| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0299028158484924, | |
| "eval_steps": 9, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00033225350942769335, | |
| "eval_loss": 2.1395199298858643, | |
| "eval_runtime": 220.9563, | |
| "eval_samples_per_second": 22.941, | |
| "eval_steps_per_second": 2.869, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00099676052828308, | |
| "grad_norm": 5.9584269523620605, | |
| "learning_rate": 1.5e-05, | |
| "loss": 9.4483, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00199352105656616, | |
| "grad_norm": 4.213073253631592, | |
| "learning_rate": 3e-05, | |
| "loss": 7.8813, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00299028158484924, | |
| "grad_norm": 4.823021411895752, | |
| "learning_rate": 4.5e-05, | |
| "loss": 7.9607, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00299028158484924, | |
| "eval_loss": 2.070302963256836, | |
| "eval_runtime": 221.304, | |
| "eval_samples_per_second": 22.905, | |
| "eval_steps_per_second": 2.865, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00398704211313232, | |
| "grad_norm": 7.021877765655518, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 9.0723, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0049838026414154, | |
| "grad_norm": 10.367716789245605, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 8.6656, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00598056316969848, | |
| "grad_norm": 3.396527051925659, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 6.5636, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00598056316969848, | |
| "eval_loss": 1.8593316078186035, | |
| "eval_runtime": 221.3173, | |
| "eval_samples_per_second": 22.904, | |
| "eval_steps_per_second": 2.865, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00697732369798156, | |
| "grad_norm": 5.275881290435791, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 7.3881, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00797408422626464, | |
| "grad_norm": 4.454370975494385, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 7.0281, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00897084475454772, | |
| "grad_norm": 4.078393936157227, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 6.9298, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00897084475454772, | |
| "eval_loss": 1.7337881326675415, | |
| "eval_runtime": 221.3769, | |
| "eval_samples_per_second": 22.898, | |
| "eval_steps_per_second": 2.864, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0099676052828308, | |
| "grad_norm": 4.427830696105957, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 6.7986, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01096436581111388, | |
| "grad_norm": 4.341389179229736, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 6.6719, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01196112633939696, | |
| "grad_norm": 6.339109420776367, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 8.2781, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01196112633939696, | |
| "eval_loss": 1.6621569395065308, | |
| "eval_runtime": 221.3208, | |
| "eval_samples_per_second": 22.903, | |
| "eval_steps_per_second": 2.865, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01295788686768004, | |
| "grad_norm": 4.891085147857666, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 6.0951, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.01395464739596312, | |
| "grad_norm": 4.013057231903076, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 6.5056, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0149514079242462, | |
| "grad_norm": 3.513582468032837, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 6.14, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0149514079242462, | |
| "eval_loss": 1.614381194114685, | |
| "eval_runtime": 221.2948, | |
| "eval_samples_per_second": 22.906, | |
| "eval_steps_per_second": 2.865, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.01594816845252928, | |
| "grad_norm": 4.265641212463379, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 4.8365, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.01694492898081236, | |
| "grad_norm": 4.8743062019348145, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 7.3449, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.01794168950909544, | |
| "grad_norm": 4.481264591217041, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 6.5566, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01794168950909544, | |
| "eval_loss": 1.5844690799713135, | |
| "eval_runtime": 221.4081, | |
| "eval_samples_per_second": 22.894, | |
| "eval_steps_per_second": 2.863, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01893845003737852, | |
| "grad_norm": 5.4833221435546875, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 6.6639, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.0199352105656616, | |
| "grad_norm": 6.972083568572998, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 6.7846, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02093197109394468, | |
| "grad_norm": 38.77021789550781, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 6.4933, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.02093197109394468, | |
| "eval_loss": 1.5598267316818237, | |
| "eval_runtime": 221.2927, | |
| "eval_samples_per_second": 22.906, | |
| "eval_steps_per_second": 2.865, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.02192873162222776, | |
| "grad_norm": 4.097354888916016, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 5.7498, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.02292549215051084, | |
| "grad_norm": 4.730719566345215, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 5.3061, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.02392225267879392, | |
| "grad_norm": 5.832185745239258, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 6.6548, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.02392225267879392, | |
| "eval_loss": 1.545784831047058, | |
| "eval_runtime": 221.2977, | |
| "eval_samples_per_second": 22.906, | |
| "eval_steps_per_second": 2.865, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.024919013207076998, | |
| "grad_norm": 3.9145100116729736, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 5.9105, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02591577373536008, | |
| "grad_norm": 4.379709720611572, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 7.0362, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.02691253426364316, | |
| "grad_norm": 4.382705211639404, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 5.4184, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.02691253426364316, | |
| "eval_loss": 1.537489891052246, | |
| "eval_runtime": 221.2804, | |
| "eval_samples_per_second": 22.908, | |
| "eval_steps_per_second": 2.865, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.02790929479192624, | |
| "grad_norm": 4.23976993560791, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 5.7423, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.02890605532020932, | |
| "grad_norm": 5.665456295013428, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 4.8259, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0299028158484924, | |
| "grad_norm": 7.587069988250732, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 6.1447, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0299028158484924, | |
| "eval_loss": 1.5353803634643555, | |
| "eval_runtime": 221.3193, | |
| "eval_samples_per_second": 22.904, | |
| "eval_steps_per_second": 2.865, | |
| "step": 90 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7402252117475328.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |