| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.001994853278541363, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.9897065570827266e-05, | |
| "grad_norm": 38.317012786865234, | |
| "learning_rate": 2e-05, | |
| "loss": 12.727, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 3.9897065570827266e-05, | |
| "eval_loss": 12.322999954223633, | |
| "eval_runtime": 900.735, | |
| "eval_samples_per_second": 11.717, | |
| "eval_steps_per_second": 5.859, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 7.979413114165453e-05, | |
| "grad_norm": 33.52197265625, | |
| "learning_rate": 4e-05, | |
| "loss": 12.4489, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0001196911967124818, | |
| "grad_norm": 36.110511779785156, | |
| "learning_rate": 6e-05, | |
| "loss": 12.454, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00015958826228330906, | |
| "grad_norm": 38.46061706542969, | |
| "learning_rate": 8e-05, | |
| "loss": 11.6869, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00019948532785413634, | |
| "grad_norm": 38.2817268371582, | |
| "learning_rate": 0.0001, | |
| "loss": 9.5625, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00019948532785413634, | |
| "eval_loss": 6.2894415855407715, | |
| "eval_runtime": 904.1441, | |
| "eval_samples_per_second": 11.673, | |
| "eval_steps_per_second": 5.836, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0002393823934249636, | |
| "grad_norm": 37.16168212890625, | |
| "learning_rate": 0.00012, | |
| "loss": 6.1292, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00027927945899579087, | |
| "grad_norm": 49.690555572509766, | |
| "learning_rate": 0.00014, | |
| "loss": 2.0369, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0003191765245666181, | |
| "grad_norm": 7.405727863311768, | |
| "learning_rate": 0.00016, | |
| "loss": 1.4215, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0003590735901374454, | |
| "grad_norm": 0.07706335932016373, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0011, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0003989706557082727, | |
| "grad_norm": 0.05455558001995087, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0008, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0003989706557082727, | |
| "eval_loss": 0.2440386563539505, | |
| "eval_runtime": 904.1353, | |
| "eval_samples_per_second": 11.673, | |
| "eval_steps_per_second": 5.837, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00043886772127909993, | |
| "grad_norm": 0.08208052068948746, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.0009, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0004787647868499272, | |
| "grad_norm": 5.9814772605896, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.4459, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0005186618524207544, | |
| "grad_norm": 11.527839660644531, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.2465, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0005585589179915817, | |
| "grad_norm": 1.4747658967971802, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0184, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0005984559835624089, | |
| "grad_norm": 5.572564601898193, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.5235, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0005984559835624089, | |
| "eval_loss": 0.22381387650966644, | |
| "eval_runtime": 904.0143, | |
| "eval_samples_per_second": 11.675, | |
| "eval_steps_per_second": 5.837, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0006383530491332362, | |
| "grad_norm": 6.26619291305542, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.3056, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0006782501147040636, | |
| "grad_norm": 3.3372812271118164, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.1254, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0007181471802748908, | |
| "grad_norm": 17.318269729614258, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.4793, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0007580442458457181, | |
| "grad_norm": 6.572148323059082, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.1594, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0007979413114165454, | |
| "grad_norm": 0.29391029477119446, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.0067, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0007979413114165454, | |
| "eval_loss": 0.07681792229413986, | |
| "eval_runtime": 904.2567, | |
| "eval_samples_per_second": 11.671, | |
| "eval_steps_per_second": 5.836, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0008378383769873726, | |
| "grad_norm": 0.07184968143701553, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.0014, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0008777354425581999, | |
| "grad_norm": 1.5820742845535278, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0236, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0009176325081290271, | |
| "grad_norm": 0.025467336177825928, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.0003, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0009575295736998544, | |
| "grad_norm": 7.2011308670043945, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.6861, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0009974266392706816, | |
| "grad_norm": 6.748629570007324, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.4156, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0009974266392706816, | |
| "eval_loss": 0.05889046937227249, | |
| "eval_runtime": 903.8768, | |
| "eval_samples_per_second": 11.676, | |
| "eval_steps_per_second": 5.838, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0010373237048415089, | |
| "grad_norm": 0.7254471182823181, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0109, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0010772207704123362, | |
| "grad_norm": 0.04494161903858185, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.0011, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0011171178359831635, | |
| "grad_norm": 0.11085151880979538, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.003, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0011570149015539908, | |
| "grad_norm": 0.44602158665657043, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.006, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0011969119671248179, | |
| "grad_norm": 0.7372945547103882, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0106, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0011969119671248179, | |
| "eval_loss": 0.058843594044446945, | |
| "eval_runtime": 904.0789, | |
| "eval_samples_per_second": 11.674, | |
| "eval_steps_per_second": 5.837, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0012368090326956452, | |
| "grad_norm": 0.4658772349357605, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.011, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0012767060982664725, | |
| "grad_norm": 4.403354167938232, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.0919, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0013166031638372998, | |
| "grad_norm": 0.10470008105039597, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.0022, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0013565002294081271, | |
| "grad_norm": 0.1977054476737976, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0033, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0013963972949789542, | |
| "grad_norm": 0.08764846622943878, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0016, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0013963972949789542, | |
| "eval_loss": 0.07193881273269653, | |
| "eval_runtime": 904.2295, | |
| "eval_samples_per_second": 11.672, | |
| "eval_steps_per_second": 5.836, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0014362943605497815, | |
| "grad_norm": 0.36496585607528687, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.0061, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0014761914261206088, | |
| "grad_norm": 4.621776580810547, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.2746, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0015160884916914361, | |
| "grad_norm": 0.016069525852799416, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0004, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0015559855572622634, | |
| "grad_norm": 0.009612571448087692, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.0002, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0015958826228330907, | |
| "grad_norm": 5.677269458770752, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.4147, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0015958826228330907, | |
| "eval_loss": 0.07546740025281906, | |
| "eval_runtime": 904.2379, | |
| "eval_samples_per_second": 11.672, | |
| "eval_steps_per_second": 5.836, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0016357796884039178, | |
| "grad_norm": 0.0074098496697843075, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.0002, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0016756767539747451, | |
| "grad_norm": 4.771692752838135, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.3291, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0017155738195455724, | |
| "grad_norm": 3.2575714588165283, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.4221, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0017554708851163997, | |
| "grad_norm": 0.08677243441343307, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0016, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.001795367950687227, | |
| "grad_norm": 0.0076282089576125145, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.0002, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.001795367950687227, | |
| "eval_loss": 0.06367920339107513, | |
| "eval_runtime": 904.1364, | |
| "eval_samples_per_second": 11.673, | |
| "eval_steps_per_second": 5.837, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0018352650162580541, | |
| "grad_norm": 3.6709728240966797, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.3988, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0018751620818288814, | |
| "grad_norm": 3.651967763900757, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.052, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0019150591473997087, | |
| "grad_norm": 3.8636350631713867, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.4492, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.001954956212970536, | |
| "grad_norm": 5.142552375793457, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0663, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.001994853278541363, | |
| "grad_norm": 0.24663817882537842, | |
| "learning_rate": 0.0, | |
| "loss": 0.0055, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.001994853278541363, | |
| "eval_loss": 0.06045769900083542, | |
| "eval_runtime": 904.2849, | |
| "eval_samples_per_second": 11.671, | |
| "eval_steps_per_second": 5.836, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 70, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.84956424814592e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |