| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05466721333880006, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005466721333880005, | |
| "eval_loss": 0.947159469127655, | |
| "eval_runtime": 422.6377, | |
| "eval_samples_per_second": 7.29, | |
| "eval_steps_per_second": 0.913, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016400164001640015, | |
| "grad_norm": 0.4066868722438812, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.9585, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.003280032800328003, | |
| "grad_norm": 0.5946331024169922, | |
| "learning_rate": 3e-05, | |
| "loss": 0.904, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004920049200492005, | |
| "grad_norm": 0.6447503566741943, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.9863, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004920049200492005, | |
| "eval_loss": 0.9110166430473328, | |
| "eval_runtime": 426.9481, | |
| "eval_samples_per_second": 7.216, | |
| "eval_steps_per_second": 0.904, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006560065600656006, | |
| "grad_norm": 0.4532524645328522, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.9468, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008200082000820008, | |
| "grad_norm": 0.45407116413116455, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.7672, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00984009840098401, | |
| "grad_norm": 0.31059959530830383, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.7584, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00984009840098401, | |
| "eval_loss": 0.7376940250396729, | |
| "eval_runtime": 426.7475, | |
| "eval_samples_per_second": 7.22, | |
| "eval_steps_per_second": 0.905, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.011480114801148012, | |
| "grad_norm": 0.23769880831241608, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.7161, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.013120131201312012, | |
| "grad_norm": 0.323109894990921, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.7092, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.014760147601476014, | |
| "grad_norm": 0.2681789696216583, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.5955, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.014760147601476014, | |
| "eval_loss": 0.6832801699638367, | |
| "eval_runtime": 426.9229, | |
| "eval_samples_per_second": 7.217, | |
| "eval_steps_per_second": 0.904, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.016400164001640016, | |
| "grad_norm": 0.2589839696884155, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.7209, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.018040180401804017, | |
| "grad_norm": 0.18874622881412506, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.6304, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01968019680196802, | |
| "grad_norm": 0.2278064489364624, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.6688, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01968019680196802, | |
| "eval_loss": 0.6574512720108032, | |
| "eval_runtime": 426.6686, | |
| "eval_samples_per_second": 7.221, | |
| "eval_steps_per_second": 0.905, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02132021320213202, | |
| "grad_norm": 0.19672754406929016, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.6362, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.022960229602296024, | |
| "grad_norm": 0.23685137927532196, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.6394, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.024600246002460024, | |
| "grad_norm": 0.2077840119600296, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.6716, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.024600246002460024, | |
| "eval_loss": 0.640738308429718, | |
| "eval_runtime": 426.7745, | |
| "eval_samples_per_second": 7.219, | |
| "eval_steps_per_second": 0.904, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.026240262402624025, | |
| "grad_norm": 0.25208452343940735, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.609, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.02788027880278803, | |
| "grad_norm": 0.22854222357273102, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.696, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.02952029520295203, | |
| "grad_norm": 0.2772148847579956, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.6943, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.02952029520295203, | |
| "eval_loss": 0.631170928478241, | |
| "eval_runtime": 426.7506, | |
| "eval_samples_per_second": 7.22, | |
| "eval_steps_per_second": 0.905, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.031160311603116032, | |
| "grad_norm": 0.23608601093292236, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.5533, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.03280032800328003, | |
| "grad_norm": 0.18795213103294373, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.5873, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03444034440344403, | |
| "grad_norm": 0.16693271696567535, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.6124, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03444034440344403, | |
| "eval_loss": 0.6260421872138977, | |
| "eval_runtime": 426.5842, | |
| "eval_samples_per_second": 7.222, | |
| "eval_steps_per_second": 0.905, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03608036080360803, | |
| "grad_norm": 0.21537937223911285, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.5806, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.03772037720377204, | |
| "grad_norm": 0.21624986827373505, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.5876, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.03936039360393604, | |
| "grad_norm": 0.2545225918292999, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.6063, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.03936039360393604, | |
| "eval_loss": 0.6232536435127258, | |
| "eval_runtime": 426.6611, | |
| "eval_samples_per_second": 7.221, | |
| "eval_steps_per_second": 0.905, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04100041000410004, | |
| "grad_norm": 0.2209480255842209, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.5878, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.04264042640426404, | |
| "grad_norm": 0.21654435992240906, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.6279, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.04428044280442804, | |
| "grad_norm": 0.1880025714635849, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.6035, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.04428044280442804, | |
| "eval_loss": 0.6215988993644714, | |
| "eval_runtime": 426.4354, | |
| "eval_samples_per_second": 7.225, | |
| "eval_steps_per_second": 0.905, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.04592045920459205, | |
| "grad_norm": 0.20597471296787262, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.64, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.04756047560475605, | |
| "grad_norm": 0.1945589780807495, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.5816, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.04920049200492005, | |
| "grad_norm": 0.22032710909843445, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.6275, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.04920049200492005, | |
| "eval_loss": 0.620901346206665, | |
| "eval_runtime": 426.7055, | |
| "eval_samples_per_second": 7.22, | |
| "eval_steps_per_second": 0.905, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.05084050840508405, | |
| "grad_norm": 0.16890203952789307, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.6217, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.05248052480524805, | |
| "grad_norm": 0.21651792526245117, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.6589, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.05412054120541206, | |
| "grad_norm": 0.236527681350708, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.6654, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.05412054120541206, | |
| "eval_loss": 0.6207596063613892, | |
| "eval_runtime": 426.6715, | |
| "eval_samples_per_second": 7.221, | |
| "eval_steps_per_second": 0.905, | |
| "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": 2.6212381734469632e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |