| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 25.098039215686274, | |
| "eval_steps": 500, | |
| "global_step": 40, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 2.3024485111236572, | |
| "eval_runtime": 52.055, | |
| "eval_samples_per_second": 1.594, | |
| "eval_steps_per_second": 0.807, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_loss": 2.2997043132781982, | |
| "eval_runtime": 52.1609, | |
| "eval_samples_per_second": 1.591, | |
| "eval_steps_per_second": 0.805, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "eval_loss": 2.2976605892181396, | |
| "eval_runtime": 52.0558, | |
| "eval_samples_per_second": 1.594, | |
| "eval_steps_per_second": 0.807, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "grad_norm": 0.0927734375, | |
| "learning_rate": 1.9238795325112867e-05, | |
| "loss": 2.2951, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "eval_loss": 2.293534278869629, | |
| "eval_runtime": 52.2245, | |
| "eval_samples_per_second": 1.589, | |
| "eval_steps_per_second": 0.804, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "eval_loss": 2.290515184402466, | |
| "eval_runtime": 52.068, | |
| "eval_samples_per_second": 1.594, | |
| "eval_steps_per_second": 0.807, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "eval_loss": 2.2852978706359863, | |
| "eval_runtime": 52.0381, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "grad_norm": 0.095703125, | |
| "learning_rate": 1.7071067811865477e-05, | |
| "loss": 2.2845, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_loss": 2.2788708209991455, | |
| "eval_runtime": 52.4054, | |
| "eval_samples_per_second": 1.584, | |
| "eval_steps_per_second": 0.801, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "eval_loss": 2.275712251663208, | |
| "eval_runtime": 52.0276, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "eval_loss": 2.2695822715759277, | |
| "eval_runtime": 52.0264, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "grad_norm": 0.09375, | |
| "learning_rate": 1.3826834323650899e-05, | |
| "loss": 2.2729, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "eval_loss": 2.266735553741455, | |
| "eval_runtime": 52.0513, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 10.67, | |
| "eval_loss": 2.2618672847747803, | |
| "eval_runtime": 52.0274, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 11.92, | |
| "eval_loss": 2.2575833797454834, | |
| "eval_runtime": 52.0229, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "grad_norm": 0.08984375, | |
| "learning_rate": 1e-05, | |
| "loss": 2.2599, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "eval_loss": 2.2554843425750732, | |
| "eval_runtime": 52.0166, | |
| "eval_samples_per_second": 1.596, | |
| "eval_steps_per_second": 0.807, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 13.8, | |
| "eval_loss": 2.2518961429595947, | |
| "eval_runtime": 52.0251, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 14.43, | |
| "eval_loss": 2.250413179397583, | |
| "eval_runtime": 52.2375, | |
| "eval_samples_per_second": 1.589, | |
| "eval_steps_per_second": 0.804, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 15.69, | |
| "grad_norm": 0.0869140625, | |
| "learning_rate": 6.173165676349103e-06, | |
| "loss": 2.2523, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 15.69, | |
| "eval_loss": 2.2478551864624023, | |
| "eval_runtime": 52.034, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "eval_loss": 2.2462329864501953, | |
| "eval_runtime": 52.0568, | |
| "eval_samples_per_second": 1.594, | |
| "eval_steps_per_second": 0.807, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 17.57, | |
| "eval_loss": 2.245408296585083, | |
| "eval_runtime": 52.0465, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "grad_norm": 0.087890625, | |
| "learning_rate": 2.9289321881345257e-06, | |
| "loss": 2.2471, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "eval_loss": 2.2442374229431152, | |
| "eval_runtime": 52.0648, | |
| "eval_samples_per_second": 1.594, | |
| "eval_steps_per_second": 0.807, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 19.45, | |
| "eval_loss": 2.2436985969543457, | |
| "eval_runtime": 52.0526, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 20.71, | |
| "eval_loss": 2.2432022094726562, | |
| "eval_runtime": 52.0653, | |
| "eval_samples_per_second": 1.594, | |
| "eval_steps_per_second": 0.807, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 21.96, | |
| "grad_norm": 0.08349609375, | |
| "learning_rate": 7.612046748871327e-07, | |
| "loss": 2.2444, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 21.96, | |
| "eval_loss": 2.24267315864563, | |
| "eval_runtime": 52.0344, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 22.59, | |
| "eval_loss": 2.2428243160247803, | |
| "eval_runtime": 52.041, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 23.84, | |
| "eval_loss": 2.2428712844848633, | |
| "eval_runtime": 52.0448, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "eval_loss": 2.2428550720214844, | |
| "eval_runtime": 52.0283, | |
| "eval_samples_per_second": 1.595, | |
| "eval_steps_per_second": 0.807, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 25.1, | |
| "grad_norm": 0.08642578125, | |
| "learning_rate": 0.0, | |
| "loss": 2.2427, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 25.1, | |
| "eval_loss": 2.2429935932159424, | |
| "eval_runtime": 52.018, | |
| "eval_samples_per_second": 1.596, | |
| "eval_steps_per_second": 0.807, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 25.1, | |
| "step": 40, | |
| "total_flos": 5.866939585627423e+18, | |
| "train_loss": 2.2623767852783203, | |
| "train_runtime": 20272.0714, | |
| "train_samples_per_second": 0.805, | |
| "train_steps_per_second": 0.002 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 40, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 5.866939585627423e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |