| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.2171581769436997, | |
| "eval_steps": 200, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17873100983020554, | |
| "grad_norm": 0.5410762429237366, | |
| "learning_rate": 9.999142157322298e-05, | |
| "loss": 0.5394, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.17873100983020554, | |
| "eval_loss": 0.2514195740222931, | |
| "eval_runtime": 56.7459, | |
| "eval_samples_per_second": 3.19, | |
| "eval_steps_per_second": 1.604, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3574620196604111, | |
| "grad_norm": 0.3244825303554535, | |
| "learning_rate": 9.95497608932715e-05, | |
| "loss": 0.2114, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3574620196604111, | |
| "eval_loss": 0.19941778481006622, | |
| "eval_runtime": 56.7365, | |
| "eval_samples_per_second": 3.19, | |
| "eval_steps_per_second": 1.604, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5361930294906166, | |
| "grad_norm": 0.23322512209415436, | |
| "learning_rate": 9.844466810070319e-05, | |
| "loss": 0.184, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.5361930294906166, | |
| "eval_loss": 0.1872645616531372, | |
| "eval_runtime": 56.8389, | |
| "eval_samples_per_second": 3.184, | |
| "eval_steps_per_second": 1.601, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.7149240393208222, | |
| "grad_norm": 0.20280934870243073, | |
| "learning_rate": 9.669093951406317e-05, | |
| "loss": 0.1797, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7149240393208222, | |
| "eval_loss": 0.18097320199012756, | |
| "eval_runtime": 56.861, | |
| "eval_samples_per_second": 3.183, | |
| "eval_steps_per_second": 1.6, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8936550491510277, | |
| "grad_norm": 0.18255303800106049, | |
| "learning_rate": 9.431205617221452e-05, | |
| "loss": 0.1719, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8936550491510277, | |
| "eval_loss": 0.17648382484912872, | |
| "eval_runtime": 58.6626, | |
| "eval_samples_per_second": 3.085, | |
| "eval_steps_per_second": 1.551, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0723860589812333, | |
| "grad_norm": 0.15965856611728668, | |
| "learning_rate": 9.133986944179754e-05, | |
| "loss": 0.1647, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.0723860589812333, | |
| "eval_loss": 0.17539893090724945, | |
| "eval_runtime": 58.6505, | |
| "eval_samples_per_second": 3.086, | |
| "eval_steps_per_second": 1.552, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.2511170688114388, | |
| "grad_norm": 0.187672421336174, | |
| "learning_rate": 8.781417455261971e-05, | |
| "loss": 0.1538, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.2511170688114388, | |
| "eval_loss": 0.1730012148618698, | |
| "eval_runtime": 58.3686, | |
| "eval_samples_per_second": 3.101, | |
| "eval_steps_per_second": 1.559, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.4298480786416443, | |
| "grad_norm": 0.21108384430408478, | |
| "learning_rate": 8.378217777099999e-05, | |
| "loss": 0.1523, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.4298480786416443, | |
| "eval_loss": 0.17163121700286865, | |
| "eval_runtime": 58.6595, | |
| "eval_samples_per_second": 3.086, | |
| "eval_steps_per_second": 1.551, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.6085790884718498, | |
| "grad_norm": 0.1646961271762848, | |
| "learning_rate": 7.929786434519532e-05, | |
| "loss": 0.1524, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.6085790884718498, | |
| "eval_loss": 0.1694817841053009, | |
| "eval_runtime": 56.1831, | |
| "eval_samples_per_second": 3.222, | |
| "eval_steps_per_second": 1.62, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.7873100983020556, | |
| "grad_norm": 0.19810634851455688, | |
| "learning_rate": 7.442127568561985e-05, | |
| "loss": 0.1487, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.7873100983020556, | |
| "eval_loss": 0.1678919941186905, | |
| "eval_runtime": 56.3246, | |
| "eval_samples_per_second": 3.214, | |
| "eval_steps_per_second": 1.616, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.966041108132261, | |
| "grad_norm": 0.2072182446718216, | |
| "learning_rate": 6.921770545784254e-05, | |
| "loss": 0.1513, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.966041108132261, | |
| "eval_loss": 0.16549332439899445, | |
| "eval_runtime": 56.2064, | |
| "eval_samples_per_second": 3.22, | |
| "eval_steps_per_second": 1.619, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.1447721179624666, | |
| "grad_norm": 0.1624869555234909, | |
| "learning_rate": 6.375682535204167e-05, | |
| "loss": 0.1338, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.1447721179624666, | |
| "eval_loss": 0.1687156856060028, | |
| "eval_runtime": 56.2284, | |
| "eval_samples_per_second": 3.219, | |
| "eval_steps_per_second": 1.618, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.323503127792672, | |
| "grad_norm": 0.22109678387641907, | |
| "learning_rate": 5.81117522341723e-05, | |
| "loss": 0.1295, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.323503127792672, | |
| "eval_loss": 0.167947456240654, | |
| "eval_runtime": 56.2002, | |
| "eval_samples_per_second": 3.221, | |
| "eval_steps_per_second": 1.619, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.5022341376228776, | |
| "grad_norm": 0.18057583272457123, | |
| "learning_rate": 5.2358069168954583e-05, | |
| "loss": 0.1295, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.5022341376228776, | |
| "eval_loss": 0.16731825470924377, | |
| "eval_runtime": 56.1976, | |
| "eval_samples_per_second": 3.221, | |
| "eval_steps_per_second": 1.619, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.680965147453083, | |
| "grad_norm": 0.18708354234695435, | |
| "learning_rate": 4.657281342241101e-05, | |
| "loss": 0.1295, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.680965147453083, | |
| "eval_loss": 0.16666428744792938, | |
| "eval_runtime": 56.1642, | |
| "eval_samples_per_second": 3.223, | |
| "eval_steps_per_second": 1.62, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.8596961572832886, | |
| "grad_norm": 0.23575575649738312, | |
| "learning_rate": 4.083344499379815e-05, | |
| "loss": 0.1287, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.8596961572832886, | |
| "eval_loss": 0.16479900479316711, | |
| "eval_runtime": 56.3182, | |
| "eval_samples_per_second": 3.214, | |
| "eval_steps_per_second": 1.616, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.038427167113494, | |
| "grad_norm": 0.20305226743221283, | |
| "learning_rate": 3.521680948747494e-05, | |
| "loss": 0.1243, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.038427167113494, | |
| "eval_loss": 0.17072516679763794, | |
| "eval_runtime": 56.3856, | |
| "eval_samples_per_second": 3.21, | |
| "eval_steps_per_second": 1.614, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 3.2171581769436997, | |
| "grad_norm": 0.2290426343679428, | |
| "learning_rate": 2.979810921103237e-05, | |
| "loss": 0.1084, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.2171581769436997, | |
| "eval_loss": 0.1712944209575653, | |
| "eval_runtime": 56.4248, | |
| "eval_samples_per_second": 3.208, | |
| "eval_steps_per_second": 1.613, | |
| "step": 3600 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 5595, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.993935653166285e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |