| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 27.11864406779661, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 1.9129917621612549, | |
| "eval_runtime": 85.8103, | |
| "eval_samples_per_second": 24.414, | |
| "eval_steps_per_second": 3.053, | |
| "eval_wer": 0.9244121629939678, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0002982, | |
| "loss": 5.0013, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "eval_loss": 0.7788760662078857, | |
| "eval_runtime": 85.5378, | |
| "eval_samples_per_second": 24.492, | |
| "eval_steps_per_second": 3.063, | |
| "eval_wer": 0.5943616890311462, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00028011999999999997, | |
| "loss": 0.6544, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "eval_loss": 0.7297950983047485, | |
| "eval_runtime": 84.8704, | |
| "eval_samples_per_second": 24.685, | |
| "eval_steps_per_second": 3.087, | |
| "eval_wer": 0.585190200664779, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.00026011999999999997, | |
| "loss": 0.4021, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "eval_loss": 0.697790265083313, | |
| "eval_runtime": 84.1609, | |
| "eval_samples_per_second": 24.893, | |
| "eval_steps_per_second": 3.113, | |
| "eval_wer": 0.5667241167056506, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.00024011999999999997, | |
| "loss": 0.3003, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "eval_loss": 0.6764048337936401, | |
| "eval_runtime": 85.0271, | |
| "eval_samples_per_second": 24.639, | |
| "eval_steps_per_second": 3.081, | |
| "eval_wer": 0.5382247937953958, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "eval_loss": 0.72489994764328, | |
| "eval_runtime": 83.797, | |
| "eval_samples_per_second": 25.001, | |
| "eval_steps_per_second": 3.127, | |
| "eval_wer": 0.5462883171242152, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 0.00022011999999999997, | |
| "loss": 0.2345, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "eval_loss": 0.7279900908470154, | |
| "eval_runtime": 81.6862, | |
| "eval_samples_per_second": 25.647, | |
| "eval_steps_per_second": 3.207, | |
| "eval_wer": 0.5124338298658131, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.00020012, | |
| "loss": 0.1993, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.85, | |
| "eval_loss": 0.7288674712181091, | |
| "eval_runtime": 81.4638, | |
| "eval_samples_per_second": 25.717, | |
| "eval_steps_per_second": 3.216, | |
| "eval_wer": 0.4690385325618614, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.00018012, | |
| "loss": 0.1617, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.2, | |
| "eval_loss": 0.7430842518806458, | |
| "eval_runtime": 81.3254, | |
| "eval_samples_per_second": 25.761, | |
| "eval_steps_per_second": 3.222, | |
| "eval_wer": 0.4732857318724609, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 0.00016011999999999998, | |
| "loss": 0.1432, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "eval_loss": 0.7448425889015198, | |
| "eval_runtime": 83.7392, | |
| "eval_samples_per_second": 25.018, | |
| "eval_steps_per_second": 3.129, | |
| "eval_wer": 0.4732857318724609, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.92, | |
| "eval_loss": 0.7745729088783264, | |
| "eval_runtime": 83.3791, | |
| "eval_samples_per_second": 25.126, | |
| "eval_steps_per_second": 3.142, | |
| "eval_wer": 0.4484796257540318, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 0.00014012, | |
| "loss": 0.1172, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 16.27, | |
| "eval_loss": 0.7588675022125244, | |
| "eval_runtime": 83.1746, | |
| "eval_samples_per_second": 25.188, | |
| "eval_steps_per_second": 3.15, | |
| "eval_wer": 0.4742090360704173, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 0.00012011999999999998, | |
| "loss": 0.1035, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.63, | |
| "eval_loss": 0.7539412975311279, | |
| "eval_runtime": 83.8808, | |
| "eval_samples_per_second": 24.976, | |
| "eval_steps_per_second": 3.123, | |
| "eval_wer": 0.4353071525298535, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 0.00010011999999999998, | |
| "loss": 0.0956, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_loss": 0.7648215293884277, | |
| "eval_runtime": 83.2596, | |
| "eval_samples_per_second": 25.162, | |
| "eval_steps_per_second": 3.147, | |
| "eval_wer": 0.44946448356518526, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 20.34, | |
| "learning_rate": 8.012e-05, | |
| "loss": 0.0845, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 20.34, | |
| "eval_loss": 0.7876783013343811, | |
| "eval_runtime": 84.6623, | |
| "eval_samples_per_second": 24.745, | |
| "eval_steps_per_second": 3.095, | |
| "eval_wer": 0.4718699987689277, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 21.69, | |
| "eval_loss": 0.7883597016334534, | |
| "eval_runtime": 84.2577, | |
| "eval_samples_per_second": 24.864, | |
| "eval_steps_per_second": 3.11, | |
| "eval_wer": 0.44343222947187, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 22.03, | |
| "learning_rate": 6.0119999999999994e-05, | |
| "loss": 0.0761, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 23.05, | |
| "eval_loss": 0.7796189188957214, | |
| "eval_runtime": 81.0587, | |
| "eval_samples_per_second": 25.845, | |
| "eval_steps_per_second": 3.232, | |
| "eval_wer": 0.4385694940292995, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 23.73, | |
| "learning_rate": 4.012e-05, | |
| "loss": 0.0634, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 24.41, | |
| "eval_loss": 0.7729123830795288, | |
| "eval_runtime": 81.4812, | |
| "eval_samples_per_second": 25.711, | |
| "eval_steps_per_second": 3.215, | |
| "eval_wer": 0.4306290779268743, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 25.42, | |
| "learning_rate": 2.0159999999999997e-05, | |
| "loss": 0.0571, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 25.76, | |
| "eval_loss": 0.7826283574104309, | |
| "eval_runtime": 81.9341, | |
| "eval_samples_per_second": 25.569, | |
| "eval_steps_per_second": 3.198, | |
| "eval_wer": 0.4298288809553121, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "learning_rate": 1.6e-07, | |
| "loss": 0.0508, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "eval_loss": 0.7805310487747192, | |
| "eval_runtime": 81.3643, | |
| "eval_samples_per_second": 25.748, | |
| "eval_steps_per_second": 3.22, | |
| "eval_wer": 0.4340145266527145, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "step": 8000, | |
| "total_flos": 1.5928886672052732e+19, | |
| "train_loss": 0.48406150817871096, | |
| "train_runtime": 11109.9321, | |
| "train_samples_per_second": 11.521, | |
| "train_steps_per_second": 0.72 | |
| } | |
| ], | |
| "max_steps": 8000, | |
| "num_train_epochs": 28, | |
| "total_flos": 1.5928886672052732e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |