| { |
| "best_global_step": 1000, |
| "best_metric": 0.4247872829437256, |
| "best_model_checkpoint": "Assignment4_Distilled_ModernBERT/run-4/checkpoint-1000", |
| "epoch": 2.0964360587002098, |
| "eval_steps": 100, |
| "global_step": 1000, |
| "is_hyper_param_search": true, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.20964360587002095, |
| "grad_norm": 11.635396957397461, |
| "learning_rate": 5.974527645836366e-05, |
| "loss": 6.4293, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.20964360587002095, |
| "eval_accuracy": 0.7254838709677419, |
| "eval_loss": 3.423161268234253, |
| "eval_runtime": 24.7311, |
| "eval_samples_per_second": 125.348, |
| "eval_steps_per_second": 15.689, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.4192872117400419, |
| "grad_norm": 7.08547830581665, |
| "learning_rate": 5.897521628664326e-05, |
| "loss": 2.4649, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.4192872117400419, |
| "eval_accuracy": 0.9012903225806451, |
| "eval_loss": 1.6392749547958374, |
| "eval_runtime": 24.334, |
| "eval_samples_per_second": 127.394, |
| "eval_steps_per_second": 15.945, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.6289308176100629, |
| "grad_norm": 2.8314898014068604, |
| "learning_rate": 5.7703135417326854e-05, |
| "loss": 1.364, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.6289308176100629, |
| "eval_accuracy": 0.9361290322580645, |
| "eval_loss": 1.0801855325698853, |
| "eval_runtime": 24.4718, |
| "eval_samples_per_second": 126.676, |
| "eval_steps_per_second": 15.855, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.8385744234800838, |
| "grad_norm": 11.189908981323242, |
| "learning_rate": 5.595107375234632e-05, |
| "loss": 0.9328, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.8385744234800838, |
| "eval_accuracy": 0.9506451612903226, |
| "eval_loss": 0.8573997616767883, |
| "eval_runtime": 24.7134, |
| "eval_samples_per_second": 125.438, |
| "eval_steps_per_second": 15.7, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.0482180293501049, |
| "grad_norm": 9.433755874633789, |
| "learning_rate": 5.3749387276298395e-05, |
| "loss": 0.708, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0482180293501049, |
| "eval_accuracy": 0.9564516129032258, |
| "eval_loss": 0.6847298741340637, |
| "eval_runtime": 24.9978, |
| "eval_samples_per_second": 124.011, |
| "eval_steps_per_second": 15.521, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.2578616352201257, |
| "grad_norm": 1.7648718357086182, |
| "learning_rate": 5.113622211274119e-05, |
| "loss": 0.4702, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.2578616352201257, |
| "eval_accuracy": 0.9616129032258065, |
| "eval_loss": 0.5767695307731628, |
| "eval_runtime": 25.0749, |
| "eval_samples_per_second": 123.63, |
| "eval_steps_per_second": 15.474, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.4675052410901468, |
| "grad_norm": 2.782994031906128, |
| "learning_rate": 4.815685360959153e-05, |
| "loss": 0.4294, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.4675052410901468, |
| "eval_accuracy": 0.9583870967741935, |
| "eval_loss": 0.5501284599304199, |
| "eval_runtime": 24.8053, |
| "eval_samples_per_second": 124.973, |
| "eval_steps_per_second": 15.642, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.6771488469601676, |
| "grad_norm": 1.1726200580596924, |
| "learning_rate": 4.486290190454053e-05, |
| "loss": 0.3644, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.6771488469601676, |
| "eval_accuracy": 0.964516129032258, |
| "eval_loss": 0.48550063371658325, |
| "eval_runtime": 25.3011, |
| "eval_samples_per_second": 122.524, |
| "eval_steps_per_second": 15.335, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.8867924528301887, |
| "grad_norm": 11.229223251342773, |
| "learning_rate": 4.13114375614956e-05, |
| "loss": 0.3547, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.8867924528301887, |
| "eval_accuracy": 0.9664516129032258, |
| "eval_loss": 0.46239548921585083, |
| "eval_runtime": 25.1571, |
| "eval_samples_per_second": 123.226, |
| "eval_steps_per_second": 15.423, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.0964360587002098, |
| "grad_norm": 1.082562804222107, |
| "learning_rate": 3.756399277367126e-05, |
| "loss": 0.2973, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.0964360587002098, |
| "eval_accuracy": 0.967741935483871, |
| "eval_loss": 0.4247872829437256, |
| "eval_runtime": 31.4174, |
| "eval_samples_per_second": 98.671, |
| "eval_steps_per_second": 12.35, |
| "step": 1000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 2385, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 448725995623620.0, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": { |
| "alpha": 0.5611126477660928, |
| "num_train_epochs": 5, |
| "temperature": 7 |
| } |
| } |
|
|
