| { | |
| "best_metric": 0.8306775689125061, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_60-samples_config-2_full_auto/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": 2.184868812561035, | |
| "learning_rate": 1e-05, | |
| "loss": 1.813, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 2.1768360137939453, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8081, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 1.6745238304138184, | |
| "eval_runtime": 5.1295, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 1.8703140020370483, | |
| "learning_rate": 4e-05, | |
| "loss": 1.7301, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 1.4640454053878784, | |
| "eval_runtime": 5.1274, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 1.2792514562606812, | |
| "learning_rate": 6e-05, | |
| "loss": 1.6047, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 1.0943307876586914, | |
| "learning_rate": 8e-05, | |
| "loss": 1.3805, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 1.227786898612976, | |
| "eval_runtime": 5.1515, | |
| "eval_samples_per_second": 2.329, | |
| "eval_steps_per_second": 2.329, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 1.516724705696106, | |
| "learning_rate": 0.0001, | |
| "loss": 1.2006, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 0.9718665480613708, | |
| "eval_runtime": 5.1309, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 1.1095489263534546, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.0008, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.732903242111206, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.8566, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.8841575980186462, | |
| "eval_runtime": 5.1316, | |
| "eval_samples_per_second": 2.338, | |
| "eval_steps_per_second": 2.338, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.6439677476882935, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.7977, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.8512413501739502, | |
| "eval_runtime": 5.1277, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.6010205149650574, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.723, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.5774854421615601, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.6809, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.8306775689125061, | |
| "eval_runtime": 5.1246, | |
| "eval_samples_per_second": 2.342, | |
| "eval_steps_per_second": 2.342, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.7789441347122192, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.6317, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8437394499778748, | |
| "eval_runtime": 5.1232, | |
| "eval_samples_per_second": 2.342, | |
| "eval_steps_per_second": 2.342, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.6131580471992493, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.5791, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 0.8791252970695496, | |
| "eval_runtime": 5.1303, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.7738478779792786, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.5323, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.7667720317840576, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.455, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 0.9140081405639648, | |
| "eval_runtime": 5.1303, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.718547523021698, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.404, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 0.9730315208435059, | |
| "eval_runtime": 5.1285, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.9068953990936279, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.3661, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.8681232929229736, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.291, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 1.08100426197052, | |
| "eval_runtime": 5.1302, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.7632395029067993, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.2496, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 1.2034608125686646, | |
| "eval_runtime": 5.1272, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 1.0477867126464844, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.2046, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.7814909219741821, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.1865, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 1.2849262952804565, | |
| "eval_runtime": 5.13, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "step": 40, | |
| "total_flos": 1.0447421976870912e+16, | |
| "train_loss": 0.785325912758708, | |
| "train_runtime": 783.1349, | |
| "train_samples_per_second": 2.937, | |
| "train_steps_per_second": 0.128 | |
| } | |
| ], | |
| "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": 1.0447421976870912e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |