| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.5444015444015444, | |
| "eval_steps": 200, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07722007722007722, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 178.8025, | |
| "eval_samples_per_second": 39.272, | |
| "eval_steps_per_second": 4.91, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15444015444015444, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 176.9522, | |
| "eval_samples_per_second": 39.683, | |
| "eval_steps_per_second": 4.962, | |
| "eval_wer": 0.9261812940672354, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.19305019305019305, | |
| "grad_norm": 3.265190362930298, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 3.9094, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.23166023166023167, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 177.1009, | |
| "eval_samples_per_second": 39.65, | |
| "eval_steps_per_second": 4.958, | |
| "eval_wer": 0.7767000376559802, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.3088803088803089, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 177.4675, | |
| "eval_samples_per_second": 39.568, | |
| "eval_steps_per_second": 4.947, | |
| "eval_wer": 0.7365379870930882, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.3861003861003861, | |
| "grad_norm": 3.0774288177490234, | |
| "learning_rate": 0.00025765714285714284, | |
| "loss": 0.9885, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3861003861003861, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 177.9576, | |
| "eval_samples_per_second": 39.459, | |
| "eval_steps_per_second": 4.934, | |
| "eval_wer": 0.6104813473050005, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.46332046332046334, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 176.7839, | |
| "eval_samples_per_second": 39.721, | |
| "eval_steps_per_second": 4.967, | |
| "eval_wer": 0.5430511731785542, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 177.8714, | |
| "eval_samples_per_second": 39.478, | |
| "eval_steps_per_second": 4.936, | |
| "eval_wer": 0.5144845675405451, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.5791505791505791, | |
| "grad_norm": 3.033381700515747, | |
| "learning_rate": 0.00021488571428571426, | |
| "loss": 0.7974, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6177606177606177, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 178.2105, | |
| "eval_samples_per_second": 39.403, | |
| "eval_steps_per_second": 4.927, | |
| "eval_wer": 0.5230675340526924, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.694980694980695, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 178.6698, | |
| "eval_samples_per_second": 39.302, | |
| "eval_steps_per_second": 4.914, | |
| "eval_wer": 0.46214275511926556, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.7722007722007722, | |
| "grad_norm": 3.2283124923706055, | |
| "learning_rate": 0.0001721142857142857, | |
| "loss": 0.7068, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7722007722007722, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 179.2886, | |
| "eval_samples_per_second": 39.166, | |
| "eval_steps_per_second": 4.897, | |
| "eval_wer": 0.4532481529092491, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.8494208494208494, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 179.5146, | |
| "eval_samples_per_second": 39.117, | |
| "eval_steps_per_second": 4.891, | |
| "eval_wer": 0.4404061651928895, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.9266409266409267, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 178.4125, | |
| "eval_samples_per_second": 39.358, | |
| "eval_steps_per_second": 4.921, | |
| "eval_wer": 0.4195006037941646, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.9652509652509652, | |
| "grad_norm": 1.9969470500946045, | |
| "learning_rate": 0.00012925714285714286, | |
| "loss": 0.6447, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0038610038610039, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 179.4629, | |
| "eval_samples_per_second": 39.128, | |
| "eval_steps_per_second": 4.892, | |
| "eval_wer": 0.4115928479607339, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 182.2946, | |
| "eval_samples_per_second": 38.52, | |
| "eval_steps_per_second": 4.816, | |
| "eval_wer": 0.38897329022372845, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.1583011583011582, | |
| "grad_norm": 0.5247506499290466, | |
| "learning_rate": 8.648571428571429e-05, | |
| "loss": 0.5318, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.1583011583011582, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.6608, | |
| "eval_samples_per_second": 38.868, | |
| "eval_steps_per_second": 4.86, | |
| "eval_wer": 0.38040330853232573, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.2355212355212355, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 178.267, | |
| "eval_samples_per_second": 39.39, | |
| "eval_steps_per_second": 4.925, | |
| "eval_wer": 0.37341747497175803, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.3127413127413128, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 179.0534, | |
| "eval_samples_per_second": 39.217, | |
| "eval_steps_per_second": 4.904, | |
| "eval_wer": 0.36753535117447705, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 0.6272704005241394, | |
| "learning_rate": 4.3799999999999994e-05, | |
| "loss": 0.494, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.3899613899613898, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 180.0451, | |
| "eval_samples_per_second": 39.001, | |
| "eval_steps_per_second": 4.877, | |
| "eval_wer": 0.3590173087660525, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.4671814671814671, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 179.7417, | |
| "eval_samples_per_second": 39.067, | |
| "eval_steps_per_second": 4.885, | |
| "eval_wer": 0.35345980548738526, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.5444015444015444, | |
| "grad_norm": 0.7731852531433105, | |
| "learning_rate": 1.0285714285714284e-06, | |
| "loss": 0.4794, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.5444015444015444, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 179.8748, | |
| "eval_samples_per_second": 39.038, | |
| "eval_steps_per_second": 4.881, | |
| "eval_wer": 0.3500967369145469, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.5444015444015444, | |
| "step": 4000, | |
| "total_flos": 8.411238439350073e+18, | |
| "train_loss": 1.0689988899230958, | |
| "train_runtime": 5961.6242, | |
| "train_samples_per_second": 10.735, | |
| "train_steps_per_second": 0.671 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.411238439350073e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |