| { | |
| "best_metric": 3.8871757984161377, | |
| "best_model_checkpoint": "./models/gpt_test9/checkpoint-10000", | |
| "epoch": 4.0, | |
| "global_step": 10592, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2.830188679245283e-05, | |
| "loss": 3.311, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 4.033947467803955, | |
| "eval_runtime": 6.7632, | |
| "eval_samples_per_second": 63.432, | |
| "eval_steps_per_second": 3.992, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 2.7041544271926144e-05, | |
| "loss": 3.1133, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 3.9776785373687744, | |
| "eval_runtime": 6.7558, | |
| "eval_samples_per_second": 63.501, | |
| "eval_steps_per_second": 3.997, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 2.3894250944187998e-05, | |
| "loss": 2.9875, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "eval_loss": 3.9545652866363525, | |
| "eval_runtime": 6.7588, | |
| "eval_samples_per_second": 63.472, | |
| "eval_steps_per_second": 3.995, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 2.0746957616449853e-05, | |
| "loss": 2.8697, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "eval_loss": 3.9269347190856934, | |
| "eval_runtime": 6.7568, | |
| "eval_samples_per_second": 63.491, | |
| "eval_steps_per_second": 3.996, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.759966428871171e-05, | |
| "loss": 2.8669, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 3.915926933288574, | |
| "eval_runtime": 6.7665, | |
| "eval_samples_per_second": 63.401, | |
| "eval_steps_per_second": 3.99, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.4452370960973563e-05, | |
| "loss": 2.7308, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 3.9065561294555664, | |
| "eval_runtime": 6.7494, | |
| "eval_samples_per_second": 63.562, | |
| "eval_steps_per_second": 4.0, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 1.1305077633235418e-05, | |
| "loss": 2.709, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "eval_loss": 3.899543285369873, | |
| "eval_runtime": 6.761, | |
| "eval_samples_per_second": 63.452, | |
| "eval_steps_per_second": 3.993, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.157784305497272e-06, | |
| "loss": 2.6979, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_loss": 3.8976171016693115, | |
| "eval_runtime": 6.7563, | |
| "eval_samples_per_second": 63.496, | |
| "eval_steps_per_second": 3.996, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 5.010490977759127e-06, | |
| "loss": 2.5878, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "eval_loss": 3.897768020629883, | |
| "eval_runtime": 6.7533, | |
| "eval_samples_per_second": 63.524, | |
| "eval_steps_per_second": 3.998, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 1.863197650020982e-06, | |
| "loss": 2.5824, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_loss": 3.8871757984161377, | |
| "eval_runtime": 6.7573, | |
| "eval_samples_per_second": 63.487, | |
| "eval_steps_per_second": 3.996, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 10592, | |
| "total_flos": 4.4280638078976e+16, | |
| "train_loss": 2.831424217569864, | |
| "train_runtime": 8708.1966, | |
| "train_samples_per_second": 19.461, | |
| "train_steps_per_second": 1.216 | |
| } | |
| ], | |
| "max_steps": 10592, | |
| "num_train_epochs": 4, | |
| "total_flos": 4.4280638078976e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |