| { | |
| "best_metric": 0.37516140937805176, | |
| "best_model_checkpoint": "data/Mistral-7B_task-3_60-samples_config-2/checkpoint-20", | |
| "epoch": 13.91304347826087, | |
| "eval_steps": 500, | |
| "global_step": 40, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 17.986906051635742, | |
| "learning_rate": 1e-05, | |
| "loss": 3.8263, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 16.079301834106445, | |
| "learning_rate": 2e-05, | |
| "loss": 3.6443, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 3.4455251693725586, | |
| "eval_runtime": 13.8084, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 14.32006549835205, | |
| "learning_rate": 4e-05, | |
| "loss": 3.1998, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 1.8330297470092773, | |
| "eval_runtime": 13.8064, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 6.900447845458984, | |
| "learning_rate": 6e-05, | |
| "loss": 2.0772, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 4.268274784088135, | |
| "learning_rate": 8e-05, | |
| "loss": 1.1687, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 0.7475635409355164, | |
| "eval_runtime": 13.8049, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 1.7821996212005615, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7861, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 0.5044365525245667, | |
| "eval_runtime": 13.806, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.9720113277435303, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.4289, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 1.0473393201828003, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.3983, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.4259566068649292, | |
| "eval_runtime": 13.8111, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.6317740678787231, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.359, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.38502585887908936, | |
| "eval_runtime": 13.8077, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.6458030343055725, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.2136, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.5151018500328064, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.2531, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.37516140937805176, | |
| "eval_runtime": 13.8069, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.46064192056655884, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.1449, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.4618639051914215, | |
| "eval_runtime": 13.8049, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.44753408432006836, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.1711, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 0.5145867466926575, | |
| "eval_runtime": 13.8075, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.706068754196167, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.1349, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.6422617435455322, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0704, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 0.6075521111488342, | |
| "eval_runtime": 13.8148, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.45618292689323425, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0479, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 0.717354953289032, | |
| "eval_runtime": 13.804, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.6182390451431274, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.0287, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.749740481376648, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0096, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 0.8061732649803162, | |
| "eval_runtime": 13.8091, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.5463645458221436, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0056, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.876629650592804, | |
| "eval_runtime": 13.8082, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.4603806734085083, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.0034, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.5449416041374207, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0024, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 0.9099900722503662, | |
| "eval_runtime": 13.8053, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "step": 40, | |
| "total_flos": 6.346894531887104e+16, | |
| "train_loss": 0.6619909038650803, | |
| "train_runtime": 2367.3473, | |
| "train_samples_per_second": 0.972, | |
| "train_steps_per_second": 0.042 | |
| } | |
| ], | |
| "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": 6.346894531887104e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |