| { | |
| "best_metric": 0.6135068535804749, | |
| "best_model_checkpoint": "autotrain-8xfer-ymqrk/checkpoint-129", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 129, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 14.928908348083496, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 1.2885, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 10.102056503295898, | |
| "learning_rate": 1.153846153846154e-05, | |
| "loss": 1.3625, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 7.909735202789307, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 1.3634, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 7.1076507568359375, | |
| "learning_rate": 2.6923076923076923e-05, | |
| "loss": 1.2046, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 9.337615966796875, | |
| "learning_rate": 3.461538461538462e-05, | |
| "loss": 1.2876, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 8.110074043273926, | |
| "learning_rate": 4.230769230769231e-05, | |
| "loss": 0.971, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 9.69324779510498, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1834, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 11.01219367980957, | |
| "learning_rate": 4.913793103448276e-05, | |
| "loss": 0.9608, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 6.818650722503662, | |
| "learning_rate": 4.827586206896552e-05, | |
| "loss": 0.9447, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 5.1624579429626465, | |
| "learning_rate": 4.741379310344828e-05, | |
| "loss": 0.6179, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 10.717084884643555, | |
| "learning_rate": 4.655172413793104e-05, | |
| "loss": 1.3452, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 4.0912017822265625, | |
| "learning_rate": 4.5689655172413794e-05, | |
| "loss": 0.8958, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 5.135016918182373, | |
| "learning_rate": 4.482758620689655e-05, | |
| "loss": 0.9057, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 9.611163139343262, | |
| "learning_rate": 4.396551724137931e-05, | |
| "loss": 1.15, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 8.786578178405762, | |
| "learning_rate": 4.3103448275862066e-05, | |
| "loss": 1.136, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 4.468399524688721, | |
| "learning_rate": 4.224137931034483e-05, | |
| "loss": 0.9317, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 6.640096664428711, | |
| "learning_rate": 4.1379310344827587e-05, | |
| "loss": 0.8963, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 4.727297306060791, | |
| "learning_rate": 4.0517241379310344e-05, | |
| "loss": 0.5773, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 4.164635181427002, | |
| "learning_rate": 3.965517241379311e-05, | |
| "loss": 0.8848, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 3.4340436458587646, | |
| "learning_rate": 3.8793103448275865e-05, | |
| "loss": 0.8131, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 8.15261459350586, | |
| "learning_rate": 3.793103448275862e-05, | |
| "loss": 0.7079, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5813953488372093, | |
| "eval_f1_macro": 0.3896430367018602, | |
| "eval_f1_micro": 0.5813953488372093, | |
| "eval_f1_weighted": 0.46328060144749605, | |
| "eval_loss": 0.8947027325630188, | |
| "eval_precision_macro": 0.3952702702702703, | |
| "eval_precision_micro": 0.5813953488372093, | |
| "eval_precision_weighted": 0.40681961030798236, | |
| "eval_recall_macro": 0.425, | |
| "eval_recall_micro": 0.5813953488372093, | |
| "eval_recall_weighted": 0.5813953488372093, | |
| "eval_runtime": 0.4917, | |
| "eval_samples_per_second": 174.9, | |
| "eval_steps_per_second": 12.202, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 11.300780296325684, | |
| "learning_rate": 3.7068965517241385e-05, | |
| "loss": 0.9546, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 6.606018543243408, | |
| "learning_rate": 3.620689655172414e-05, | |
| "loss": 0.8938, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 4.41719913482666, | |
| "learning_rate": 3.53448275862069e-05, | |
| "loss": 0.6425, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 10.717499732971191, | |
| "learning_rate": 3.4482758620689657e-05, | |
| "loss": 0.5271, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 3.899660348892212, | |
| "learning_rate": 3.3620689655172414e-05, | |
| "loss": 0.7768, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 5.042697906494141, | |
| "learning_rate": 3.275862068965517e-05, | |
| "loss": 0.6497, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 6.32042121887207, | |
| "learning_rate": 3.1896551724137935e-05, | |
| "loss": 0.7292, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 6.479010105133057, | |
| "learning_rate": 3.103448275862069e-05, | |
| "loss": 0.6227, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 8.43276596069336, | |
| "learning_rate": 3.017241379310345e-05, | |
| "loss": 0.6197, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 6.091739654541016, | |
| "learning_rate": 2.9310344827586206e-05, | |
| "loss": 0.4838, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 5.038337230682373, | |
| "learning_rate": 2.844827586206897e-05, | |
| "loss": 0.5271, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 6.296905517578125, | |
| "learning_rate": 2.7586206896551727e-05, | |
| "loss": 0.7801, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 9.464910507202148, | |
| "learning_rate": 2.672413793103448e-05, | |
| "loss": 0.6315, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 10.5482177734375, | |
| "learning_rate": 2.5862068965517244e-05, | |
| "loss": 1.1036, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 6.3730645179748535, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.8945, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 9.688776969909668, | |
| "learning_rate": 2.413793103448276e-05, | |
| "loss": 0.7347, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 10.05594253540039, | |
| "learning_rate": 2.327586206896552e-05, | |
| "loss": 0.7423, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 8.176011085510254, | |
| "learning_rate": 2.2413793103448276e-05, | |
| "loss": 0.7391, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 10.692707061767578, | |
| "learning_rate": 2.1551724137931033e-05, | |
| "loss": 0.5511, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 11.206247329711914, | |
| "learning_rate": 2.0689655172413793e-05, | |
| "loss": 0.6857, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 5.92320442199707, | |
| "learning_rate": 1.9827586206896554e-05, | |
| "loss": 0.5739, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 7.509181976318359, | |
| "learning_rate": 1.896551724137931e-05, | |
| "loss": 0.6399, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6627906976744186, | |
| "eval_f1_macro": 0.5486309523809524, | |
| "eval_f1_micro": 0.6627906976744186, | |
| "eval_f1_weighted": 0.6009828349944628, | |
| "eval_loss": 0.7699998617172241, | |
| "eval_precision_macro": 0.8307971014492754, | |
| "eval_precision_micro": 0.6627906976744186, | |
| "eval_precision_weighted": 0.7592686215032018, | |
| "eval_recall_macro": 0.5194444444444445, | |
| "eval_recall_micro": 0.6627906976744186, | |
| "eval_recall_weighted": 0.6627906976744186, | |
| "eval_runtime": 0.4513, | |
| "eval_samples_per_second": 190.547, | |
| "eval_steps_per_second": 13.294, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 6.413642406463623, | |
| "learning_rate": 1.810344827586207e-05, | |
| "loss": 0.6215, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 7.694431781768799, | |
| "learning_rate": 1.7241379310344828e-05, | |
| "loss": 0.6688, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 6.910383224487305, | |
| "learning_rate": 1.6379310344827585e-05, | |
| "loss": 0.5048, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 6.3850579261779785, | |
| "learning_rate": 1.5517241379310346e-05, | |
| "loss": 0.7519, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 9.831122398376465, | |
| "learning_rate": 1.4655172413793103e-05, | |
| "loss": 0.7617, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 9.89468765258789, | |
| "learning_rate": 1.3793103448275863e-05, | |
| "loss": 0.9211, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 3.081348180770874, | |
| "learning_rate": 1.2931034482758622e-05, | |
| "loss": 0.2957, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 13.976354598999023, | |
| "learning_rate": 1.206896551724138e-05, | |
| "loss": 0.6079, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 14.941506385803223, | |
| "learning_rate": 1.163793103448276e-05, | |
| "loss": 0.7309, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 8.57857894897461, | |
| "learning_rate": 1.0775862068965516e-05, | |
| "loss": 0.763, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 4.113162040710449, | |
| "learning_rate": 9.913793103448277e-06, | |
| "loss": 0.3743, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 10.615869522094727, | |
| "learning_rate": 9.051724137931036e-06, | |
| "loss": 0.6419, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 6.8312530517578125, | |
| "learning_rate": 8.189655172413793e-06, | |
| "loss": 0.493, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 4.930009841918945, | |
| "learning_rate": 7.3275862068965514e-06, | |
| "loss": 0.578, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 4.636520862579346, | |
| "learning_rate": 6.465517241379311e-06, | |
| "loss": 0.5522, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 6.69807243347168, | |
| "learning_rate": 5.603448275862069e-06, | |
| "loss": 0.513, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 15.026963233947754, | |
| "learning_rate": 4.741379310344828e-06, | |
| "loss": 1.2114, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 5.072807312011719, | |
| "learning_rate": 3.8793103448275865e-06, | |
| "loss": 0.5483, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 5.802096366882324, | |
| "learning_rate": 3.017241379310345e-06, | |
| "loss": 0.6085, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "grad_norm": 8.00621223449707, | |
| "learning_rate": 2.1551724137931035e-06, | |
| "loss": 0.6149, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 6.807849884033203, | |
| "learning_rate": 1.293103448275862e-06, | |
| "loss": 0.3146, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7325581395348837, | |
| "eval_f1_macro": 0.6642156862745098, | |
| "eval_f1_micro": 0.7325581395348836, | |
| "eval_f1_weighted": 0.7024623803009576, | |
| "eval_loss": 0.6135068535804749, | |
| "eval_precision_macro": 0.8029661016949152, | |
| "eval_precision_micro": 0.7325581395348837, | |
| "eval_precision_weighted": 0.762908947575877, | |
| "eval_recall_macro": 0.630297157622739, | |
| "eval_recall_micro": 0.7325581395348837, | |
| "eval_recall_weighted": 0.7325581395348837, | |
| "eval_runtime": 0.457, | |
| "eval_samples_per_second": 188.186, | |
| "eval_steps_per_second": 13.129, | |
| "step": 129 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 129, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 7.97406866688983e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |