_runtime": 0.6915, "eval_openbookqa_pairs_samples_per_second": 216.931, "eval_openbookqa_pairs_steps_per_second": 2.892, "step": 2506 }, { "epoch": 0.7, "eval_msmarco_pairs_loss": 0.5778762698173523, "eval_msmarco_pairs_runtime": 1.1608, "eval_msmarco_pairs_samples_per_second": 129.225, "eval_msmarco_pairs_steps_per_second": 1.723, "step": 2506 }, { "epoch": 0.7, "eval_nq_pairs_loss": 0.573172390460968, "eval_nq_pairs_runtime": 2.2658, "eval_nq_pairs_samples_per_second": 66.202, "eval_nq_pairs_steps_per_second": 0.883, "step": 2506 }, { "epoch": 0.7, "eval_trivia_pairs_loss": 1.1024057865142822, "eval_trivia_pairs_runtime": 3.1397, "eval_trivia_pairs_samples_per_second": 47.776, "eval_trivia_pairs_steps_per_second": 0.637, "step": 2506 }, { "epoch": 0.7, "eval_quora_pairs_loss": 0.19568383693695068, "eval_quora_pairs_runtime": 0.3839, "eval_quora_pairs_samples_per_second": 390.744, "eval_quora_pairs_steps_per_second": 5.21, "step": 2506 }, { "epoch": 0.7, "eval_gooaq_pairs_loss": 0.5365306735038757, "eval_gooaq_pairs_runtime": 0.7966, "eval_gooaq_pairs_samples_per_second": 188.307, "eval_gooaq_pairs_steps_per_second": 2.511, "step": 2506 }, { "epoch": 0.7, "eval_mrpc_pairs_loss": 0.055629413574934006, "eval_mrpc_pairs_runtime": 0.1021, "eval_mrpc_pairs_samples_per_second": 1468.536, "eval_mrpc_pairs_steps_per_second": 19.58, "step": 2506 }, { "epoch": 0.7039106145251397, "grad_norm": 4.487581253051758, "learning_rate": 1.0874065283223303e-05, "loss": 0.8282, "step": 2520 }, { "epoch": 0.7089385474860335, "grad_norm": 4.290740966796875, "learning_rate": 1.1258245871942228e-05, "loss": 1.0656, "step": 2538 }, { "epoch": 0.7139664804469273, "grad_norm": 19.562671661376953, "learning_rate": 1.1641955870674848e-05, "loss": 0.8865, "step": 2556 }, { "epoch": 0.7189944134078212, "grad_norm": 1.9158836603164673, "learning_rate": 1.2024496061727222e-05, "loss": 1.2046, "step": 2574 }, { "epoch": 0.7240223463687151, "grad_norm": 16.630388259887695, "learning_rate": 1.2405169359094e-05, "loss": 0.8117, "step": 2592 }, { "epoch": 0.729050279329609, "grad_norm": 5.154221057891846, "learning_rate": 1.2783282078727288e-05, "loss": 1.0844, "step": 2610 }, { "epoch": 0.7340782122905027, "grad_norm": 7.845170021057129, "learning_rate": 1.3158145202606363e-05, "loss": 1.0884, "step": 2628 }, { "epoch": 0.7391061452513966, "grad_norm": 11.965822219848633, "learning_rate": 1.3529075634304503e-05, "loss": 0.8428, "step": 2646 }, { "epoch": 0.7441340782122905, "grad_norm": 0.3276151418685913, "learning_rate": 1.3895397443765348e-05, "loss": 0.9549, "step": 2664 }, { "epoch": 0.7491620111731844, "grad_norm": 1.0129306316375732, "learning_rate": 1.4256443099020256e-05, "loss": 0.5632, "step": 2682 }, { "epoch": 0.75, "eval_nli-pairs_loss": 0.8929781317710876, "eval_nli-pairs_runtime": 3.3406, "eval_nli-pairs_samples_per_second": 44.902, "eval_nli-pairs_steps_per_second": 0.599, "eval_sts-test_pearson_cosine": 0.7952756314257663, "eval_sts-test_pearson_dot": 0.5371812831147439, "eval_sts-test_pearson_euclidean": 0.744816506031064, "eval_sts-test_pearson_manhattan": 0.741366674634691, "eval_sts-test_pearson_max": 0.7952756314257663, "eval_sts-test_spearman_cosine": 0.8078984683362199, "eval_sts-test_spearman_dot": 0.5136140187219894, "eval_sts-test_spearman_euclidean": 0.7347138642005089, "eval_sts-test_spearman_manhattan": 0.733563679240801, "eval_sts-test_spearman_max": 0.8078984683362199, "step": 2685 }, { "epoch": 0.75, "eval_vitaminc-pairs_loss": 4.211962699890137, "eval_vitaminc-pairs_runtime": 0.8026, "eval_vitaminc-pairs_samples_per_second": 155.743, "eval_vitaminc-pairs_steps_per_second": 2.492, "step": 2685 }, { "epoch": 0.75, "eval_qnli-contrastive_loss": 0.24487823247909546, "eval_qnli-contrastive_runtime": 0.117, "eval_qnli-contrastive_samples_per_second": 1281.614, "eval_qnli-contrastive_steps_per_second": 17.088, "step": 2685 }, { "epoch": 0.75, "eval_scitail-pairs-qa_loss": 0.06043863296508789, "eval_scitail-pairs-qa_runtime": 0.6999, "eval_scitail-pairs-qa_samples_per_second": 214.315, "eval_scitail-pairs-qa_steps_per_second": 2.858, "step": 2685 }, { "epoch": 0.75, "eval_scitail-pairs-pos_loss": 0.2817319333553314, "eval_scitail-pairs-pos_runtime": 1.8256, "eval_scitail-pairs-pos_samples_per_second": 82.165, "eval_scitail-pairs-pos_steps_per_second": 1.096, "step": 2685 }, { "epoch": 0.75, "eval_xsum-pairs_loss": 0.47503286600112915, "eval_xsum-pairs_runtime": 0.1513, "eval_xsum-pairs_samples_per_second": 541.963, "eval_xsum-pairs_steps_per_second": 6.609, "step": 2685 }, { "epoch": 0.75, "eval_compression-pairs_loss": 0.09204871207475662, "eval_compression-pairs_runtime": 0.1036, "eval_compression-pairs_samples_per_second": 1448.093, "eval_compression-pairs_steps_per_second": 19.308, "step": 2685 }, { "epoch": 0.75, "eval_sciq_pairs_loss": 0.16384734213352203, "eval_sciq_pairs_runtime": 4.7989, "eval_sciq_pairs_samples_per_second": 31.257, "eval_sciq_pairs_steps_per_second": 0.417, "step": 2685 }, { "epoch": 0.75, "eval_qasc_pairs_loss": 0.2411985844373703, "eval_qasc_pairs_runtime": 0.6347, "eval_qasc_pairs_samples_per_second": 236.317, "eval_qasc_pairs_steps_per_second": 3.151, "step": 2685 }, { "epoch": 0.75, "eval_openbookqa_pairs_loss": 1.5304628610610962, "eval_openbookqa_pairs_runtime": 0.6807, "eval_openbookqa_pairs_samples_per_second": 220.369, "eval_openbookqa_pairs_steps_per_second": 2.938, "step": 2685 }, { "epoch": 0.75, "eval_msmarco_pairs_loss": 0.5842806100845337, "eval_msmarco_pairs_runtime": 1.1452, "eval_msmarco_pairs_samples_per_second": 130.985, "eval_msmarco_pairs_steps_per_second": 1.746, "step": 2685 }, { "epoch": 0.75, "eval_nq_pairs_loss": 0.5671815276145935, "eval_nq_pairs_runtime": 2.2692, "eval_nq_pairs_samples_per_second": 66.103, "eval_nq_pairs_steps_per_second": 0.881, "step": 2685 }, { "epoch": 0.75, "eval_trivia_pairs_loss": 1.141603946685791, "eval_trivia_pairs_runtime": 3.1531, "eval_trivia_pairs_samples_per_second": 47.573, "eval_trivia_pairs_steps_per_second": 0.634, "step": 2685 }, { "epoch": 0.75, "eval_quora_pairs_loss": 0.22496861219406128, "eval_quora_pairs_runtime": 0.3722, "eval_quora_pairs_samples_per_second": 403.006, "eval_quora_pairs_steps_per_second": 5.373, "step": 2685 }, { "epoch": 0.75, "eval_gooaq_pairs_loss": 0.5398057103157043, "eval_gooaq_pairs_runtime": 0.7603, "eval_gooaq_pairs_samples_per_second": 197.281, "eval_gooaq_pairs_steps_per_second": 2.63, "step": 2685 }, { "epoch": 0.75, "eval_mrpc_pairs_loss": 0.05283508449792862, "eval_mrpc_pairs_runtime": 0.1013, "eval_mrpc_pairs_samples_per_second": 1480.799, "eval_mrpc_pairs_steps_per_second": 19.744, "step": 2685 }, { "epoch": 0.7541899441340782, "grad_norm": 0.8120360970497131, "learning_rate": 1.4611554682602087e-05, "loss": 0.7252, "step": 2700 }, { "epoch": 0.759217877094972, "grad_norm": 1.0719642639160156, "learning_rate": 1.4960085090439056e-05, "loss": 0.633, "step": 2718 }, { "epoch": 0.7642458100558659, "grad_norm": 23.75746726989746, "learning_rate": 1.5301399211043744e-05, "loss": 1.0046, "step": 2736 }, { "epoch": 0.7692737430167598, "grad_norm": 7.527226448059082, "learning_rate": 1.563487508284853e-05, "loss": 0.8778, "step": 2754 }, { "epoch": 0.7743016759776536, "grad_norm": 1.64340078830719, "learning_rate": 1.595990502757857e-05, "loss": 0.7953, "step": 2772 }, { "epoch": 0.7793296089385475, "grad_norm": 0.6890662312507629, "learning_rate": 1.627589675759695e-05, "loss": 0.8779, "step": 2790 }, { "epoch": 0.7843575418994413, "grad_norm": 3.8019280433654785, "learning_rate": 1.658227445520409e-05, "loss": 1.1106, "step": 2808 }, { "epoch": 0.7893854748603352, "grad_norm": 9.067117691040039, "learning_rate": 1.6878479821924834e-05, "loss": 0.9071, "step": 2826 }, { "epoch": 0.794413407821229, "grad_norm": 1.4688657522201538, "learning_rate": 1.7163973095870953e-05, "loss": 0.8823, "step": 2844 }, { "epoch": 0.7994413407821229, "grad_norm": 9.774276733398438, "learning_rate": 1.7438234035325246e-05, "loss": 0.9773, "step": 2862 }, { "epoch": 0.8, "eval_nli-pairs_loss": 0.8836459517478943, "eval_nli-pairs_runtime": 3.3741, "eval_nli-pairs_samples_per_second": 44.456, "eval_nli-pairs_steps_per_second": 0.593, "eval_sts-test_pearson_cosine": 0.7936478561473643, "eval_sts-test_pearson_dot": 0.5464208315752136, "eval_sts-test_pearson_euclidean": 0.744815303847436, "eval_sts-test_pearson_manhattan": 0.7406874236561795, "eval_sts-test_pearson_max": 0.7936478561473643, "eval_sts-test_spearman_cosine": 0.8087901362229535, "eval_sts-test_spearman_dot": 0.522859828440202, "eval_sts-test_spearman_euclidean": 0.7366541166806573, "eval_sts-test_spearman_manhattan": 0.7345184313041931, "eval_sts-test_spearman_max": 0.8087901362229535, "step": 2864 }, { "epoch": 0.8, "eval_vitaminc-pairs_loss": 4.289100646972656, "eval_vitaminc-pairs_runtime": 0.8298, "eval_vitaminc-pairs_samples_per_second": 150.644, "eval_vitaminc-pairs_steps_per_second": 2.41, "step": 2864 }, { "epoch": 0.8, "eval_qnli-contrastive_loss": 0.234215646982193, "eval_qnli-contrastive_runtime": 0.1154, "eval_qnli-contrastive_samples_per_second": 1299.968, "eval_qnli-contrastive_steps_per_second": 17.333, "step": 2864 }, { "epoch": 0.8, "eval_scitail-pairs-qa_loss": 0.06133409962058067, "eval_scitail-pairs-qa_runtime": 0.666, "eval_scitail-pairs-qa_samples_per_second": 225.222, "eval_scitail-pairs-qa_steps_per_second": 3.003, "step": 2864 }, { "epoch": 0.8, "eval_scitail-pairs-pos_loss": 0.27694422006607056, "eval_scitail-pairs-pos_runtime": 1.7662, "eval_scitail-pairs-pos_samples_per_second": 84.927, "eval_scitail-pairs-pos_steps_per_second": 1.132, "step": 2864 }, { "epoch": 0.8, "eval_xsum-pairs_loss": 0.4549708962440491, "eval_xsum-pairs_runtime": 0.1519, "eval_xsum-pairs_samples_per_second": 539.692, "eval_xsum-pairs_steps_per_second": 6.582, "step": 2864 }, { "epoch": 0.8, "eval_compression-pairs_loss": 0.0935271754860878, "eval_compression-pairs_runtime": 0.1052, "eval_compression-pairs_samples_per_second": 1425.638, "eval_compression-pairs_steps_per_second": 19.009, "step": 2864 }, { "epoch": 0.8, "eval_sciq_pairs_loss": 0.16447363793849945, "eval_sciq_pairs_runtime": 4.8534, "eval_sciq_pairs_samples_per_second": 30.906, "eval_sciq_pairs_steps_per_second": 0.412, "step": 2864 }, { "epoch": 0.8, "eval_qasc_pairs_loss": 0.24154072999954224, "eval_qasc_pairs_runtime": 0.6435, "eval_qasc_pairs_samples_per_second": 233.112, "eval_qasc_pairs_steps_per_second": 3.108, "step": 2864 }, { "epoch": 0.8, "eval_openbookqa_pairs_loss": 1.5259324312210083, "eval_openbookqa_pairs_runtime": 0.6829, "eval_openbookqa_pairs_samples_per_second": 219.654, "eval_openbookqa_pairs_steps_per_second": 2.929, "step": 2864 }, { "epoch": 0.8, "eval_msmarco_pairs_loss": 0.5627599358558655, "eval_msmarco_pairs_runtime": 1.1467, "eval_msmarco_pairs_samples_per_second": 130.815, "eval_msmarco_pairs_steps_per_second": 1.744, "step": 2864 }, { "epoch": 0.8, "eval_nq_pairs_loss": 0.5455084443092346, "eval_nq_pairs_runtime": 2.2531, "eval_nq_pairs_samples_per_second": 66.576, "eval_nq_pairs_steps_per_second": 0.888, "step": 2864 }, { "epoch": 0.8, "eval_trivia_pairs_loss": 1.1112126111984253, "eval_trivia_pairs_runtime": 3.1293, "eval_trivia_pairs_samples_per_second": 47.934, "eval_trivia_pairs_steps_per_second": 0.639, "step": 2864 }, { "epoch": 0.8, "eval_quora_pairs_loss": 0.1693161278963089, "eval_quora_pairs_runtime": 0.4078, "eval_quora_pairs_samples_per_second": 367.828, "eval_quora_pairs_steps_per_second": 4.904, "step": 2864 }, { "epoch": 0.8, "eval_gooaq_pairs_loss": 0.5442405939102173, "eval_gooaq_pairs_runtime": 0.7657, "eval_gooaq_pairs_samples_per_second": 195.905, "eval_gooaq_pairs_steps_per_second": 2.612, "step": 2864 }, { "epoch": 0.8, "eval_mrpc_pairs_loss": 0.05412428081035614, "eval_mrpc_pairs_runtime": 0.1012, "eval_mrpc_pairs_samples_per_second": 1482.925, "eval_mrpc_pairs_steps_per_second": 19.772, "step": 2864 } ], "logging_steps": 18, "max_steps": 7160, "num_input_tokens_seen": 0, "num_train_epochs": 2, "save_steps": 716, "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": 82, "trial_name": null, "trial_params": null }