| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 300, | |
| "global_step": 7070, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4243281471004243, | |
| "eval_loss": 5.990268707275391, | |
| "eval_runtime": 138.1214, | |
| "eval_samples_per_second": 40.949, | |
| "eval_steps_per_second": 5.119, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7072135785007072, | |
| "grad_norm": 2.022184371948242, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 7.061, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8486562942008486, | |
| "eval_loss": 3.0451483726501465, | |
| "eval_runtime": 135.2034, | |
| "eval_samples_per_second": 41.833, | |
| "eval_steps_per_second": 5.229, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.272984441301273, | |
| "eval_loss": 2.9642043113708496, | |
| "eval_runtime": 136.8192, | |
| "eval_samples_per_second": 41.339, | |
| "eval_steps_per_second": 5.167, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4144271570014144, | |
| "grad_norm": 0.7561541795730591, | |
| "learning_rate": 4.624048706240488e-05, | |
| "loss": 3.0081, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6973125884016973, | |
| "eval_loss": 2.956415891647339, | |
| "eval_runtime": 136.6446, | |
| "eval_samples_per_second": 41.392, | |
| "eval_steps_per_second": 5.174, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.1216407355021216, | |
| "grad_norm": 0.3668934106826782, | |
| "learning_rate": 4.245053272450533e-05, | |
| "loss": 2.9733, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.1216407355021216, | |
| "eval_loss": 2.947998285293579, | |
| "eval_runtime": 136.4831, | |
| "eval_samples_per_second": 41.441, | |
| "eval_steps_per_second": 5.18, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.545968882602546, | |
| "eval_loss": 2.945077657699585, | |
| "eval_runtime": 136.3573, | |
| "eval_samples_per_second": 41.479, | |
| "eval_steps_per_second": 5.185, | |
| "eval_wer": 1.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.828854314002829, | |
| "grad_norm": 1.441468596458435, | |
| "learning_rate": 3.866057838660579e-05, | |
| "loss": 2.9454, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.9702970297029703, | |
| "eval_loss": 2.9147346019744873, | |
| "eval_runtime": 135.861, | |
| "eval_samples_per_second": 41.631, | |
| "eval_steps_per_second": 5.204, | |
| "eval_wer": 1.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.3946251768033946, | |
| "eval_loss": 2.9019417762756348, | |
| "eval_runtime": 136.0133, | |
| "eval_samples_per_second": 41.584, | |
| "eval_steps_per_second": 5.198, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "grad_norm": 0.46694883704185486, | |
| "learning_rate": 3.487062404870624e-05, | |
| "loss": 2.9064, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.818953323903819, | |
| "eval_loss": 2.884958505630493, | |
| "eval_runtime": 136.0891, | |
| "eval_samples_per_second": 41.561, | |
| "eval_steps_per_second": 5.195, | |
| "eval_wer": 1.0, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 4.243281471004243, | |
| "grad_norm": 0.4952280819416046, | |
| "learning_rate": 3.10882800608828e-05, | |
| "loss": 2.9048, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.243281471004243, | |
| "eval_loss": 2.8812334537506104, | |
| "eval_runtime": 136.4568, | |
| "eval_samples_per_second": 41.449, | |
| "eval_steps_per_second": 5.181, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.667609618104668, | |
| "eval_loss": 2.884371042251587, | |
| "eval_runtime": 136.787, | |
| "eval_samples_per_second": 41.349, | |
| "eval_steps_per_second": 5.169, | |
| "eval_wer": 1.0, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.9504950495049505, | |
| "grad_norm": 0.8865047097206116, | |
| "learning_rate": 2.7290715372907157e-05, | |
| "loss": 2.8965, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.091937765205092, | |
| "eval_loss": 2.9125277996063232, | |
| "eval_runtime": 136.3564, | |
| "eval_samples_per_second": 41.48, | |
| "eval_steps_per_second": 5.185, | |
| "eval_wer": 1.0, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 5.516265912305516, | |
| "eval_loss": 2.898144006729126, | |
| "eval_runtime": 136.0768, | |
| "eval_samples_per_second": 41.565, | |
| "eval_steps_per_second": 5.196, | |
| "eval_wer": 1.0, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 5.657708628005658, | |
| "grad_norm": 0.3529145121574402, | |
| "learning_rate": 2.3508371385083716e-05, | |
| "loss": 2.9261, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.9405940594059405, | |
| "eval_loss": 2.905318260192871, | |
| "eval_runtime": 136.6781, | |
| "eval_samples_per_second": 41.382, | |
| "eval_steps_per_second": 5.173, | |
| "eval_wer": 1.0, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 6.364922206506365, | |
| "grad_norm": 0.22229251265525818, | |
| "learning_rate": 1.971841704718417e-05, | |
| "loss": 2.9273, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.364922206506365, | |
| "eval_loss": 2.916677951812744, | |
| "eval_runtime": 136.7502, | |
| "eval_samples_per_second": 41.36, | |
| "eval_steps_per_second": 5.17, | |
| "eval_wer": 1.0, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 6.789250353606789, | |
| "eval_loss": 2.911259651184082, | |
| "eval_runtime": 136.484, | |
| "eval_samples_per_second": 41.441, | |
| "eval_steps_per_second": 5.18, | |
| "eval_wer": 1.0, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 7.072135785007072, | |
| "grad_norm": 1.7586228847503662, | |
| "learning_rate": 1.592846270928463e-05, | |
| "loss": 2.9302, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 7.2135785007072135, | |
| "eval_loss": 2.9133317470550537, | |
| "eval_runtime": 135.9523, | |
| "eval_samples_per_second": 41.603, | |
| "eval_steps_per_second": 5.2, | |
| "eval_wer": 1.0, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 7.637906647807638, | |
| "eval_loss": 2.921302080154419, | |
| "eval_runtime": 136.5228, | |
| "eval_samples_per_second": 41.429, | |
| "eval_steps_per_second": 5.179, | |
| "eval_wer": 1.0, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 7.779349363507779, | |
| "grad_norm": 0.6302638649940491, | |
| "learning_rate": 1.2146118721461187e-05, | |
| "loss": 2.9397, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 8.062234794908063, | |
| "eval_loss": 2.9251174926757812, | |
| "eval_runtime": 136.4335, | |
| "eval_samples_per_second": 41.456, | |
| "eval_steps_per_second": 5.182, | |
| "eval_wer": 1.0, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 8.486562942008486, | |
| "grad_norm": 0.5835816860198975, | |
| "learning_rate": 8.340943683409437e-06, | |
| "loss": 2.937, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.486562942008486, | |
| "eval_loss": 2.921030282974243, | |
| "eval_runtime": 136.1229, | |
| "eval_samples_per_second": 41.551, | |
| "eval_steps_per_second": 5.194, | |
| "eval_wer": 1.0, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 8.910891089108912, | |
| "eval_loss": 2.92145037651062, | |
| "eval_runtime": 137.1094, | |
| "eval_samples_per_second": 41.252, | |
| "eval_steps_per_second": 5.156, | |
| "eval_wer": 1.0, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 9.193776520509195, | |
| "grad_norm": 0.7211419939994812, | |
| "learning_rate": 4.558599695585997e-06, | |
| "loss": 2.9406, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 9.335219236209335, | |
| "eval_loss": 2.917142629623413, | |
| "eval_runtime": 136.4229, | |
| "eval_samples_per_second": 41.459, | |
| "eval_steps_per_second": 5.182, | |
| "eval_wer": 1.0, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 9.75954738330976, | |
| "eval_loss": 2.9176828861236572, | |
| "eval_runtime": 137.2177, | |
| "eval_samples_per_second": 41.219, | |
| "eval_steps_per_second": 5.152, | |
| "eval_wer": 1.0, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 9.900990099009901, | |
| "grad_norm": 0.0, | |
| "learning_rate": 7.762557077625571e-07, | |
| "loss": 2.9378, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 7070, | |
| "total_flos": 2.4662883830172946e+19, | |
| "train_loss": 3.228043903960534, | |
| "train_runtime": 12366.081, | |
| "train_samples_per_second": 18.293, | |
| "train_steps_per_second": 0.572 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7070, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 600, | |
| "total_flos": 2.4662883830172946e+19, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |