| { | |
| "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": 5.332692623138428, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5551, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7167919799498746, | |
| "eval_f1": 0.6337397746565828, | |
| "eval_loss": 0.49750953912734985, | |
| "eval_precision": 0.6484917150545249, | |
| "eval_recall": 0.6271140207310419, | |
| "eval_runtime": 1.7156, | |
| "eval_samples_per_second": 232.578, | |
| "eval_steps_per_second": 29.145, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.58549690246582, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4753, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7418546365914787, | |
| "eval_f1": 0.7171128258430446, | |
| "eval_loss": 0.468201607465744, | |
| "eval_precision": 0.7117669953295278, | |
| "eval_recall": 0.747363156937625, | |
| "eval_runtime": 1.7118, | |
| "eval_samples_per_second": 233.092, | |
| "eval_steps_per_second": 29.21, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.9045073986053467, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.412, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.7803582113367107, | |
| "eval_loss": 0.38815850019454956, | |
| "eval_precision": 0.7993592785951591, | |
| "eval_recall": 0.7676395708310602, | |
| "eval_runtime": 1.7097, | |
| "eval_samples_per_second": 233.375, | |
| "eval_steps_per_second": 29.245, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.453057289123535, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3498, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8421052631578947, | |
| "eval_f1": 0.8090546545634092, | |
| "eval_loss": 0.3690779209136963, | |
| "eval_precision": 0.8098422078713293, | |
| "eval_recall": 0.808283324240771, | |
| "eval_runtime": 1.7281, | |
| "eval_samples_per_second": 230.885, | |
| "eval_steps_per_second": 28.933, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.748023986816406, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3361, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8145363408521303, | |
| "eval_f1": 0.7897487752079297, | |
| "eval_loss": 0.3794577121734619, | |
| "eval_precision": 0.7788623404833017, | |
| "eval_recall": 0.8112838697945082, | |
| "eval_runtime": 1.7098, | |
| "eval_samples_per_second": 233.358, | |
| "eval_steps_per_second": 29.243, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.2061963081359863, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3081, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7969924812030075, | |
| "eval_f1": 0.7760554889583348, | |
| "eval_loss": 0.4142220616340637, | |
| "eval_precision": 0.7665010460251046, | |
| "eval_recall": 0.8088743407892345, | |
| "eval_runtime": 1.7108, | |
| "eval_samples_per_second": 233.229, | |
| "eval_steps_per_second": 29.227, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.4639934301376343, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2918, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8234833375639119, | |
| "eval_loss": 0.3555023968219757, | |
| "eval_precision": 0.8130172220979647, | |
| "eval_recall": 0.8393344244408074, | |
| "eval_runtime": 1.7094, | |
| "eval_samples_per_second": 233.421, | |
| "eval_steps_per_second": 29.251, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.498388767242432, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2739, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8315033783783784, | |
| "eval_loss": 0.33167070150375366, | |
| "eval_precision": 0.8438775510204082, | |
| "eval_recall": 0.8217403164211674, | |
| "eval_runtime": 1.7093, | |
| "eval_samples_per_second": 233.426, | |
| "eval_steps_per_second": 29.251, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 10.206114768981934, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2586, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8336296003686878, | |
| "eval_loss": 0.35962939262390137, | |
| "eval_precision": 0.8243105209397344, | |
| "eval_recall": 0.8464266230223677, | |
| "eval_runtime": 1.7093, | |
| "eval_samples_per_second": 233.426, | |
| "eval_steps_per_second": 29.251, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.042139530181885, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2509, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8349466368826011, | |
| "eval_loss": 0.329887330532074, | |
| "eval_precision": 0.8325716845878136, | |
| "eval_recall": 0.8374704491725768, | |
| "eval_runtime": 1.7094, | |
| "eval_samples_per_second": 233.418, | |
| "eval_steps_per_second": 29.25, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 8.531843185424805, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2468, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8401647707947546, | |
| "eval_loss": 0.32244208455085754, | |
| "eval_precision": 0.8393298751432535, | |
| "eval_recall": 0.8410165484633569, | |
| "eval_runtime": 1.7111, | |
| "eval_samples_per_second": 233.186, | |
| "eval_steps_per_second": 29.221, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 8.214770317077637, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2372, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8313646980313647, | |
| "eval_loss": 0.32940539717674255, | |
| "eval_precision": 0.825091575091575, | |
| "eval_recall": 0.8389252591380251, | |
| "eval_runtime": 1.7104, | |
| "eval_samples_per_second": 233.277, | |
| "eval_steps_per_second": 29.233, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 3.748082160949707, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2305, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8411818738518064, | |
| "eval_loss": 0.31342318654060364, | |
| "eval_precision": 0.8448542607834644, | |
| "eval_recall": 0.8377886888525186, | |
| "eval_runtime": 1.7084, | |
| "eval_samples_per_second": 233.547, | |
| "eval_steps_per_second": 29.267, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 14.465522766113281, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2249, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8457993935430168, | |
| "eval_loss": 0.32253342866897583, | |
| "eval_precision": 0.8398540145985401, | |
| "eval_recall": 0.8527914166212038, | |
| "eval_runtime": 1.7086, | |
| "eval_samples_per_second": 233.518, | |
| "eval_steps_per_second": 29.263, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.3240478038787842, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2193, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8503151260504201, | |
| "eval_loss": 0.3187616467475891, | |
| "eval_precision": 0.8470628455912955, | |
| "eval_recall": 0.8538370612838698, | |
| "eval_runtime": 1.7079, | |
| "eval_samples_per_second": 233.622, | |
| "eval_steps_per_second": 29.276, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 5.576813697814941, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2061, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8277920424868874, | |
| "eval_loss": 0.3392154574394226, | |
| "eval_precision": 0.8186499829758257, | |
| "eval_recall": 0.8403800691034733, | |
| "eval_runtime": 1.7081, | |
| "eval_samples_per_second": 233.595, | |
| "eval_steps_per_second": 29.273, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 9.40192985534668, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.21, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8541488972828073, | |
| "eval_loss": 0.31219902634620667, | |
| "eval_precision": 0.8559859154929578, | |
| "eval_recall": 0.8523822513184216, | |
| "eval_runtime": 1.7119, | |
| "eval_samples_per_second": 233.08, | |
| "eval_steps_per_second": 29.208, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 7.254611492156982, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2112, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8303372434017595, | |
| "eval_loss": 0.3332211673259735, | |
| "eval_precision": 0.8215506437279252, | |
| "eval_recall": 0.8421531187488634, | |
| "eval_runtime": 1.7073, | |
| "eval_samples_per_second": 233.698, | |
| "eval_steps_per_second": 29.285, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.143869876861572, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2002, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8514869535493182, | |
| "eval_loss": 0.3120571970939636, | |
| "eval_precision": 0.8523821128305106, | |
| "eval_recall": 0.8506092016730314, | |
| "eval_runtime": 1.7097, | |
| "eval_samples_per_second": 233.375, | |
| "eval_steps_per_second": 29.245, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.309845924377441, | |
| "learning_rate": 0.0, | |
| "loss": 0.2041, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8495838108450449, | |
| "eval_loss": 0.3129415214061737, | |
| "eval_precision": 0.8478991596638655, | |
| "eval_recall": 0.8513366066557555, | |
| "eval_runtime": 1.7073, | |
| "eval_samples_per_second": 233.696, | |
| "eval_steps_per_second": 29.285, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7662265464912000.0, | |
| "train_loss": 0.28510660187142794, | |
| "train_runtime": 615.9137, | |
| "train_samples_per_second": 118.133, | |
| "train_steps_per_second": 3.962 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7662265464912000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |