| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.8293545534924847, | |
| "eval_steps": 100, | |
| "global_step": 3200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08841732979664015, | |
| "eval_accuracy": 0.1989687629116883, | |
| "eval_loss": 4.78662633895874, | |
| "eval_runtime": 26.1083, | |
| "eval_samples_per_second": 16.7, | |
| "eval_steps_per_second": 0.536, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1768346595932803, | |
| "eval_accuracy": 0.330860012697261, | |
| "eval_loss": 4.04963493347168, | |
| "eval_runtime": 26.2098, | |
| "eval_samples_per_second": 16.635, | |
| "eval_steps_per_second": 0.534, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.26525198938992045, | |
| "eval_accuracy": 0.37791993303243177, | |
| "eval_loss": 3.652527332305908, | |
| "eval_runtime": 26.1862, | |
| "eval_samples_per_second": 16.65, | |
| "eval_steps_per_second": 0.535, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.3536693191865606, | |
| "eval_accuracy": 0.42577323076228935, | |
| "eval_loss": 3.2409958839416504, | |
| "eval_runtime": 26.1526, | |
| "eval_samples_per_second": 16.671, | |
| "eval_steps_per_second": 0.535, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4420866489832007, | |
| "grad_norm": 195.0, | |
| "learning_rate": 4.263188918361332e-05, | |
| "loss": 3.9116, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4420866489832007, | |
| "eval_accuracy": 0.39117474036699323, | |
| "eval_loss": 3.6305172443389893, | |
| "eval_runtime": 26.2335, | |
| "eval_samples_per_second": 16.62, | |
| "eval_steps_per_second": 0.534, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.5305039787798409, | |
| "eval_accuracy": 0.4406458382696409, | |
| "eval_loss": 3.177022695541382, | |
| "eval_runtime": 26.2979, | |
| "eval_samples_per_second": 16.579, | |
| "eval_steps_per_second": 0.532, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.618921308576481, | |
| "eval_accuracy": 0.5199022247813888, | |
| "eval_loss": 2.4477944374084473, | |
| "eval_runtime": 26.2814, | |
| "eval_samples_per_second": 16.59, | |
| "eval_steps_per_second": 0.533, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.7073386383731212, | |
| "eval_accuracy": 0.5508474097348983, | |
| "eval_loss": 2.238290548324585, | |
| "eval_runtime": 26.1919, | |
| "eval_samples_per_second": 16.646, | |
| "eval_steps_per_second": 0.535, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7957559681697612, | |
| "eval_accuracy": 0.5634974176979812, | |
| "eval_loss": 2.1547296047210693, | |
| "eval_runtime": 26.0445, | |
| "eval_samples_per_second": 16.741, | |
| "eval_steps_per_second": 0.538, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.8841732979664014, | |
| "grad_norm": 1.6484375, | |
| "learning_rate": 3.526377836722665e-05, | |
| "loss": 2.4568, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8841732979664014, | |
| "eval_accuracy": 0.5759096753818043, | |
| "eval_loss": 2.086787223815918, | |
| "eval_runtime": 26.1107, | |
| "eval_samples_per_second": 16.698, | |
| "eval_steps_per_second": 0.536, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.9725906277630415, | |
| "eval_accuracy": 0.5819702172527925, | |
| "eval_loss": 2.0399203300476074, | |
| "eval_runtime": 26.206, | |
| "eval_samples_per_second": 16.637, | |
| "eval_steps_per_second": 0.534, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.0610079575596818, | |
| "eval_accuracy": 0.5872521470246358, | |
| "eval_loss": 2.0102362632751465, | |
| "eval_runtime": 26.1934, | |
| "eval_samples_per_second": 16.645, | |
| "eval_steps_per_second": 0.534, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.1494252873563218, | |
| "eval_accuracy": 0.5896846032104907, | |
| "eval_loss": 1.9805158376693726, | |
| "eval_runtime": 26.091, | |
| "eval_samples_per_second": 16.711, | |
| "eval_steps_per_second": 0.537, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.237842617152962, | |
| "eval_accuracy": 0.5955343118975784, | |
| "eval_loss": 1.9590190649032593, | |
| "eval_runtime": 26.2439, | |
| "eval_samples_per_second": 16.613, | |
| "eval_steps_per_second": 0.533, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.3262599469496021, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 2.7895667550839967e-05, | |
| "loss": 1.9305, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.3262599469496021, | |
| "eval_accuracy": 0.598231793709266, | |
| "eval_loss": 1.9380624294281006, | |
| "eval_runtime": 26.2429, | |
| "eval_samples_per_second": 16.614, | |
| "eval_steps_per_second": 0.533, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.4146772767462423, | |
| "eval_accuracy": 0.5994681725660957, | |
| "eval_loss": 1.92489492893219, | |
| "eval_runtime": 26.2375, | |
| "eval_samples_per_second": 16.617, | |
| "eval_steps_per_second": 0.534, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.5030946065428825, | |
| "eval_accuracy": 0.6016819917312428, | |
| "eval_loss": 1.922256588935852, | |
| "eval_runtime": 26.2419, | |
| "eval_samples_per_second": 16.615, | |
| "eval_steps_per_second": 0.533, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.5915119363395225, | |
| "eval_accuracy": 0.6037235246198407, | |
| "eval_loss": 1.9091354608535767, | |
| "eval_runtime": 26.0733, | |
| "eval_samples_per_second": 16.722, | |
| "eval_steps_per_second": 0.537, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.6799292661361627, | |
| "eval_accuracy": 0.6042424379334493, | |
| "eval_loss": 1.9038457870483398, | |
| "eval_runtime": 26.0298, | |
| "eval_samples_per_second": 16.75, | |
| "eval_steps_per_second": 0.538, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.7683465959328029, | |
| "grad_norm": 1.703125, | |
| "learning_rate": 2.0527556734453286e-05, | |
| "loss": 1.8511, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.7683465959328029, | |
| "eval_accuracy": 0.6045487728273357, | |
| "eval_loss": 1.898189663887024, | |
| "eval_runtime": 26.1744, | |
| "eval_samples_per_second": 16.658, | |
| "eval_steps_per_second": 0.535, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.8567639257294428, | |
| "eval_accuracy": 0.60601247039271, | |
| "eval_loss": 1.8923863172531128, | |
| "eval_runtime": 26.2245, | |
| "eval_samples_per_second": 16.626, | |
| "eval_steps_per_second": 0.534, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.9451812555260832, | |
| "eval_accuracy": 0.6072253419074359, | |
| "eval_loss": 1.8843563795089722, | |
| "eval_runtime": 26.2182, | |
| "eval_samples_per_second": 16.63, | |
| "eval_steps_per_second": 0.534, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.033598585322723, | |
| "eval_accuracy": 0.6086919582677255, | |
| "eval_loss": 1.8872811794281006, | |
| "eval_runtime": 26.2095, | |
| "eval_samples_per_second": 16.635, | |
| "eval_steps_per_second": 0.534, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.1220159151193636, | |
| "eval_accuracy": 0.606785049583925, | |
| "eval_loss": 1.8889071941375732, | |
| "eval_runtime": 26.1074, | |
| "eval_samples_per_second": 16.7, | |
| "eval_steps_per_second": 0.536, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.2104332449160036, | |
| "grad_norm": 1.9609375, | |
| "learning_rate": 1.315944591806661e-05, | |
| "loss": 1.8197, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.2104332449160036, | |
| "eval_accuracy": 0.6080460875913235, | |
| "eval_loss": 1.8847737312316895, | |
| "eval_runtime": 26.1528, | |
| "eval_samples_per_second": 16.671, | |
| "eval_steps_per_second": 0.535, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.2988505747126435, | |
| "eval_accuracy": 0.6090524026863879, | |
| "eval_loss": 1.8735781908035278, | |
| "eval_runtime": 26.1209, | |
| "eval_samples_per_second": 16.692, | |
| "eval_steps_per_second": 0.536, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.387267904509284, | |
| "eval_accuracy": 0.6072484329163864, | |
| "eval_loss": 1.8857922554016113, | |
| "eval_runtime": 26.0234, | |
| "eval_samples_per_second": 16.754, | |
| "eval_steps_per_second": 0.538, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.475685234305924, | |
| "eval_accuracy": 0.6088222983902897, | |
| "eval_loss": 1.8814462423324585, | |
| "eval_runtime": 26.1989, | |
| "eval_samples_per_second": 16.642, | |
| "eval_steps_per_second": 0.534, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.564102564102564, | |
| "eval_accuracy": 0.6103062409699992, | |
| "eval_loss": 1.8648868799209595, | |
| "eval_runtime": 26.2353, | |
| "eval_samples_per_second": 16.619, | |
| "eval_steps_per_second": 0.534, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.6525198938992043, | |
| "grad_norm": 1.6640625, | |
| "learning_rate": 5.7913351016799295e-06, | |
| "loss": 1.8116, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.6525198938992043, | |
| "eval_accuracy": 0.6091320072332731, | |
| "eval_loss": 1.8647068738937378, | |
| "eval_runtime": 26.1308, | |
| "eval_samples_per_second": 16.685, | |
| "eval_steps_per_second": 0.536, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.7409372236958442, | |
| "eval_accuracy": 0.6100501488617454, | |
| "eval_loss": 1.8755285739898682, | |
| "eval_runtime": 26.2337, | |
| "eval_samples_per_second": 16.62, | |
| "eval_steps_per_second": 0.534, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.8293545534924847, | |
| "eval_accuracy": 0.6099107960519622, | |
| "eval_loss": 1.8754957914352417, | |
| "eval_runtime": 26.1992, | |
| "eval_samples_per_second": 16.642, | |
| "eval_steps_per_second": 0.534, | |
| "step": 3200 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3393, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "total_flos": 1.6292390296249958e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |