| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.653179190751445, | |
| "eval_steps": 774, | |
| "global_step": 7740, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.965317919075145, | |
| "grad_norm": 2.1615042686462402, | |
| "learning_rate": 9.002583979328166e-06, | |
| "loss": 2.686, | |
| "step": 774 | |
| }, | |
| { | |
| "epoch": 5.965317919075145, | |
| "eval_accuracy": 0.32318619989852865, | |
| "eval_loss": 2.3315887451171875, | |
| "eval_runtime": 25.0742, | |
| "eval_samples_per_second": 314.426, | |
| "eval_steps_per_second": 15.753, | |
| "step": 774 | |
| }, | |
| { | |
| "epoch": 11.93063583815029, | |
| "grad_norm": 2.2875025272369385, | |
| "learning_rate": 8.002583979328166e-06, | |
| "loss": 2.5095, | |
| "step": 1548 | |
| }, | |
| { | |
| "epoch": 11.93063583815029, | |
| "eval_accuracy": 0.3167808219178082, | |
| "eval_loss": 2.3506176471710205, | |
| "eval_runtime": 25.0863, | |
| "eval_samples_per_second": 314.275, | |
| "eval_steps_per_second": 15.746, | |
| "step": 1548 | |
| }, | |
| { | |
| "epoch": 17.895953757225435, | |
| "grad_norm": 3.3261446952819824, | |
| "learning_rate": 7.003875968992248e-06, | |
| "loss": 2.4304, | |
| "step": 2322 | |
| }, | |
| { | |
| "epoch": 17.895953757225435, | |
| "eval_accuracy": 0.31033316421444274, | |
| "eval_loss": 2.4180381298065186, | |
| "eval_runtime": 25.0532, | |
| "eval_samples_per_second": 314.69, | |
| "eval_steps_per_second": 15.766, | |
| "step": 2322 | |
| }, | |
| { | |
| "epoch": 23.86127167630058, | |
| "grad_norm": 2.2134969234466553, | |
| "learning_rate": 6.003875968992249e-06, | |
| "loss": 2.3871, | |
| "step": 3096 | |
| }, | |
| { | |
| "epoch": 23.86127167630058, | |
| "eval_accuracy": 0.3051750380517504, | |
| "eval_loss": 2.4722793102264404, | |
| "eval_runtime": 25.1716, | |
| "eval_samples_per_second": 313.21, | |
| "eval_steps_per_second": 15.692, | |
| "step": 3096 | |
| }, | |
| { | |
| "epoch": 29.826589595375722, | |
| "grad_norm": 2.5758585929870605, | |
| "learning_rate": 5.003875968992249e-06, | |
| "loss": 2.3556, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 29.826589595375722, | |
| "eval_accuracy": 0.3, | |
| "eval_loss": 2.5127205848693848, | |
| "eval_runtime": 25.002, | |
| "eval_samples_per_second": 315.335, | |
| "eval_steps_per_second": 15.799, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 35.79190751445087, | |
| "grad_norm": 2.606534242630005, | |
| "learning_rate": 4.005167958656331e-06, | |
| "loss": 2.3325, | |
| "step": 4644 | |
| }, | |
| { | |
| "epoch": 35.79190751445087, | |
| "eval_accuracy": 0.2964865550481989, | |
| "eval_loss": 2.5233418941497803, | |
| "eval_runtime": 25.6388, | |
| "eval_samples_per_second": 307.503, | |
| "eval_steps_per_second": 15.406, | |
| "step": 4644 | |
| }, | |
| { | |
| "epoch": 41.75722543352601, | |
| "grad_norm": 2.7333884239196777, | |
| "learning_rate": 3.0051679586563307e-06, | |
| "loss": 2.3155, | |
| "step": 5418 | |
| }, | |
| { | |
| "epoch": 41.75722543352601, | |
| "eval_accuracy": 0.2930165978111184, | |
| "eval_loss": 2.5572330951690674, | |
| "eval_runtime": 25.7526, | |
| "eval_samples_per_second": 306.144, | |
| "eval_steps_per_second": 15.338, | |
| "step": 5418 | |
| }, | |
| { | |
| "epoch": 47.72254335260116, | |
| "grad_norm": 3.1406445503234863, | |
| "learning_rate": 2.0064599483204137e-06, | |
| "loss": 2.3137, | |
| "step": 6192 | |
| }, | |
| { | |
| "epoch": 47.72254335260116, | |
| "eval_accuracy": 0.29025558092338916, | |
| "eval_loss": 2.5638859272003174, | |
| "eval_runtime": 25.0057, | |
| "eval_samples_per_second": 315.288, | |
| "eval_steps_per_second": 15.796, | |
| "step": 6192 | |
| }, | |
| { | |
| "epoch": 53.6878612716763, | |
| "grad_norm": 2.6950843334198, | |
| "learning_rate": 1.0064599483204135e-06, | |
| "loss": 2.2978, | |
| "step": 6966 | |
| }, | |
| { | |
| "epoch": 53.6878612716763, | |
| "eval_accuracy": 0.2877839788037657, | |
| "eval_loss": 2.574859380722046, | |
| "eval_runtime": 25.1615, | |
| "eval_samples_per_second": 313.336, | |
| "eval_steps_per_second": 15.699, | |
| "step": 6966 | |
| }, | |
| { | |
| "epoch": 59.653179190751445, | |
| "grad_norm": 2.301842212677002, | |
| "learning_rate": 6.4599483204134375e-09, | |
| "loss": 2.2964, | |
| "step": 7740 | |
| }, | |
| { | |
| "epoch": 59.653179190751445, | |
| "eval_accuracy": 0.28579401319127345, | |
| "eval_loss": 2.5783193111419678, | |
| "eval_runtime": 24.745, | |
| "eval_samples_per_second": 318.609, | |
| "eval_steps_per_second": 15.963, | |
| "step": 7740 | |
| }, | |
| { | |
| "epoch": 59.653179190751445, | |
| "step": 7740, | |
| "total_flos": 9.747443743335875e+17, | |
| "train_loss": 2.392465943329094, | |
| "train_runtime": 26168.6752, | |
| "train_samples_per_second": 142.723, | |
| "train_steps_per_second": 0.296 | |
| } | |
| ], | |
| "logging_steps": 774, | |
| "max_steps": 7740, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.747443743335875e+17, | |
| "train_batch_size": 60, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |