| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07079646017699115, | |
| "eval_steps": 40, | |
| "global_step": 80, | |
| "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 | |
| } | |
| ], | |
| "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 | |
| } | |