| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3878550440744368, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0039177277179236044, | |
| "eval_loss": 10.374105453491211, | |
| "eval_runtime": 0.767, | |
| "eval_samples_per_second": 560.637, | |
| "eval_steps_per_second": 70.406, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011753183153770812, | |
| "grad_norm": 0.04419707879424095, | |
| "learning_rate": 3e-05, | |
| "loss": 10.3757, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.023506366307541625, | |
| "grad_norm": 0.03822525963187218, | |
| "learning_rate": 6e-05, | |
| "loss": 10.3745, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03525954946131244, | |
| "grad_norm": 0.03418227285146713, | |
| "learning_rate": 9e-05, | |
| "loss": 10.3751, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03525954946131244, | |
| "eval_loss": 10.373617172241211, | |
| "eval_runtime": 0.7461, | |
| "eval_samples_per_second": 576.299, | |
| "eval_steps_per_second": 72.372, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04701273261508325, | |
| "grad_norm": 0.04289775341749191, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 10.3756, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.058765915768854066, | |
| "grad_norm": 0.03956589475274086, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 10.3729, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07051909892262488, | |
| "grad_norm": 0.0488935187458992, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 10.3732, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07051909892262488, | |
| "eval_loss": 10.37240982055664, | |
| "eval_runtime": 0.7837, | |
| "eval_samples_per_second": 548.666, | |
| "eval_steps_per_second": 68.902, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08227228207639568, | |
| "grad_norm": 0.033634647727012634, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 10.3756, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0940254652301665, | |
| "grad_norm": 0.05794714018702507, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 10.373, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.10577864838393732, | |
| "grad_norm": 0.035457514226436615, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 10.3694, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10577864838393732, | |
| "eval_loss": 10.371187210083008, | |
| "eval_runtime": 0.7714, | |
| "eval_samples_per_second": 557.405, | |
| "eval_steps_per_second": 70.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.11753183153770813, | |
| "grad_norm": 0.04388291388750076, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 10.3715, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.12928501469147893, | |
| "grad_norm": 0.04818139225244522, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 10.3729, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.14103819784524976, | |
| "grad_norm": 0.04022936895489693, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 10.3716, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.14103819784524976, | |
| "eval_loss": 10.369961738586426, | |
| "eval_runtime": 0.7491, | |
| "eval_samples_per_second": 574.0, | |
| "eval_steps_per_second": 72.084, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.15279138099902057, | |
| "grad_norm": 0.05035528540611267, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 10.3703, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.16454456415279137, | |
| "grad_norm": 0.04766154661774635, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 10.3692, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.1762977473065622, | |
| "grad_norm": 0.03941637650132179, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 10.3696, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1762977473065622, | |
| "eval_loss": 10.368770599365234, | |
| "eval_runtime": 0.784, | |
| "eval_samples_per_second": 548.49, | |
| "eval_steps_per_second": 68.88, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.188050930460333, | |
| "grad_norm": 0.03891783580183983, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 10.3699, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.19980411361410383, | |
| "grad_norm": 0.04591111093759537, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 10.3683, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.21155729676787463, | |
| "grad_norm": 0.06579438596963882, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 10.3693, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.21155729676787463, | |
| "eval_loss": 10.367687225341797, | |
| "eval_runtime": 0.7705, | |
| "eval_samples_per_second": 558.084, | |
| "eval_steps_per_second": 70.085, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.22331047992164543, | |
| "grad_norm": 0.048035670071840286, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 10.3687, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.23506366307541626, | |
| "grad_norm": 0.046677302569150925, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 10.3667, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.24681684622918706, | |
| "grad_norm": 0.04819753021001816, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 10.3666, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.24681684622918706, | |
| "eval_loss": 10.366778373718262, | |
| "eval_runtime": 0.7477, | |
| "eval_samples_per_second": 575.108, | |
| "eval_steps_per_second": 72.223, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.25857002938295787, | |
| "grad_norm": 0.05676067993044853, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 10.3669, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.2703232125367287, | |
| "grad_norm": 0.05819647014141083, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 10.3661, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.2820763956904995, | |
| "grad_norm": 0.056551020592451096, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 10.3681, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2820763956904995, | |
| "eval_loss": 10.366111755371094, | |
| "eval_runtime": 0.7821, | |
| "eval_samples_per_second": 549.813, | |
| "eval_steps_per_second": 69.046, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.2938295788442703, | |
| "grad_norm": 0.06585326790809631, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 10.3683, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.30558276199804113, | |
| "grad_norm": 0.04930694028735161, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 10.365, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.31733594515181196, | |
| "grad_norm": 0.06866071373224258, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 10.367, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.31733594515181196, | |
| "eval_loss": 10.365713119506836, | |
| "eval_runtime": 0.7481, | |
| "eval_samples_per_second": 574.788, | |
| "eval_steps_per_second": 72.183, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.32908912830558273, | |
| "grad_norm": 0.04585767164826393, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 10.3658, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.34084231145935356, | |
| "grad_norm": 0.05678807944059372, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 10.367, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.3525954946131244, | |
| "grad_norm": 0.056517504155635834, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 10.3663, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3525954946131244, | |
| "eval_loss": 10.365554809570312, | |
| "eval_runtime": 0.754, | |
| "eval_samples_per_second": 570.3, | |
| "eval_steps_per_second": 71.619, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3643486777668952, | |
| "grad_norm": 0.05382026731967926, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 10.366, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.376101860920666, | |
| "grad_norm": 0.05993453785777092, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 10.3649, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.3878550440744368, | |
| "grad_norm": 0.05342447757720947, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 10.3646, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.3878550440744368, | |
| "eval_loss": 10.365524291992188, | |
| "eval_runtime": 0.7745, | |
| "eval_samples_per_second": 555.185, | |
| "eval_steps_per_second": 69.721, | |
| "step": 99 | |
| } | |
| ], | |
| "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": 10355884425216.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |