| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06830892712291338, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013661785424582676, | |
| "grad_norm": 0.44257715344429016, | |
| "learning_rate": 5e-05, | |
| "loss": 2.692, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013661785424582676, | |
| "eval_loss": 5.153928279876709, | |
| "eval_runtime": 3.4156, | |
| "eval_samples_per_second": 14.639, | |
| "eval_steps_per_second": 3.806, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002732357084916535, | |
| "grad_norm": 0.7517432570457458, | |
| "learning_rate": 0.0001, | |
| "loss": 3.4072, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.004098535627374803, | |
| "grad_norm": 0.8371504545211792, | |
| "learning_rate": 9.990365154573717e-05, | |
| "loss": 3.6247, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00546471416983307, | |
| "grad_norm": 1.030019998550415, | |
| "learning_rate": 9.961501876182148e-05, | |
| "loss": 3.8374, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.006830892712291338, | |
| "grad_norm": 1.288458228111267, | |
| "learning_rate": 9.913533761814537e-05, | |
| "loss": 4.0069, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008197071254749605, | |
| "grad_norm": 1.3096429109573364, | |
| "learning_rate": 9.846666218300807e-05, | |
| "loss": 3.9824, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.009563249797207872, | |
| "grad_norm": 1.4815852642059326, | |
| "learning_rate": 9.761185582727977e-05, | |
| "loss": 4.0691, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01092942833966614, | |
| "grad_norm": 1.670554280281067, | |
| "learning_rate": 9.657457896300791e-05, | |
| "loss": 4.0622, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.012295606882124407, | |
| "grad_norm": 2.384780168533325, | |
| "learning_rate": 9.535927336897098e-05, | |
| "loss": 4.1641, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013661785424582676, | |
| "grad_norm": 2.171551465988159, | |
| "learning_rate": 9.397114317029975e-05, | |
| "loss": 4.0371, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.015027963967040942, | |
| "grad_norm": 2.5893049240112305, | |
| "learning_rate": 9.241613255361455e-05, | |
| "loss": 4.323, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.01639414250949921, | |
| "grad_norm": 3.286071538925171, | |
| "learning_rate": 9.070090031310558e-05, | |
| "loss": 4.489, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01776032105195748, | |
| "grad_norm": 2.760322332382202, | |
| "learning_rate": 8.883279133655399e-05, | |
| "loss": 2.6212, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.019126499594415744, | |
| "grad_norm": 2.987971782684326, | |
| "learning_rate": 8.681980515339464e-05, | |
| "loss": 2.9064, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.020492678136874012, | |
| "grad_norm": 2.283176898956299, | |
| "learning_rate": 8.467056167950311e-05, | |
| "loss": 3.2186, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02185885667933228, | |
| "grad_norm": 1.3560705184936523, | |
| "learning_rate": 8.239426430539243e-05, | |
| "loss": 3.1604, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02322503522179055, | |
| "grad_norm": 1.1246516704559326, | |
| "learning_rate": 8.000066048588211e-05, | |
| "loss": 3.1155, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.024591213764248814, | |
| "grad_norm": 1.4898251295089722, | |
| "learning_rate": 7.75e-05, | |
| "loss": 3.4359, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.025957392306707083, | |
| "grad_norm": 1.5308696031570435, | |
| "learning_rate": 7.490299105985507e-05, | |
| "loss": 3.3464, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.02732357084916535, | |
| "grad_norm": 1.7177456617355347, | |
| "learning_rate": 7.222075445642904e-05, | |
| "loss": 3.7012, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02868974939162362, | |
| "grad_norm": 1.730107069015503, | |
| "learning_rate": 6.946477593864228e-05, | |
| "loss": 3.7408, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.030055927934081884, | |
| "grad_norm": 1.8491281270980835, | |
| "learning_rate": 6.664685702961344e-05, | |
| "loss": 3.834, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03142210647654015, | |
| "grad_norm": 1.7509467601776123, | |
| "learning_rate": 6.377906449072578e-05, | |
| "loss": 3.9101, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.03278828501899842, | |
| "grad_norm": 2.118955373764038, | |
| "learning_rate": 6.087367864990233e-05, | |
| "loss": 3.9045, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03415446356145669, | |
| "grad_norm": 2.91212797164917, | |
| "learning_rate": 5.794314081535644e-05, | |
| "loss": 4.185, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03415446356145669, | |
| "eval_loss": 3.4046623706817627, | |
| "eval_runtime": 3.487, | |
| "eval_samples_per_second": 14.339, | |
| "eval_steps_per_second": 3.728, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03552064210391496, | |
| "grad_norm": 1.5169587135314941, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 2.3497, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03688682064637322, | |
| "grad_norm": 2.4556190967559814, | |
| "learning_rate": 5.205685918464356e-05, | |
| "loss": 2.9366, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03825299918883149, | |
| "grad_norm": 2.6385529041290283, | |
| "learning_rate": 4.912632135009769e-05, | |
| "loss": 2.97, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.039619177731289756, | |
| "grad_norm": 2.0972061157226562, | |
| "learning_rate": 4.6220935509274235e-05, | |
| "loss": 3.172, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.040985356273748025, | |
| "grad_norm": 1.6529752016067505, | |
| "learning_rate": 4.3353142970386564e-05, | |
| "loss": 3.3134, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04235153481620629, | |
| "grad_norm": 1.3257697820663452, | |
| "learning_rate": 4.053522406135775e-05, | |
| "loss": 3.3341, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04371771335866456, | |
| "grad_norm": 1.5197049379348755, | |
| "learning_rate": 3.777924554357096e-05, | |
| "loss": 3.3917, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04508389190112283, | |
| "grad_norm": 1.694000244140625, | |
| "learning_rate": 3.509700894014496e-05, | |
| "loss": 3.6008, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0464500704435811, | |
| "grad_norm": 1.9344284534454346, | |
| "learning_rate": 3.250000000000001e-05, | |
| "loss": 3.4473, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04781624898603936, | |
| "grad_norm": 2.178778648376465, | |
| "learning_rate": 2.9999339514117912e-05, | |
| "loss": 3.5018, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04918242752849763, | |
| "grad_norm": 2.6502668857574463, | |
| "learning_rate": 2.760573569460757e-05, | |
| "loss": 3.7114, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0505486060709559, | |
| "grad_norm": 2.9071402549743652, | |
| "learning_rate": 2.53294383204969e-05, | |
| "loss": 4.0033, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.051914784613414165, | |
| "grad_norm": 0.6881694793701172, | |
| "learning_rate": 2.3180194846605367e-05, | |
| "loss": 2.4487, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.053280963155872434, | |
| "grad_norm": 0.6594279408454895, | |
| "learning_rate": 2.1167208663446025e-05, | |
| "loss": 2.6734, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0546471416983307, | |
| "grad_norm": 0.7833492159843445, | |
| "learning_rate": 1.9299099686894423e-05, | |
| "loss": 2.8635, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05601332024078897, | |
| "grad_norm": 0.9415630102157593, | |
| "learning_rate": 1.758386744638546e-05, | |
| "loss": 3.1077, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.05737949878324724, | |
| "grad_norm": 0.9522256255149841, | |
| "learning_rate": 1.602885682970026e-05, | |
| "loss": 3.099, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0587456773257055, | |
| "grad_norm": 0.9785342812538147, | |
| "learning_rate": 1.464072663102903e-05, | |
| "loss": 3.0967, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.06011185586816377, | |
| "grad_norm": 1.0899083614349365, | |
| "learning_rate": 1.3425421036992098e-05, | |
| "loss": 3.2681, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.06147803441062204, | |
| "grad_norm": 1.2303138971328735, | |
| "learning_rate": 1.2388144172720251e-05, | |
| "loss": 3.3943, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0628442129530803, | |
| "grad_norm": 1.5075228214263916, | |
| "learning_rate": 1.1533337816991932e-05, | |
| "loss": 3.438, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06421039149553857, | |
| "grad_norm": 1.4767671823501587, | |
| "learning_rate": 1.0864662381854632e-05, | |
| "loss": 3.6421, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.06557657003799684, | |
| "grad_norm": 1.618712306022644, | |
| "learning_rate": 1.0384981238178534e-05, | |
| "loss": 3.5912, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06694274858045511, | |
| "grad_norm": 1.9819996356964111, | |
| "learning_rate": 1.0096348454262845e-05, | |
| "loss": 3.6884, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.06830892712291338, | |
| "grad_norm": 3.463073968887329, | |
| "learning_rate": 1e-05, | |
| "loss": 4.0977, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06830892712291338, | |
| "eval_loss": 3.257638931274414, | |
| "eval_runtime": 3.4994, | |
| "eval_samples_per_second": 14.288, | |
| "eval_steps_per_second": 3.715, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.969948642440643e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |