| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.1422121896162527, | |
| "eval_steps": 25, | |
| "global_step": 42, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07223476297968397, | |
| "grad_norm": 0.008988643996417522, | |
| "learning_rate": 5e-05, | |
| "loss": 11.9328, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07223476297968397, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1775, | |
| "eval_samples_per_second": 281.628, | |
| "eval_steps_per_second": 73.223, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.14446952595936793, | |
| "grad_norm": 0.009210865013301373, | |
| "learning_rate": 0.0001, | |
| "loss": 11.933, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.21670428893905191, | |
| "grad_norm": 0.009502468630671501, | |
| "learning_rate": 9.986128001799077e-05, | |
| "loss": 11.9332, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.28893905191873587, | |
| "grad_norm": 0.00922730565071106, | |
| "learning_rate": 9.94459753267812e-05, | |
| "loss": 11.9325, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.3611738148984199, | |
| "grad_norm": 0.009089686907827854, | |
| "learning_rate": 9.875664641789545e-05, | |
| "loss": 11.9326, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.43340857787810383, | |
| "grad_norm": 0.009873133152723312, | |
| "learning_rate": 9.779754323328192e-05, | |
| "loss": 11.933, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.5056433408577878, | |
| "grad_norm": 0.011123728938400745, | |
| "learning_rate": 9.657457896300791e-05, | |
| "loss": 11.9327, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5778781038374717, | |
| "grad_norm": 0.012063764967024326, | |
| "learning_rate": 9.509529358847655e-05, | |
| "loss": 11.9327, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.6501128668171557, | |
| "grad_norm": 0.011134549975395203, | |
| "learning_rate": 9.336880739593416e-05, | |
| "loss": 11.933, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.7223476297968398, | |
| "grad_norm": 0.01320857834070921, | |
| "learning_rate": 9.140576474687264e-05, | |
| "loss": 11.9324, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7945823927765236, | |
| "grad_norm": 0.012652316130697727, | |
| "learning_rate": 8.921826845200139e-05, | |
| "loss": 11.9326, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.8668171557562077, | |
| "grad_norm": 0.013287683948874474, | |
| "learning_rate": 8.681980515339464e-05, | |
| "loss": 11.9325, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.9390519187358917, | |
| "grad_norm": 0.014345780946314335, | |
| "learning_rate": 8.422516217485826e-05, | |
| "loss": 11.9323, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.0474040632054176, | |
| "grad_norm": 0.02747885324060917, | |
| "learning_rate": 8.14503363531613e-05, | |
| "loss": 20.5712, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.1196388261851016, | |
| "grad_norm": 0.015461836010217667, | |
| "learning_rate": 7.85124354122177e-05, | |
| "loss": 11.7716, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.1918735891647856, | |
| "grad_norm": 0.018558837473392487, | |
| "learning_rate": 7.542957248827961e-05, | |
| "loss": 13.575, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.2641083521444696, | |
| "grad_norm": 0.015563758090138435, | |
| "learning_rate": 7.222075445642904e-05, | |
| "loss": 9.4655, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.3363431151241536, | |
| "grad_norm": 0.01990383490920067, | |
| "learning_rate": 6.890576474687263e-05, | |
| "loss": 12.6392, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.4085778781038374, | |
| "grad_norm": 0.021893231198191643, | |
| "learning_rate": 6.550504137351576e-05, | |
| "loss": 13.4199, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.4808126410835214, | |
| "grad_norm": 0.011745378375053406, | |
| "learning_rate": 6.203955092681039e-05, | |
| "loss": 6.7549, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.5530474040632054, | |
| "grad_norm": 0.026938999071717262, | |
| "learning_rate": 5.8530659307753036e-05, | |
| "loss": 15.6081, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.6252821670428894, | |
| "grad_norm": 0.01988409459590912, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 11.9145, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.6975169300225734, | |
| "grad_norm": 0.009685815311968327, | |
| "learning_rate": 5.1469340692246995e-05, | |
| "loss": 6.0489, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.7697516930022572, | |
| "grad_norm": 0.030963227152824402, | |
| "learning_rate": 4.7960449073189606e-05, | |
| "loss": 17.7904, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.8419864559819414, | |
| "grad_norm": 0.022139111533761024, | |
| "learning_rate": 4.4494958626484276e-05, | |
| "loss": 12.1609, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.8419864559819414, | |
| "eval_loss": NaN, | |
| "eval_runtime": 0.1732, | |
| "eval_samples_per_second": 288.718, | |
| "eval_steps_per_second": 75.067, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.9142212189616252, | |
| "grad_norm": 0.030451903119683266, | |
| "learning_rate": 4.109423525312738e-05, | |
| "loss": 15.3316, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.0225733634311513, | |
| "grad_norm": 0.03348216786980629, | |
| "learning_rate": 3.777924554357096e-05, | |
| "loss": 16.8198, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.094808126410835, | |
| "grad_norm": 0.02313702367246151, | |
| "learning_rate": 3.45704275117204e-05, | |
| "loss": 11.3532, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.1670428893905194, | |
| "grad_norm": 0.02855922095477581, | |
| "learning_rate": 3.1487564587782306e-05, | |
| "loss": 13.9424, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.239277652370203, | |
| "grad_norm": 0.013586718589067459, | |
| "learning_rate": 2.854966364683872e-05, | |
| "loss": 7.0922, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.311512415349887, | |
| "grad_norm": 0.03476588800549507, | |
| "learning_rate": 2.577483782514174e-05, | |
| "loss": 14.6245, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.383747178329571, | |
| "grad_norm": 0.0276983343064785, | |
| "learning_rate": 2.3180194846605367e-05, | |
| "loss": 12.0846, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.455981941309255, | |
| "grad_norm": 0.017914237454533577, | |
| "learning_rate": 2.0781731547998614e-05, | |
| "loss": 7.768, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.528216704288939, | |
| "grad_norm": 0.03817462921142578, | |
| "learning_rate": 1.8594235253127375e-05, | |
| "loss": 16.043, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.600451467268623, | |
| "grad_norm": 0.02697724476456642, | |
| "learning_rate": 1.6631192604065855e-05, | |
| "loss": 11.9938, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.672686230248307, | |
| "grad_norm": 0.01927335001528263, | |
| "learning_rate": 1.490470641152345e-05, | |
| "loss": 9.3809, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.744920993227991, | |
| "grad_norm": 0.03292158618569374, | |
| "learning_rate": 1.3425421036992098e-05, | |
| "loss": 14.5154, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.8171557562076748, | |
| "grad_norm": 0.03068559244275093, | |
| "learning_rate": 1.2202456766718093e-05, | |
| "loss": 12.0212, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.889390519187359, | |
| "grad_norm": 0.03927991911768913, | |
| "learning_rate": 1.1243353582104556e-05, | |
| "loss": 16.2666, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.961625282167043, | |
| "grad_norm": 0.032981183379888535, | |
| "learning_rate": 1.0554024673218807e-05, | |
| "loss": 14.3015, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.069977426636569, | |
| "grad_norm": 0.02991604059934616, | |
| "learning_rate": 1.0138719982009242e-05, | |
| "loss": 12.4303, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 3.1422121896162527, | |
| "grad_norm": 0.026301313191652298, | |
| "learning_rate": 1e-05, | |
| "loss": 11.9028, | |
| "step": 42 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 42, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 877486964736.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |