| { | |
| "best_metric": 0.8033270239830017, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_60-samples_config-2_full/checkpoint-31", | |
| "epoch": 17.73913043478261, | |
| "eval_steps": 500, | |
| "global_step": 51, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 1.1966744661331177, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2051, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 1.0429316759109497, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1698, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 1.1608541011810303, | |
| "eval_runtime": 16.1427, | |
| "eval_samples_per_second": 0.743, | |
| "eval_steps_per_second": 0.743, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.9423496723175049, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1595, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 1.0642138719558716, | |
| "eval_runtime": 16.133, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.6271457076072693, | |
| "learning_rate": 6e-05, | |
| "loss": 1.1013, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.38807761669158936, | |
| "learning_rate": 8e-05, | |
| "loss": 1.0152, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 0.9874032139778137, | |
| "eval_runtime": 16.1254, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.3477388620376587, | |
| "learning_rate": 0.0001, | |
| "loss": 0.9495, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 0.9347397685050964, | |
| "eval_runtime": 16.1105, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.36352431774139404, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.937, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.48189738392829895, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.8878, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.8747708201408386, | |
| "eval_runtime": 16.1079, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.29127490520477295, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.823, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.8342945575714111, | |
| "eval_runtime": 16.1078, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.22186493873596191, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.7857, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.17112106084823608, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.7669, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.8199789524078369, | |
| "eval_runtime": 16.1152, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.1760251373052597, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.7616, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8117280006408691, | |
| "eval_runtime": 16.1097, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.15678618848323822, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.7438, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 0.8073273301124573, | |
| "eval_runtime": 16.1117, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.16403785347938538, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.7263, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.16462087631225586, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.7119, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 0.803439199924469, | |
| "eval_runtime": 16.1126, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.16713577508926392, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.7076, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 0.8033270239830017, | |
| "eval_runtime": 16.1138, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.19175440073013306, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.6671, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.19275148212909698, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.6789, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 0.8045647740364075, | |
| "eval_runtime": 16.1085, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.2089928835630417, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.6409, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.8046860694885254, | |
| "eval_runtime": 16.1081, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.24371260404586792, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.6434, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.24846789240837097, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.625, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 0.8139411807060242, | |
| "eval_runtime": 16.1116, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.25238463282585144, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.581, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 0.8231260776519775, | |
| "eval_runtime": 16.109, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.26100218296051025, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.5833, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.44007062911987305, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.5582, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.8310089707374573, | |
| "eval_runtime": 16.1053, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "grad_norm": 0.450304239988327, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.5045, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "eval_loss": 0.8541346192359924, | |
| "eval_runtime": 16.106, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 17.391304347826086, | |
| "grad_norm": 0.344973087310791, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.507, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 17.73913043478261, | |
| "eval_loss": 0.8567993640899658, | |
| "eval_runtime": 16.1067, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 17.73913043478261, | |
| "step": 51, | |
| "total_flos": 9.91420990137303e+16, | |
| "train_loss": 0.7725417123121374, | |
| "train_runtime": 3617.3626, | |
| "train_samples_per_second": 0.636, | |
| "train_steps_per_second": 0.028 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.91420990137303e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |