| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "global_step": 50000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.1964e-05, | |
| "loss": 1.6251, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.3952e-05, | |
| "loss": 1.0682, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.9991929541758342e-05, | |
| "loss": 0.8531, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2.9926636060938516e-05, | |
| "loss": 0.8384, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.9796072198520905e-05, | |
| "loss": 0.8226, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.96010361969696e-05, | |
| "loss": 0.8162, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.9342281162131742e-05, | |
| "loss": 0.802, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.902058372494256e-05, | |
| "loss": 0.8099, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.8637577842487164e-05, | |
| "loss": 0.8511, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.8194938302038455e-05, | |
| "loss": 0.8355, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.8438622275145564, | |
| "eval_loss": 0.6963315010070801, | |
| "eval_runtime": 212.1584, | |
| "eval_samples_per_second": 23.567, | |
| "eval_steps_per_second": 0.589, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.7695129103900878e-05, | |
| "loss": 0.8541, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.7139335532657592e-05, | |
| "loss": 0.8387, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.6530459735832775e-05, | |
| "loss": 0.8294, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.5871164172365078e-05, | |
| "loss": 0.8318, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.5164331774003996e-05, | |
| "loss": 0.8036, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 2.441459810041615e-05, | |
| "loss": 0.8184, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 2.3622237699269646e-05, | |
| "loss": 0.8164, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.2792174446485267e-05, | |
| "loss": 0.8178, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2.192803799746817e-05, | |
| "loss": 0.8153, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.1033607001041156e-05, | |
| "loss": 0.8333, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.8461335925242551, | |
| "eval_loss": 0.6882692575454712, | |
| "eval_runtime": 222.9746, | |
| "eval_samples_per_second": 22.424, | |
| "eval_steps_per_second": 0.561, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 2.011372523790927e-05, | |
| "loss": 0.8297, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.9171527138850477e-05, | |
| "loss": 0.8453, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.8210155373841292e-05, | |
| "loss": 0.8384, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.7234746416166368e-05, | |
| "loss": 0.828, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.624956548101695e-05, | |
| "loss": 0.8226, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1.5259912447470205e-05, | |
| "loss": 0.8188, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.4268134252092541e-05, | |
| "loss": 0.807, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 1.3279556319416353e-05, | |
| "loss": 0.8235, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.2298501449209877e-05, | |
| "loss": 0.802, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.1330221470267496e-05, | |
| "loss": 0.8327, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.8488932773912237, | |
| "eval_loss": 0.6712204217910767, | |
| "eval_runtime": 182.5232, | |
| "eval_samples_per_second": 27.394, | |
| "eval_steps_per_second": 0.685, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.0377012633509545e-05, | |
| "loss": 0.8178, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 9.444018944243678e-06, | |
| "loss": 0.8298, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 8.535320146484432e-06, | |
| "loss": 0.8438, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 7.655754766677537e-06, | |
| "loss": 0.8532, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 6.807408671366549e-06, | |
| "loss": 0.8595, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 5.995680124886882e-06, | |
| "loss": 0.8587, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 5.222494120486821e-06, | |
| "loss": 0.8702, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.492770657745851e-06, | |
| "loss": 0.8598, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 3.808240256977668e-06, | |
| "loss": 0.8558, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3.1726485739618023e-06, | |
| "loss": 0.8659, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.8498886317841439, | |
| "eval_loss": 0.6668724417686462, | |
| "eval_runtime": 209.1326, | |
| "eval_samples_per_second": 23.908, | |
| "eval_steps_per_second": 0.598, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 2.5893320607754046e-06, | |
| "loss": 0.8543, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 2.0596740374389e-06, | |
| "loss": 0.8245, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.587044866629908e-06, | |
| "loss": 0.8451, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.1725653761999294e-06, | |
| "loss": 0.8217, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 8.185497391612495e-07, | |
| "loss": 0.7959, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 5.265459758197716e-07, | |
| "loss": 0.8161, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.980276904853785e-07, | |
| "loss": 0.811, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.3353680835990033e-07, | |
| "loss": 0.8219, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 3.412003109892648e-08, | |
| "loss": 0.8384, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.3123017789107294e-11, | |
| "loss": 0.8297, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8507072999639782, | |
| "eval_loss": 0.6616092920303345, | |
| "eval_runtime": 184.1184, | |
| "eval_samples_per_second": 27.156, | |
| "eval_steps_per_second": 0.679, | |
| "step": 50000 | |
| } | |
| ], | |
| "max_steps": 50000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.578331275264e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |