| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 28.8135593220339, | |
| "global_step": 8500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 4.94e-05, | |
| "loss": 6.7713, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "eval_loss": 2.9682114124298096, | |
| "eval_runtime": 24.5806, | |
| "eval_samples_per_second": 15.663, | |
| "eval_steps_per_second": 1.993, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 9.94e-05, | |
| "loss": 1.7717, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "eval_loss": 1.053725004196167, | |
| "eval_runtime": 25.0326, | |
| "eval_samples_per_second": 15.38, | |
| "eval_steps_per_second": 1.957, | |
| "eval_wer": 0.7996899224806202, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 9.370700636942676e-05, | |
| "loss": 0.8188, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "eval_loss": 0.723295271396637, | |
| "eval_runtime": 24.8888, | |
| "eval_samples_per_second": 15.469, | |
| "eval_steps_per_second": 1.969, | |
| "eval_wer": 0.6505426356589147, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 8.73375796178344e-05, | |
| "loss": 0.5633, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "eval_loss": 0.6577075719833374, | |
| "eval_runtime": 24.4626, | |
| "eval_samples_per_second": 15.738, | |
| "eval_steps_per_second": 2.003, | |
| "eval_wer": 0.6083720930232558, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 8.096815286624205e-05, | |
| "loss": 0.4201, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "eval_loss": 0.6801705360412598, | |
| "eval_runtime": 23.9758, | |
| "eval_samples_per_second": 16.058, | |
| "eval_steps_per_second": 2.044, | |
| "eval_wer": 0.5937984496124031, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 7.459872611464968e-05, | |
| "loss": 0.3465, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "eval_loss": 0.734240710735321, | |
| "eval_runtime": 24.3607, | |
| "eval_samples_per_second": 15.804, | |
| "eval_steps_per_second": 2.011, | |
| "eval_wer": 0.5792248062015504, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 6.822929936305733e-05, | |
| "loss": 0.2812, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "eval_loss": 0.654410183429718, | |
| "eval_runtime": 25.1886, | |
| "eval_samples_per_second": 15.285, | |
| "eval_steps_per_second": 1.945, | |
| "eval_wer": 0.56, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 6.185987261146497e-05, | |
| "loss": 0.2362, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "eval_loss": 0.6740626692771912, | |
| "eval_runtime": 24.8977, | |
| "eval_samples_per_second": 15.463, | |
| "eval_steps_per_second": 1.968, | |
| "eval_wer": 0.5311627906976745, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 5.549044585987262e-05, | |
| "loss": 0.2042, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "eval_loss": 0.7329800724983215, | |
| "eval_runtime": 24.5766, | |
| "eval_samples_per_second": 15.665, | |
| "eval_steps_per_second": 1.994, | |
| "eval_wer": 0.5221705426356589, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 4.912101910828026e-05, | |
| "loss": 0.1881, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "eval_loss": 0.7085126042366028, | |
| "eval_runtime": 25.0265, | |
| "eval_samples_per_second": 15.384, | |
| "eval_steps_per_second": 1.958, | |
| "eval_wer": 0.5184496124031007, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 4.27515923566879e-05, | |
| "loss": 0.1632, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "eval_loss": 0.6873669624328613, | |
| "eval_runtime": 24.3859, | |
| "eval_samples_per_second": 15.788, | |
| "eval_steps_per_second": 2.009, | |
| "eval_wer": 0.5255813953488372, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 20.34, | |
| "learning_rate": 3.638216560509554e-05, | |
| "loss": 0.1502, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 20.34, | |
| "eval_loss": 0.7731661200523376, | |
| "eval_runtime": 25.086, | |
| "eval_samples_per_second": 15.347, | |
| "eval_steps_per_second": 1.953, | |
| "eval_wer": 0.5193798449612403, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 22.03, | |
| "learning_rate": 3.0012738853503187e-05, | |
| "loss": 0.1338, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 22.03, | |
| "eval_loss": 0.7070124745368958, | |
| "eval_runtime": 24.9391, | |
| "eval_samples_per_second": 15.438, | |
| "eval_steps_per_second": 1.965, | |
| "eval_wer": 0.5041860465116279, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 23.73, | |
| "learning_rate": 2.3656050955414013e-05, | |
| "loss": 0.1295, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 23.73, | |
| "eval_loss": 0.7655993103981018, | |
| "eval_runtime": 24.8534, | |
| "eval_samples_per_second": 15.491, | |
| "eval_steps_per_second": 1.972, | |
| "eval_wer": 0.4951937984496124, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 25.42, | |
| "learning_rate": 1.7286624203821657e-05, | |
| "loss": 0.1143, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 25.42, | |
| "eval_loss": 0.7407109141349792, | |
| "eval_runtime": 24.926, | |
| "eval_samples_per_second": 15.446, | |
| "eval_steps_per_second": 1.966, | |
| "eval_wer": 0.4951937984496124, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "learning_rate": 1.09171974522293e-05, | |
| "loss": 0.104, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "eval_loss": 0.7474448680877686, | |
| "eval_runtime": 25.1716, | |
| "eval_samples_per_second": 15.295, | |
| "eval_steps_per_second": 1.947, | |
| "eval_wer": 0.4846511627906977, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 28.81, | |
| "learning_rate": 4.5477707006369424e-06, | |
| "loss": 0.1077, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 28.81, | |
| "eval_loss": 0.7391781210899353, | |
| "eval_runtime": 24.7928, | |
| "eval_samples_per_second": 15.529, | |
| "eval_steps_per_second": 1.976, | |
| "eval_wer": 0.47968992248062015, | |
| "step": 8500 | |
| } | |
| ], | |
| "max_steps": 8850, | |
| "num_train_epochs": 30, | |
| "total_flos": 2.7043654893523635e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |