| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.569611072540283, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5438, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7218045112781954, | |
| "eval_f1": 0.6545993371027491, | |
| "eval_loss": 0.49882158637046814, | |
| "eval_precision": 0.6600553802562947, | |
| "eval_recall": 0.6506637570467357, | |
| "eval_runtime": 5.2428, | |
| "eval_samples_per_second": 76.104, | |
| "eval_steps_per_second": 9.537, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.2452878952026367, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4428, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8161454307628278, | |
| "eval_loss": 0.3788329064846039, | |
| "eval_precision": 0.8107299270072992, | |
| "eval_recall": 0.8225586470267321, | |
| "eval_runtime": 5.1661, | |
| "eval_samples_per_second": 77.234, | |
| "eval_steps_per_second": 9.679, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.797173500061035, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3441, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8179269882659713, | |
| "eval_loss": 0.3289283514022827, | |
| "eval_precision": 0.8510239760239761, | |
| "eval_recall": 0.7981905801054737, | |
| "eval_runtime": 5.1072, | |
| "eval_samples_per_second": 78.125, | |
| "eval_steps_per_second": 9.79, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.9981327056884766, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2986, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8533986527862829, | |
| "eval_loss": 0.28838610649108887, | |
| "eval_precision": 0.8572003218020917, | |
| "eval_recall": 0.8498817966903074, | |
| "eval_runtime": 5.1209, | |
| "eval_samples_per_second": 77.916, | |
| "eval_steps_per_second": 9.764, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.763756513595581, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2667, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.850729517396184, | |
| "eval_loss": 0.26981133222579956, | |
| "eval_precision": 0.8535087719298247, | |
| "eval_recall": 0.8481087470449173, | |
| "eval_runtime": 5.1232, | |
| "eval_samples_per_second": 77.881, | |
| "eval_steps_per_second": 9.759, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.7370102405548096, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2524, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8609292598654301, | |
| "eval_loss": 0.27233511209487915, | |
| "eval_precision": 0.8609292598654301, | |
| "eval_recall": 0.8609292598654301, | |
| "eval_runtime": 5.1586, | |
| "eval_samples_per_second": 77.347, | |
| "eval_steps_per_second": 9.693, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.2413272857666016, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2343, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8265664241097301, | |
| "eval_loss": 0.3179858326911926, | |
| "eval_precision": 0.8532894736842105, | |
| "eval_recall": 0.8092380432805966, | |
| "eval_runtime": 5.1498, | |
| "eval_samples_per_second": 77.479, | |
| "eval_steps_per_second": 9.709, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.469871997833252, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2212, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.852937255424767, | |
| "eval_loss": 0.29489362239837646, | |
| "eval_precision": 0.8674217731421121, | |
| "eval_recall": 0.8416530278232406, | |
| "eval_runtime": 5.155, | |
| "eval_samples_per_second": 77.4, | |
| "eval_steps_per_second": 9.699, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.085997581481934, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2142, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8564658408408408, | |
| "eval_loss": 0.2828481197357178, | |
| "eval_precision": 0.8697278911564625, | |
| "eval_recall": 0.8459265320967448, | |
| "eval_runtime": 5.1873, | |
| "eval_samples_per_second": 76.919, | |
| "eval_steps_per_second": 9.639, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.2397539615631104, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1958, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8457993935430168, | |
| "eval_loss": 0.28871509432792664, | |
| "eval_precision": 0.8398540145985401, | |
| "eval_recall": 0.8527914166212038, | |
| "eval_runtime": 5.1373, | |
| "eval_samples_per_second": 77.667, | |
| "eval_steps_per_second": 9.733, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.4002835750579834, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1855, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8602993213495533, | |
| "eval_loss": 0.2867955267429352, | |
| "eval_precision": 0.8547653958944281, | |
| "eval_recall": 0.8666575741043827, | |
| "eval_runtime": 5.1317, | |
| "eval_samples_per_second": 77.752, | |
| "eval_steps_per_second": 9.743, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.633034348487854, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1742, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8448388501742161, | |
| "eval_loss": 0.29811105132102966, | |
| "eval_precision": 0.8551721930610677, | |
| "eval_recall": 0.8363338788870704, | |
| "eval_runtime": 5.1494, | |
| "eval_samples_per_second": 77.484, | |
| "eval_steps_per_second": 9.71, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 2.338294506072998, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1601, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8556004584112431, | |
| "eval_loss": 0.29304954409599304, | |
| "eval_precision": 0.8538865546218487, | |
| "eval_recall": 0.85738316057465, | |
| "eval_runtime": 5.0982, | |
| "eval_samples_per_second": 78.263, | |
| "eval_steps_per_second": 9.807, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.3382197618484497, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1602, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8589607635206786, | |
| "eval_loss": 0.29793980717658997, | |
| "eval_precision": 0.8503875968992248, | |
| "eval_recall": 0.8698854337152209, | |
| "eval_runtime": 5.1508, | |
| "eval_samples_per_second": 77.464, | |
| "eval_steps_per_second": 9.707, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.5800030827522278, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1497, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8662440310793597, | |
| "eval_loss": 0.29690659046173096, | |
| "eval_precision": 0.8606158357771261, | |
| "eval_recall": 0.872704128023277, | |
| "eval_runtime": 5.1461, | |
| "eval_samples_per_second": 77.534, | |
| "eval_steps_per_second": 9.716, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 6.755856037139893, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1447, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8847117794486216, | |
| "eval_f1": 0.8616171059774413, | |
| "eval_loss": 0.29627636075019836, | |
| "eval_precision": 0.859873949579832, | |
| "eval_recall": 0.8634297144935443, | |
| "eval_runtime": 5.1639, | |
| "eval_samples_per_second": 77.267, | |
| "eval_steps_per_second": 9.683, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 2.1538662910461426, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1394, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8589543987905864, | |
| "eval_loss": 0.30176377296447754, | |
| "eval_precision": 0.8564068100358423, | |
| "eval_recall": 0.8616566648481543, | |
| "eval_runtime": 5.17, | |
| "eval_samples_per_second": 77.176, | |
| "eval_steps_per_second": 9.671, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 2.54630708694458, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1333, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.858259325044405, | |
| "eval_loss": 0.30650317668914795, | |
| "eval_precision": 0.8573798178418481, | |
| "eval_recall": 0.8591562102200401, | |
| "eval_runtime": 5.128, | |
| "eval_samples_per_second": 77.807, | |
| "eval_steps_per_second": 9.75, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.7696042656898499, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1406, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.858259325044405, | |
| "eval_loss": 0.30623340606689453, | |
| "eval_precision": 0.8573798178418481, | |
| "eval_recall": 0.8591562102200401, | |
| "eval_runtime": 5.1122, | |
| "eval_samples_per_second": 78.049, | |
| "eval_steps_per_second": 9.781, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.025254487991333, | |
| "learning_rate": 0.0, | |
| "loss": 0.1243, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.858259325044405, | |
| "eval_loss": 0.30716511607170105, | |
| "eval_precision": 0.8573798178418481, | |
| "eval_recall": 0.8591562102200401, | |
| "eval_runtime": 5.0928, | |
| "eval_samples_per_second": 78.347, | |
| "eval_steps_per_second": 9.818, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8460375995160000.0, | |
| "train_loss": 0.22629446436147221, | |
| "train_runtime": 1956.0503, | |
| "train_samples_per_second": 37.269, | |
| "train_steps_per_second": 1.247 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8460375995160000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |