| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 35, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 9.303076493883763, | |
| "learning_rate": 0.0, | |
| "loss": 0.9792, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 9.717944059478823, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9867, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 4.596472947786303, | |
| "learning_rate": 9.978670881475173e-06, | |
| "loss": 0.8843, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 2.3617356503729585, | |
| "learning_rate": 9.91486549841951e-06, | |
| "loss": 0.8005, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 2.2676316282563183, | |
| "learning_rate": 9.809128215864096e-06, | |
| "loss": 0.7994, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 2.246406385939518, | |
| "learning_rate": 9.66236114702178e-06, | |
| "loss": 0.7365, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.8834051955191673, | |
| "learning_rate": 9.475816456775313e-06, | |
| "loss": 0.7101, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 1.3376525812163877, | |
| "learning_rate": 9.251085678648072e-06, | |
| "loss": 0.6625, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.2857142857142856, | |
| "grad_norm": 1.0334003924250443, | |
| "learning_rate": 8.990086136401199e-06, | |
| "loss": 0.626, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.1270833913483167, | |
| "learning_rate": 8.695044586103297e-06, | |
| "loss": 0.656, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5714285714285714, | |
| "grad_norm": 0.914746603740364, | |
| "learning_rate": 8.368478218232787e-06, | |
| "loss": 0.5878, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 0.818200644259025, | |
| "learning_rate": 8.013173181896283e-06, | |
| "loss": 0.5721, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.8571428571428572, | |
| "grad_norm": 0.7861145956482819, | |
| "learning_rate": 7.63216081438678e-06, | |
| "loss": 0.5933, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.670471249816862, | |
| "learning_rate": 7.2286917788826926e-06, | |
| "loss": 0.5497, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 0.6783701796467253, | |
| "learning_rate": 6.806208330935766e-06, | |
| "loss": 0.5418, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.7130802321996181, | |
| "learning_rate": 6.368314950360416e-06, | |
| "loss": 0.5445, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.4285714285714284, | |
| "grad_norm": 0.7237611676544982, | |
| "learning_rate": 5.918747589082853e-06, | |
| "loss": 0.5367, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 0.6234236353208849, | |
| "learning_rate": 5.46134179731651e-06, | |
| "loss": 0.5187, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.7142857142857144, | |
| "grad_norm": 0.5483554337403505, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4862, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 0.5729608404090003, | |
| "learning_rate": 4.53865820268349e-06, | |
| "loss": 0.4859, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.6525783085630301, | |
| "learning_rate": 4.081252410917148e-06, | |
| "loss": 0.4726, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.6234485276585319, | |
| "learning_rate": 3.6316850496395863e-06, | |
| "loss": 0.5075, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.2857142857142856, | |
| "grad_norm": 0.5624289294983126, | |
| "learning_rate": 3.1937916690642356e-06, | |
| "loss": 0.4833, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 0.4801723823114017, | |
| "learning_rate": 2.771308221117309e-06, | |
| "loss": 0.468, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 0.5258811679866904, | |
| "learning_rate": 2.3678391856132203e-06, | |
| "loss": 0.4651, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.7644061359252237, | |
| "learning_rate": 1.9868268181037186e-06, | |
| "loss": 0.4682, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.857142857142857, | |
| "grad_norm": 0.6929173124252979, | |
| "learning_rate": 1.6315217817672142e-06, | |
| "loss": 0.4514, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5297753035187348, | |
| "learning_rate": 1.3049554138967052e-06, | |
| "loss": 0.4789, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.142857142857143, | |
| "grad_norm": 0.5415909130252555, | |
| "learning_rate": 1.0099138635988026e-06, | |
| "loss": 0.4695, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.5282132594942947, | |
| "learning_rate": 7.489143213519301e-07, | |
| "loss": 0.4236, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.428571428571429, | |
| "grad_norm": 0.5270375217775852, | |
| "learning_rate": 5.241835432246888e-07, | |
| "loss": 0.4557, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 0.47428708038578266, | |
| "learning_rate": 3.3763885297822153e-07, | |
| "loss": 0.4666, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.714285714285714, | |
| "grad_norm": 0.44077245455561864, | |
| "learning_rate": 1.908717841359048e-07, | |
| "loss": 0.4402, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 0.4840165144194508, | |
| "learning_rate": 8.513450158049109e-08, | |
| "loss": 0.4539, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.45298550256117076, | |
| "learning_rate": 2.1329118524827662e-08, | |
| "loss": 0.4573, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 35, | |
| "total_flos": 9984310984704.0, | |
| "train_loss": 0.5776977070740291, | |
| "train_runtime": 1052.4329, | |
| "train_samples_per_second": 0.48, | |
| "train_steps_per_second": 0.033 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 35, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 42, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9984310984704.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |