| { | |
| "best_metric": 0.7326732673267327, | |
| "best_model_checkpoint": "distilhubert-finetuned-not-a-word2/run-7/checkpoint-8", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 80, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.9305598735809326, | |
| "learning_rate": 1.0388926727261681e-05, | |
| "loss": 0.6973, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.68402099609375, | |
| "eval_runtime": 1.3601, | |
| "eval_samples_per_second": 47.054, | |
| "eval_steps_per_second": 5.882, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.470983624458313, | |
| "learning_rate": 1.6160552686851503e-05, | |
| "loss": 0.6748, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.3902422785758972, | |
| "learning_rate": 1.5006227494933539e-05, | |
| "loss": 0.6423, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.6836051940917969, | |
| "eval_runtime": 1.3646, | |
| "eval_samples_per_second": 46.9, | |
| "eval_steps_per_second": 5.863, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.4280204772949219, | |
| "learning_rate": 1.3851902303015575e-05, | |
| "loss": 0.6311, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.6989364624023438, | |
| "eval_runtime": 1.3707, | |
| "eval_samples_per_second": 46.693, | |
| "eval_steps_per_second": 5.837, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.3204731047153473, | |
| "learning_rate": 1.269757711109761e-05, | |
| "loss": 0.5902, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.26261427998542786, | |
| "learning_rate": 1.1543251919179645e-05, | |
| "loss": 0.5967, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.7159881591796875, | |
| "eval_runtime": 1.3893, | |
| "eval_samples_per_second": 46.065, | |
| "eval_steps_per_second": 5.758, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "grad_norm": 0.611359179019928, | |
| "learning_rate": 1.0388926727261681e-05, | |
| "loss": 0.6029, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.3782584071159363, | |
| "learning_rate": 9.234601535343717e-06, | |
| "loss": 0.5986, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.7192001342773438, | |
| "eval_runtime": 1.3984, | |
| "eval_samples_per_second": 45.766, | |
| "eval_steps_per_second": 5.721, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "grad_norm": 0.8225945234298706, | |
| "learning_rate": 8.080276343425752e-06, | |
| "loss": 0.6027, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.7113037109375, | |
| "eval_runtime": 1.3638, | |
| "eval_samples_per_second": 46.927, | |
| "eval_steps_per_second": 5.866, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.4112108051776886, | |
| "learning_rate": 6.9259511515077876e-06, | |
| "loss": 0.5633, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "grad_norm": 0.43113499879837036, | |
| "learning_rate": 5.771625959589823e-06, | |
| "loss": 0.6274, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.7047462463378906, | |
| "eval_runtime": 1.3741, | |
| "eval_samples_per_second": 46.577, | |
| "eval_steps_per_second": 5.822, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.7747747302055359, | |
| "learning_rate": 4.617300767671859e-06, | |
| "loss": 0.5591, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.70208740234375, | |
| "eval_runtime": 1.3676, | |
| "eval_samples_per_second": 46.799, | |
| "eval_steps_per_second": 5.85, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "grad_norm": 0.4591865837574005, | |
| "learning_rate": 3.4629755757538938e-06, | |
| "loss": 0.5763, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 0.6979279518127441, | |
| "learning_rate": 2.3086503838359293e-06, | |
| "loss": 0.5406, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.7020683288574219, | |
| "eval_runtime": 1.37, | |
| "eval_samples_per_second": 46.716, | |
| "eval_steps_per_second": 5.84, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "grad_norm": 0.3886575698852539, | |
| "learning_rate": 1.1543251919179647e-06, | |
| "loss": 0.5971, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.48529914021492004, | |
| "learning_rate": 0.0, | |
| "loss": 0.6026, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_f1": 0.7326732673267327, | |
| "eval_loss": 0.7013587951660156, | |
| "eval_runtime": 1.3618, | |
| "eval_samples_per_second": 46.998, | |
| "eval_steps_per_second": 5.875, | |
| "step": 80 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 80, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.0210264286168832e+16, | |
| "train_batch_size": 24, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 1.662228276361869e-05, | |
| "per_device_train_batch_size": 24 | |
| } | |
| } | |