| { | |
| "best_metric": 0.8736686110496521, | |
| "best_model_checkpoint": "data/Mistral-7B_task-3_60-samples_config-2_full/checkpoint-28", | |
| "epoch": 16.695652173913043, | |
| "eval_steps": 500, | |
| "global_step": 48, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 1.0153508186340332, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2657, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.9736206531524658, | |
| "learning_rate": 2e-05, | |
| "loss": 1.2329, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 1.227666974067688, | |
| "eval_runtime": 13.8044, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.853352427482605, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2254, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 1.1423181295394897, | |
| "eval_runtime": 13.8079, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.5523448586463928, | |
| "learning_rate": 6e-05, | |
| "loss": 1.1766, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.41505715250968933, | |
| "learning_rate": 8e-05, | |
| "loss": 1.0984, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 1.0653983354568481, | |
| "eval_runtime": 13.8066, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.377344012260437, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0266, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 1.0051709413528442, | |
| "eval_runtime": 13.8011, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.37405651807785034, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.0095, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.5535106062889099, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.948, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.9384298920631409, | |
| "eval_runtime": 13.8031, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.33964699506759644, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.884, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.9019761681556702, | |
| "eval_runtime": 13.8018, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.2418847680091858, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.8502, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.21523350477218628, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.8253, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.8885052800178528, | |
| "eval_runtime": 13.8026, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.19641552865505219, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.822, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8788420557975769, | |
| "eval_runtime": 13.8056, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.18110164999961853, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.7994, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 0.8753540515899658, | |
| "eval_runtime": 13.8028, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.18521583080291748, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.779, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.20691977441310883, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.7669, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 0.8736686110496521, | |
| "eval_runtime": 13.804, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.20081867277622223, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.7596, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 0.873776376247406, | |
| "eval_runtime": 13.8051, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.2356591522693634, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.7106, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.2318907231092453, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.7256, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 0.8750200867652893, | |
| "eval_runtime": 13.8049, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.24235422909259796, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.6803, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.8834443688392639, | |
| "eval_runtime": 13.8008, | |
| "eval_samples_per_second": 0.87, | |
| "eval_steps_per_second": 0.87, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.2748987674713135, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.682, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.2720317542552948, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.6584, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 0.8950894474983215, | |
| "eval_runtime": 13.8031, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.3152175545692444, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.6131, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 0.902263343334198, | |
| "eval_runtime": 13.8028, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.38017743825912476, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.6056, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.8462388515472412, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.5777, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.9154355525970459, | |
| "eval_runtime": 13.8024, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "grad_norm": 0.7009825110435486, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.5158, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "eval_loss": 0.9439125061035156, | |
| "eval_runtime": 13.8015, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 16.695652173913043, | |
| "step": 48, | |
| "total_flos": 7.706943345288806e+16, | |
| "train_loss": 0.8436332593361536, | |
| "train_runtime": 2874.4335, | |
| "train_samples_per_second": 0.8, | |
| "train_steps_per_second": 0.035 | |
| } | |
| ], | |
| "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": 7.706943345288806e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |