| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.031203682034480068, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0003120368203448007, | |
| "eval_loss": 2.0700910091400146, | |
| "eval_runtime": 99.8915, | |
| "eval_samples_per_second": 54.039, | |
| "eval_steps_per_second": 6.757, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.000936110461034402, | |
| "grad_norm": 0.7371235489845276, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.0984, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.001872220922068804, | |
| "grad_norm": 0.8994637131690979, | |
| "learning_rate": 3e-05, | |
| "loss": 2.0438, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.002808331383103206, | |
| "grad_norm": 0.9350093007087708, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.981, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.002808331383103206, | |
| "eval_loss": 2.015099048614502, | |
| "eval_runtime": 100.6931, | |
| "eval_samples_per_second": 53.608, | |
| "eval_steps_per_second": 6.704, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.003744441844137608, | |
| "grad_norm": 1.0670744180679321, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.9626, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00468055230517201, | |
| "grad_norm": 0.9438199996948242, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.8207, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005616662766206412, | |
| "grad_norm": 0.8364858031272888, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.8106, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.005616662766206412, | |
| "eval_loss": 1.6422984600067139, | |
| "eval_runtime": 99.4187, | |
| "eval_samples_per_second": 54.296, | |
| "eval_steps_per_second": 6.789, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.006552773227240814, | |
| "grad_norm": 0.6442909240722656, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.5737, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.007488883688275216, | |
| "grad_norm": 0.6347518563270569, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.6011, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.008424994149309618, | |
| "grad_norm": 0.7162268161773682, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.4523, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.008424994149309618, | |
| "eval_loss": 1.364453673362732, | |
| "eval_runtime": 100.9594, | |
| "eval_samples_per_second": 53.467, | |
| "eval_steps_per_second": 6.686, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00936110461034402, | |
| "grad_norm": 0.586948812007904, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.315, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.010297215071378422, | |
| "grad_norm": 0.5078171491622925, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.404, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.011233325532412824, | |
| "grad_norm": 0.42401692271232605, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.2328, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.011233325532412824, | |
| "eval_loss": 1.2021371126174927, | |
| "eval_runtime": 101.4814, | |
| "eval_samples_per_second": 53.192, | |
| "eval_steps_per_second": 6.651, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.012169435993447227, | |
| "grad_norm": 0.4761180579662323, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.2958, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.013105546454481629, | |
| "grad_norm": 0.4139047861099243, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.2068, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.01404165691551603, | |
| "grad_norm": 0.45918866991996765, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 1.0381, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.01404165691551603, | |
| "eval_loss": 1.118904948234558, | |
| "eval_runtime": 98.9835, | |
| "eval_samples_per_second": 54.534, | |
| "eval_steps_per_second": 6.819, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.014977767376550433, | |
| "grad_norm": 0.4343072772026062, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.0742, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.015913877837584835, | |
| "grad_norm": 0.428012490272522, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.092, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.016849988298619237, | |
| "grad_norm": 0.3722698986530304, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.974, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.016849988298619237, | |
| "eval_loss": 1.0687353610992432, | |
| "eval_runtime": 99.0201, | |
| "eval_samples_per_second": 54.514, | |
| "eval_steps_per_second": 6.817, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01778609875965364, | |
| "grad_norm": 0.40713074803352356, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.1488, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.01872220922068804, | |
| "grad_norm": 0.4211384057998657, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.0259, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.019658319681722443, | |
| "grad_norm": 0.3947371244430542, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 1.0927, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.019658319681722443, | |
| "eval_loss": 1.0422518253326416, | |
| "eval_runtime": 98.9882, | |
| "eval_samples_per_second": 54.532, | |
| "eval_steps_per_second": 6.819, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.020594430142756845, | |
| "grad_norm": 0.3636433780193329, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.9972, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.021530540603791247, | |
| "grad_norm": 0.40044713020324707, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 1.0174, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.02246665106482565, | |
| "grad_norm": 0.40297916531562805, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 1.0661, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.02246665106482565, | |
| "eval_loss": 1.0281081199645996, | |
| "eval_runtime": 98.8829, | |
| "eval_samples_per_second": 54.59, | |
| "eval_steps_per_second": 6.826, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.02340276152586005, | |
| "grad_norm": 0.37431633472442627, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 1.0517, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.024338871986894453, | |
| "grad_norm": 0.41993048787117004, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 1.0507, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.025274982447928855, | |
| "grad_norm": 0.3987724781036377, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 1.0236, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.025274982447928855, | |
| "eval_loss": 1.021734356880188, | |
| "eval_runtime": 98.832, | |
| "eval_samples_per_second": 54.618, | |
| "eval_steps_per_second": 6.83, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.026211092908963257, | |
| "grad_norm": 0.42730259895324707, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 1.0945, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.02714720336999766, | |
| "grad_norm": 0.3776666820049286, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.9927, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.02808331383103206, | |
| "grad_norm": 0.3661386966705322, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.976, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.02808331383103206, | |
| "eval_loss": 1.0193862915039062, | |
| "eval_runtime": 98.9139, | |
| "eval_samples_per_second": 54.573, | |
| "eval_steps_per_second": 6.824, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.029019424292066463, | |
| "grad_norm": 0.4045335054397583, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.987, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.029955534753100865, | |
| "grad_norm": 0.3902008831501007, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 1.0379, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.030891645214135267, | |
| "grad_norm": 0.37765341997146606, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 1.033, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.030891645214135267, | |
| "eval_loss": 1.0190349817276, | |
| "eval_runtime": 98.9222, | |
| "eval_samples_per_second": 54.568, | |
| "eval_steps_per_second": 6.824, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.05873754406912e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |