| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.090909090909092, | |
| "eval_steps": 500, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.9201496675098015, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.0754, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 2.2528992553888285, | |
| "learning_rate": 2e-05, | |
| "loss": 2.0635, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_loss": 1.912642002105713, | |
| "eval_runtime": 2.1083, | |
| "eval_samples_per_second": 5.218, | |
| "eval_steps_per_second": 0.474, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 1.0331613684070078, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 1.9282, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.881392002105713, | |
| "eval_runtime": 2.0915, | |
| "eval_samples_per_second": 5.259, | |
| "eval_steps_per_second": 0.478, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 0.8045841138030898, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 1.7541, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 2.265625, | |
| "eval_runtime": 1.9591, | |
| "eval_samples_per_second": 5.615, | |
| "eval_steps_per_second": 0.51, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 0.6586571154914392, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 1.5669, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.21875, | |
| "eval_runtime": 2.0902, | |
| "eval_samples_per_second": 5.263, | |
| "eval_steps_per_second": 0.478, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "grad_norm": 0.6838298205894413, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 1.3975, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "eval_loss": 2.2542612552642822, | |
| "eval_runtime": 1.9645, | |
| "eval_samples_per_second": 5.599, | |
| "eval_steps_per_second": 0.509, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "grad_norm": 0.8229750350265765, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 1.2431, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.3338067531585693, | |
| "eval_runtime": 2.0783, | |
| "eval_samples_per_second": 5.293, | |
| "eval_steps_per_second": 0.481, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "grad_norm": 0.7937525174359349, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 1.1081, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "eval_loss": 2.34375, | |
| "eval_runtime": 1.9615, | |
| "eval_samples_per_second": 5.608, | |
| "eval_steps_per_second": 0.51, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "grad_norm": 0.5774017687858329, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 1.0212, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.4275567531585693, | |
| "eval_runtime": 2.0761, | |
| "eval_samples_per_second": 5.298, | |
| "eval_steps_per_second": 0.482, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "grad_norm": 0.5148979329896105, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.9554, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "eval_loss": 2.4176137447357178, | |
| "eval_runtime": 1.9472, | |
| "eval_samples_per_second": 5.649, | |
| "eval_steps_per_second": 0.514, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 0.5705108161470246, | |
| "learning_rate": 0.0, | |
| "loss": 0.9463, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_loss": 2.4176137447357178, | |
| "eval_runtime": 2.0605, | |
| "eval_samples_per_second": 5.338, | |
| "eval_steps_per_second": 0.485, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "step": 50, | |
| "total_flos": 11674778075136.0, | |
| "train_loss": 1.3986669921875, | |
| "train_runtime": 576.3463, | |
| "train_samples_per_second": 11.677, | |
| "train_steps_per_second": 0.087 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 400, | |
| "total_flos": 11674778075136.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |