| { |
| "best_global_step": 2000, |
| "best_metric": 0.47075214982032776, |
| "best_model_checkpoint": "Assignment4_Distilled_ModernBERT/run-1/checkpoint-2000", |
| "epoch": 4.1928721174004195, |
| "eval_steps": 100, |
| "global_step": 2000, |
| "is_hyper_param_search": true, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.20964360587002095, |
| "grad_norm": 24.706209182739258, |
| "learning_rate": 5.990041263622974e-05, |
| "loss": 12.0071, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.20964360587002095, |
| "eval_accuracy": 0.7661290322580645, |
| "eval_loss": 5.098243236541748, |
| "eval_runtime": 24.2956, |
| "eval_samples_per_second": 127.595, |
| "eval_steps_per_second": 15.97, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.4192872117400419, |
| "grad_norm": 13.007184982299805, |
| "learning_rate": 5.959829353527136e-05, |
| "loss": 3.4751, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.4192872117400419, |
| "eval_accuracy": 0.9196774193548387, |
| "eval_loss": 2.054624557495117, |
| "eval_runtime": 24.5401, |
| "eval_samples_per_second": 126.324, |
| "eval_steps_per_second": 15.811, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.6289308176100629, |
| "grad_norm": 6.618763446807861, |
| "learning_rate": 5.9095679050071424e-05, |
| "loss": 1.8181, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.6289308176100629, |
| "eval_accuracy": 0.9454838709677419, |
| "eval_loss": 1.4840706586837769, |
| "eval_runtime": 24.5263, |
| "eval_samples_per_second": 126.395, |
| "eval_steps_per_second": 15.82, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.8385744234800838, |
| "grad_norm": 17.116268157958984, |
| "learning_rate": 5.839597383246284e-05, |
| "loss": 1.3857, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.8385744234800838, |
| "eval_accuracy": 0.9558064516129032, |
| "eval_loss": 1.2064718008041382, |
| "eval_runtime": 24.6049, |
| "eval_samples_per_second": 125.991, |
| "eval_steps_per_second": 15.769, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.0482180293501049, |
| "grad_norm": 10.49691104888916, |
| "learning_rate": 5.750391760388572e-05, |
| "loss": 1.0929, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.0482180293501049, |
| "eval_accuracy": 0.9603225806451613, |
| "eval_loss": 1.0366302728652954, |
| "eval_runtime": 24.6779, |
| "eval_samples_per_second": 125.619, |
| "eval_steps_per_second": 15.723, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.2578616352201257, |
| "grad_norm": 4.32417106628418, |
| "learning_rate": 5.642555304906785e-05, |
| "loss": 0.8031, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.2578616352201257, |
| "eval_accuracy": 0.9651612903225807, |
| "eval_loss": 0.8905357718467712, |
| "eval_runtime": 25.2308, |
| "eval_samples_per_second": 122.866, |
| "eval_steps_per_second": 15.378, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.4675052410901468, |
| "grad_norm": 5.406693935394287, |
| "learning_rate": 5.51681848835839e-05, |
| "loss": 0.7532, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.4675052410901468, |
| "eval_accuracy": 0.9609677419354838, |
| "eval_loss": 0.8550653457641602, |
| "eval_runtime": 26.4505, |
| "eval_samples_per_second": 117.2, |
| "eval_steps_per_second": 14.669, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.6771488469601676, |
| "grad_norm": 3.1182005405426025, |
| "learning_rate": 5.374033037256506e-05, |
| "loss": 0.6439, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.6771488469601676, |
| "eval_accuracy": 0.9654838709677419, |
| "eval_loss": 0.7534056901931763, |
| "eval_runtime": 24.5816, |
| "eval_samples_per_second": 126.111, |
| "eval_steps_per_second": 15.784, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.8867924528301887, |
| "grad_norm": 8.852187156677246, |
| "learning_rate": 5.215166163573972e-05, |
| "loss": 0.6199, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.8867924528301887, |
| "eval_accuracy": 0.9690322580645161, |
| "eval_loss": 0.7397432327270508, |
| "eval_runtime": 24.6609, |
| "eval_samples_per_second": 125.705, |
| "eval_steps_per_second": 15.733, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.0964360587002098, |
| "grad_norm": 2.526475429534912, |
| "learning_rate": 5.041294012962368e-05, |
| "loss": 0.5397, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.0964360587002098, |
| "eval_accuracy": 0.9683870967741935, |
| "eval_loss": 0.6956890225410461, |
| "eval_runtime": 24.6718, |
| "eval_samples_per_second": 125.649, |
| "eval_steps_per_second": 15.726, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.3060796645702304, |
| "grad_norm": 2.916027307510376, |
| "learning_rate": 4.85359437506704e-05, |
| "loss": 0.4462, |
| "step": 1100 |
| }, |
| { |
| "epoch": 2.3060796645702304, |
| "eval_accuracy": 0.9648387096774194, |
| "eval_loss": 0.6581540107727051, |
| "eval_runtime": 24.7875, |
| "eval_samples_per_second": 125.063, |
| "eval_steps_per_second": 15.653, |
| "step": 1100 |
| }, |
| { |
| "epoch": 2.5157232704402515, |
| "grad_norm": 3.283611297607422, |
| "learning_rate": 4.6533387053175734e-05, |
| "loss": 0.4288, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.5157232704402515, |
| "eval_accuracy": 0.9719354838709677, |
| "eval_loss": 0.6225466132164001, |
| "eval_runtime": 28.7571, |
| "eval_samples_per_second": 107.799, |
| "eval_steps_per_second": 13.492, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.7253668763102725, |
| "grad_norm": 2.559558868408203, |
| "learning_rate": 4.441883512237173e-05, |
| "loss": 0.4, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.7253668763102725, |
| "eval_accuracy": 0.9687096774193549, |
| "eval_loss": 0.5823577642440796, |
| "eval_runtime": 24.4941, |
| "eval_samples_per_second": 126.561, |
| "eval_steps_per_second": 15.841, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.9350104821802936, |
| "grad_norm": 2.1637282371520996, |
| "learning_rate": 4.2206611686123325e-05, |
| "loss": 0.3764, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.9350104821802936, |
| "eval_accuracy": 0.9703225806451613, |
| "eval_loss": 0.570899486541748, |
| "eval_runtime": 24.7744, |
| "eval_samples_per_second": 125.129, |
| "eval_steps_per_second": 15.661, |
| "step": 1400 |
| }, |
| { |
| "epoch": 3.1446540880503147, |
| "grad_norm": 3.660550594329834, |
| "learning_rate": 3.991170208766829e-05, |
| "loss": 0.3398, |
| "step": 1500 |
| }, |
| { |
| "epoch": 3.1446540880503147, |
| "eval_accuracy": 0.9693548387096774, |
| "eval_loss": 0.5414220690727234, |
| "eval_runtime": 24.8467, |
| "eval_samples_per_second": 124.765, |
| "eval_steps_per_second": 15.616, |
| "step": 1500 |
| }, |
| { |
| "epoch": 3.3542976939203353, |
| "grad_norm": 2.297179698944092, |
| "learning_rate": 3.7549651776652014e-05, |
| "loss": 0.3155, |
| "step": 1600 |
| }, |
| { |
| "epoch": 3.3542976939203353, |
| "eval_accuracy": 0.9732258064516129, |
| "eval_loss": 0.5297242403030396, |
| "eval_runtime": 24.8119, |
| "eval_samples_per_second": 124.94, |
| "eval_steps_per_second": 15.638, |
| "step": 1600 |
| }, |
| { |
| "epoch": 3.5639412997903563, |
| "grad_norm": 1.6558480262756348, |
| "learning_rate": 3.513646100606681e-05, |
| "loss": 0.301, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.5639412997903563, |
| "eval_accuracy": 0.9703225806451613, |
| "eval_loss": 0.5193695425987244, |
| "eval_runtime": 29.3532, |
| "eval_samples_per_second": 105.61, |
| "eval_steps_per_second": 13.218, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.7735849056603774, |
| "grad_norm": 2.271939277648926, |
| "learning_rate": 3.268847644840649e-05, |
| "loss": 0.2918, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.7735849056603774, |
| "eval_accuracy": 0.9719354838709677, |
| "eval_loss": 0.49459728598594666, |
| "eval_runtime": 24.5731, |
| "eval_samples_per_second": 126.154, |
| "eval_steps_per_second": 15.79, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.9832285115303985, |
| "grad_norm": 2.7767817974090576, |
| "learning_rate": 3.0222280465215667e-05, |
| "loss": 0.286, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.9832285115303985, |
| "eval_accuracy": 0.9712903225806452, |
| "eval_loss": 0.4878614842891693, |
| "eval_runtime": 24.7879, |
| "eval_samples_per_second": 125.061, |
| "eval_steps_per_second": 15.653, |
| "step": 1900 |
| }, |
| { |
| "epoch": 4.1928721174004195, |
| "grad_norm": 1.6416629552841187, |
| "learning_rate": 2.7754578780108955e-05, |
| "loss": 0.2506, |
| "step": 2000 |
| }, |
| { |
| "epoch": 4.1928721174004195, |
| "eval_accuracy": 0.9722580645161291, |
| "eval_loss": 0.47075214982032776, |
| "eval_runtime": 24.6817, |
| "eval_samples_per_second": 125.599, |
| "eval_steps_per_second": 15.72, |
| "step": 2000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 3816, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 8, |
| "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": 896601881570736.0, |
| "train_batch_size": 32, |
| "trial_name": null, |
| "trial_params": { |
| "alpha": 0.29055745323517945, |
| "num_train_epochs": 8, |
| "temperature": 5 |
| } |
| } |
|
|
