| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 0.013269430728931421, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1829, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 0.014592487970564175, | |
| "learning_rate": 4e-05, | |
| "loss": 0.182, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 0.012501483172816132, | |
| "learning_rate": 6.000000000000001e-05, | |
| "loss": 0.1679, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 0.014675726288326245, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1713, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 0.018115438584048463, | |
| "learning_rate": 7.983897175980957e-05, | |
| "loss": 0.1797, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 0.021165587756301594, | |
| "learning_rate": 7.93571835439452e-05, | |
| "loss": 0.1729, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 0.021651982021657683, | |
| "learning_rate": 7.855851442783414e-05, | |
| "loss": 0.1616, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 0.019996298484323216, | |
| "learning_rate": 7.74493948255895e-05, | |
| "loss": 0.1546, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 0.022086733590685532, | |
| "learning_rate": 7.603875471609677e-05, | |
| "loss": 0.1574, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 0.023460046923696968, | |
| "learning_rate": 7.433795174407465e-05, | |
| "loss": 0.1469, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 0.01966702846768547, | |
| "learning_rate": 7.236067977499791e-05, | |
| "loss": 0.1449, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 0.015864251345271908, | |
| "learning_rate": 7.012285864014445e-05, | |
| "loss": 0.1903, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.01668395990001702, | |
| "learning_rate": 6.76425059594746e-05, | |
| "loss": 0.1316, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 0.013306248852551723, | |
| "learning_rate": 6.493959207434934e-05, | |
| "loss": 0.1209, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 0.014444361115276272, | |
| "learning_rate": 6.20358792580841e-05, | |
| "loss": 0.1247, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 0.012694063625630069, | |
| "learning_rate": 5.895474649891995e-05, | |
| "loss": 0.1239, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 0.011335043878086336, | |
| "learning_rate": 5.572100126615695e-05, | |
| "loss": 0.1174, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 0.011121516477788954, | |
| "learning_rate": 5.23606797749979e-05, | |
| "loss": 0.1087, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 0.012626138240166873, | |
| "learning_rate": 4.890083735825258e-05, | |
| "loss": 0.122, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.010374180218655115, | |
| "learning_rate": 4.5369330632706223e-05, | |
| "loss": 0.1109, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 0.010019764004911724, | |
| "learning_rate": 4.17945932140206e-05, | |
| "loss": 0.1118, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 0.010217187797954923, | |
| "learning_rate": 3.820540678597942e-05, | |
| "loss": 0.1239, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 0.010140518188043878, | |
| "learning_rate": 3.4630669367293797e-05, | |
| "loss": 0.1005, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 0.008967088657592456, | |
| "learning_rate": 3.109916264174743e-05, | |
| "loss": 0.1112, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 0.009714335125854632, | |
| "learning_rate": 2.7639320225002108e-05, | |
| "loss": 0.1113, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.0099737196975216, | |
| "learning_rate": 2.427899873384306e-05, | |
| "loss": 0.0941, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 0.010178619725172516, | |
| "learning_rate": 2.1045253501080058e-05, | |
| "loss": 0.0918, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 0.009143314439920697, | |
| "learning_rate": 1.7964120741915905e-05, | |
| "loss": 0.0982, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 0.010956613331760715, | |
| "learning_rate": 1.5060407925650662e-05, | |
| "loss": 0.0843, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 0.009454005573198707, | |
| "learning_rate": 1.2357494040525416e-05, | |
| "loss": 0.0857, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 0.009235678327303279, | |
| "learning_rate": 9.877141359855567e-06, | |
| "loss": 0.0926, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.010279720845112944, | |
| "learning_rate": 7.639320225002106e-06, | |
| "loss": 0.1007, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 0.01108522661450021, | |
| "learning_rate": 5.662048255925357e-06, | |
| "loss": 0.1029, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 0.00970747378677267, | |
| "learning_rate": 3.961245283903239e-06, | |
| "loss": 0.0842, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 0.00911401689896878, | |
| "learning_rate": 2.550605174410512e-06, | |
| "loss": 0.0889, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 0.008503792418277522, | |
| "learning_rate": 1.4414855721658705e-06, | |
| "loss": 0.1013, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 0.009065765434226318, | |
| "learning_rate": 6.428164560548134e-07, | |
| "loss": 0.0892, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 0.008244343474705498, | |
| "learning_rate": 1.6102824019043728e-07, | |
| "loss": 0.0819, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.008506553510928293, | |
| "learning_rate": 0.0, | |
| "loss": 0.0818, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 39, | |
| "total_flos": 76752027779072.0, | |
| "train_loss": 0.12329554576904346, | |
| "train_runtime": 1209.8305, | |
| "train_samples_per_second": 0.496, | |
| "train_steps_per_second": 0.032 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 39, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 76752027779072.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |