| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 19200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.535132032339264e-05, | |
| "loss": 1.5294, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6997061800943186, | |
| "eval_loss": 1.505336046218872, | |
| "eval_runtime": 59.2609, | |
| "eval_samples_per_second": 128.314, | |
| "eval_steps_per_second": 0.81, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.6900880215595094e-05, | |
| "loss": 1.3178, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7119164174032121, | |
| "eval_loss": 1.4108233451843262, | |
| "eval_runtime": 57.8828, | |
| "eval_samples_per_second": 131.369, | |
| "eval_steps_per_second": 0.829, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.7807314645155048e-05, | |
| "loss": 1.2464, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.716069757803329, | |
| "eval_loss": 1.3826147317886353, | |
| "eval_runtime": 57.5, | |
| "eval_samples_per_second": 132.244, | |
| "eval_steps_per_second": 0.835, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.8450440107797548e-05, | |
| "loss": 1.2004, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7212539141360841, | |
| "eval_loss": 1.3423837423324585, | |
| "eval_runtime": 57.7225, | |
| "eval_samples_per_second": 131.734, | |
| "eval_steps_per_second": 0.832, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.894928697180815e-05, | |
| "loss": 1.1671, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.723119868068725, | |
| "eval_loss": 1.3292474746704102, | |
| "eval_runtime": 57.4277, | |
| "eval_samples_per_second": 132.41, | |
| "eval_steps_per_second": 0.836, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 1.93568745373575e-05, | |
| "loss": 1.1405, | |
| "step": 5760 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7279360245602308, | |
| "eval_loss": 1.3002619743347168, | |
| "eval_runtime": 57.5343, | |
| "eval_samples_per_second": 132.165, | |
| "eval_steps_per_second": 0.834, | |
| "step": 5760 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.9701484913790247e-05, | |
| "loss": 1.1195, | |
| "step": 6720 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7286793766001343, | |
| "eval_loss": 1.2911103963851929, | |
| "eval_runtime": 57.3971, | |
| "eval_samples_per_second": 132.481, | |
| "eval_steps_per_second": 0.836, | |
| "step": 6720 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0984, | |
| "step": 7680 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.731524743026423, | |
| "eval_loss": 1.2744665145874023, | |
| "eval_runtime": 58.0408, | |
| "eval_samples_per_second": 131.011, | |
| "eval_steps_per_second": 0.827, | |
| "step": 7680 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0802, | |
| "step": 8640 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7321408265476137, | |
| "eval_loss": 1.2718857526779175, | |
| "eval_runtime": 57.237, | |
| "eval_samples_per_second": 132.851, | |
| "eval_steps_per_second": 0.839, | |
| "step": 8640 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0678, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7324240391543068, | |
| "eval_loss": 1.261126160621643, | |
| "eval_runtime": 57.4079, | |
| "eval_samples_per_second": 132.456, | |
| "eval_steps_per_second": 0.836, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.055, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7351396479197737, | |
| "eval_loss": 1.2426605224609375, | |
| "eval_runtime": 57.4037, | |
| "eval_samples_per_second": 132.465, | |
| "eval_steps_per_second": 0.836, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0426, | |
| "step": 11520 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.738889192995377, | |
| "eval_loss": 1.226466417312622, | |
| "eval_runtime": 57.6438, | |
| "eval_samples_per_second": 131.914, | |
| "eval_steps_per_second": 0.833, | |
| "step": 11520 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0329, | |
| "step": 12480 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7372212557134691, | |
| "eval_loss": 1.234350323677063, | |
| "eval_runtime": 57.4326, | |
| "eval_samples_per_second": 132.399, | |
| "eval_steps_per_second": 0.836, | |
| "step": 12480 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0211, | |
| "step": 13440 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7388404872505433, | |
| "eval_loss": 1.2182461023330688, | |
| "eval_runtime": 57.2924, | |
| "eval_samples_per_second": 132.723, | |
| "eval_steps_per_second": 0.838, | |
| "step": 13440 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0118, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7408094590258935, | |
| "eval_loss": 1.214383602142334, | |
| "eval_runtime": 62.5016, | |
| "eval_samples_per_second": 121.661, | |
| "eval_steps_per_second": 0.768, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0038, | |
| "step": 15360 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.740329327948473, | |
| "eval_loss": 1.2094308137893677, | |
| "eval_runtime": 57.7614, | |
| "eval_samples_per_second": 131.645, | |
| "eval_steps_per_second": 0.831, | |
| "step": 15360 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9952, | |
| "step": 16320 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7427147172532104, | |
| "eval_loss": 1.1963080167770386, | |
| "eval_runtime": 57.4115, | |
| "eval_samples_per_second": 132.447, | |
| "eval_steps_per_second": 0.836, | |
| "step": 16320 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.988, | |
| "step": 17280 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7429841884959819, | |
| "eval_loss": 1.1982266902923584, | |
| "eval_runtime": 57.4423, | |
| "eval_samples_per_second": 132.376, | |
| "eval_steps_per_second": 0.836, | |
| "step": 17280 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9808, | |
| "step": 18240 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.7437883910645381, | |
| "eval_loss": 1.1920512914657593, | |
| "eval_runtime": 57.5877, | |
| "eval_samples_per_second": 132.042, | |
| "eval_steps_per_second": 0.834, | |
| "step": 18240 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9747, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7435962484051245, | |
| "eval_loss": 1.1889218091964722, | |
| "eval_runtime": 57.672, | |
| "eval_samples_per_second": 131.849, | |
| "eval_steps_per_second": 0.832, | |
| "step": 19200 | |
| } | |
| ], | |
| "max_steps": 38400, | |
| "num_train_epochs": 40, | |
| "total_flos": 3062931751174144.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |