| { | |
| "best_metric": 0.7544093132019043, | |
| "best_model_checkpoint": "xblock-social-screenshots-3/checkpoint-1245", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1245, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 13.465712547302246, | |
| "learning_rate": 7.2e-06, | |
| "loss": 2.4572, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 9.6524076461792, | |
| "learning_rate": 1.52e-05, | |
| "loss": 2.079, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 9.261948585510254, | |
| "learning_rate": 2.32e-05, | |
| "loss": 1.7089, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 9.193083763122559, | |
| "learning_rate": 3.12e-05, | |
| "loss": 1.4402, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 7.320693492889404, | |
| "learning_rate": 3.9200000000000004e-05, | |
| "loss": 1.407, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 7.089232444763184, | |
| "learning_rate": 4.72e-05, | |
| "loss": 1.2287, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 9.907115936279297, | |
| "learning_rate": 4.941964285714286e-05, | |
| "loss": 1.2205, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 6.660067081451416, | |
| "learning_rate": 4.852678571428572e-05, | |
| "loss": 1.0867, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 8.426629066467285, | |
| "learning_rate": 4.7633928571428573e-05, | |
| "loss": 1.097, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 9.229801177978516, | |
| "learning_rate": 4.674107142857143e-05, | |
| "loss": 1.4158, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 6.415766716003418, | |
| "learning_rate": 4.584821428571429e-05, | |
| "loss": 1.2353, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 9.118013381958008, | |
| "learning_rate": 4.4955357142857144e-05, | |
| "loss": 1.0926, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 8.883289337158203, | |
| "learning_rate": 4.40625e-05, | |
| "loss": 1.1642, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 6.868738651275635, | |
| "learning_rate": 4.3169642857142864e-05, | |
| "loss": 1.0041, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 2.424619436264038, | |
| "learning_rate": 4.2276785714285714e-05, | |
| "loss": 1.1074, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 10.195758819580078, | |
| "learning_rate": 4.138392857142857e-05, | |
| "loss": 0.8968, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 8.857401847839355, | |
| "learning_rate": 4.0491071428571434e-05, | |
| "loss": 1.2484, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 7.8076605796813965, | |
| "learning_rate": 3.9598214285714284e-05, | |
| "loss": 0.998, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 6.418033599853516, | |
| "learning_rate": 3.870535714285715e-05, | |
| "loss": 0.9212, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 7.953289985656738, | |
| "learning_rate": 3.78125e-05, | |
| "loss": 1.1387, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7156626506024096, | |
| "eval_f1_macro": 0.24337876592111682, | |
| "eval_f1_micro": 0.7156626506024096, | |
| "eval_f1_weighted": 0.6701704805513728, | |
| "eval_loss": 0.9069176912307739, | |
| "eval_precision_macro": 0.28735578588970384, | |
| "eval_precision_micro": 0.7156626506024096, | |
| "eval_precision_weighted": 0.6398455360285946, | |
| "eval_recall_macro": 0.2455658658820232, | |
| "eval_recall_micro": 0.7156626506024096, | |
| "eval_recall_weighted": 0.7156626506024096, | |
| "eval_runtime": 32.7637, | |
| "eval_samples_per_second": 25.333, | |
| "eval_steps_per_second": 1.587, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 10.716949462890625, | |
| "learning_rate": 3.691964285714286e-05, | |
| "loss": 0.974, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 6.424170970916748, | |
| "learning_rate": 3.602678571428572e-05, | |
| "loss": 0.9156, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 5.7230000495910645, | |
| "learning_rate": 3.513392857142857e-05, | |
| "loss": 1.0882, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 6.781800746917725, | |
| "learning_rate": 3.424107142857143e-05, | |
| "loss": 0.8575, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 7.278835296630859, | |
| "learning_rate": 3.334821428571429e-05, | |
| "loss": 0.8842, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 8.899985313415527, | |
| "learning_rate": 3.2455357142857145e-05, | |
| "loss": 1.0324, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 10.547853469848633, | |
| "learning_rate": 3.15625e-05, | |
| "loss": 0.9993, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 5.870782852172852, | |
| "learning_rate": 3.066964285714286e-05, | |
| "loss": 1.0766, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 7.413377285003662, | |
| "learning_rate": 2.9776785714285715e-05, | |
| "loss": 0.8224, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 8.232254981994629, | |
| "learning_rate": 2.888392857142857e-05, | |
| "loss": 0.8699, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 7.721341609954834, | |
| "learning_rate": 2.7991071428571432e-05, | |
| "loss": 1.0413, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 10.697331428527832, | |
| "learning_rate": 2.7098214285714285e-05, | |
| "loss": 0.995, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 5.213534355163574, | |
| "learning_rate": 2.6205357142857145e-05, | |
| "loss": 1.1162, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 4.949966907501221, | |
| "learning_rate": 2.53125e-05, | |
| "loss": 0.7447, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 8.088377952575684, | |
| "learning_rate": 2.441964285714286e-05, | |
| "loss": 0.7466, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 6.961575984954834, | |
| "learning_rate": 2.3526785714285715e-05, | |
| "loss": 0.8619, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 12.139084815979004, | |
| "learning_rate": 2.2633928571428572e-05, | |
| "loss": 0.6702, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 4.518425941467285, | |
| "learning_rate": 2.174107142857143e-05, | |
| "loss": 0.8649, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 7.617171764373779, | |
| "learning_rate": 2.084821428571429e-05, | |
| "loss": 0.7682, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 9.569235801696777, | |
| "learning_rate": 1.9955357142857146e-05, | |
| "loss": 0.9026, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 8.944718360900879, | |
| "learning_rate": 1.90625e-05, | |
| "loss": 0.8327, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.744578313253012, | |
| "eval_f1_macro": 0.328500713177248, | |
| "eval_f1_micro": 0.744578313253012, | |
| "eval_f1_weighted": 0.7121093039405768, | |
| "eval_loss": 0.7984326481819153, | |
| "eval_precision_macro": 0.4303939410339066, | |
| "eval_precision_micro": 0.744578313253012, | |
| "eval_precision_weighted": 0.7177621227889628, | |
| "eval_recall_macro": 0.30809268775929033, | |
| "eval_recall_micro": 0.744578313253012, | |
| "eval_recall_weighted": 0.744578313253012, | |
| "eval_runtime": 30.47, | |
| "eval_samples_per_second": 27.24, | |
| "eval_steps_per_second": 1.707, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 6.886695861816406, | |
| "learning_rate": 1.8169642857142856e-05, | |
| "loss": 0.8599, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 3.803279399871826, | |
| "learning_rate": 1.7276785714285716e-05, | |
| "loss": 0.7697, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 8.693037033081055, | |
| "learning_rate": 1.6383928571428573e-05, | |
| "loss": 0.7916, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 10.645078659057617, | |
| "learning_rate": 1.549107142857143e-05, | |
| "loss": 0.6787, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 6.220460414886475, | |
| "learning_rate": 1.4598214285714288e-05, | |
| "loss": 0.7976, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 10.7758150100708, | |
| "learning_rate": 1.3705357142857145e-05, | |
| "loss": 0.6825, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 11.92375373840332, | |
| "learning_rate": 1.28125e-05, | |
| "loss": 0.853, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 8.55823040008545, | |
| "learning_rate": 1.1919642857142858e-05, | |
| "loss": 0.7647, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 10.385089874267578, | |
| "learning_rate": 1.1026785714285715e-05, | |
| "loss": 0.7351, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 6.002354621887207, | |
| "learning_rate": 1.0133928571428572e-05, | |
| "loss": 0.8779, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 7.847404956817627, | |
| "learning_rate": 9.24107142857143e-06, | |
| "loss": 0.7776, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 5.796429634094238, | |
| "learning_rate": 8.348214285714285e-06, | |
| "loss": 0.743, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 8.282535552978516, | |
| "learning_rate": 7.455357142857143e-06, | |
| "loss": 0.8694, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 9.527463912963867, | |
| "learning_rate": 6.5625e-06, | |
| "loss": 0.7111, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 7.947423458099365, | |
| "learning_rate": 5.669642857142857e-06, | |
| "loss": 0.9326, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 8.990558624267578, | |
| "learning_rate": 4.776785714285715e-06, | |
| "loss": 0.7542, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 8.609009742736816, | |
| "learning_rate": 3.883928571428571e-06, | |
| "loss": 0.7247, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 3.701921224594116, | |
| "learning_rate": 2.991071428571429e-06, | |
| "loss": 0.6855, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 6.839054584503174, | |
| "learning_rate": 2.0982142857142857e-06, | |
| "loss": 0.7191, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "grad_norm": 5.237525939941406, | |
| "learning_rate": 1.205357142857143e-06, | |
| "loss": 0.592, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 6.806934356689453, | |
| "learning_rate": 3.5714285714285716e-07, | |
| "loss": 0.7655, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7626506024096386, | |
| "eval_f1_macro": 0.34243943617676403, | |
| "eval_f1_micro": 0.7626506024096386, | |
| "eval_f1_weighted": 0.7295452326605777, | |
| "eval_loss": 0.7544093132019043, | |
| "eval_precision_macro": 0.5376629516266399, | |
| "eval_precision_micro": 0.7626506024096386, | |
| "eval_precision_weighted": 0.7581766519951105, | |
| "eval_recall_macro": 0.3164834046818783, | |
| "eval_recall_micro": 0.7626506024096386, | |
| "eval_recall_weighted": 0.7626506024096386, | |
| "eval_runtime": 30.5059, | |
| "eval_samples_per_second": 27.208, | |
| "eval_steps_per_second": 1.705, | |
| "step": 1245 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 1245, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 7.702550117363405e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |