| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.952830188679245, | |
| "eval_steps": 10, | |
| "global_step": 210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 4.999720254525684e-05, | |
| "loss": 1.458, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.9930094929529506e-05, | |
| "loss": 1.4236, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.972077065562821e-05, | |
| "loss": 1.3366, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 1.278298020362854, | |
| "eval_runtime": 24.9289, | |
| "eval_samples_per_second": 5.175, | |
| "eval_steps_per_second": 1.725, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.937319780454559e-05, | |
| "loss": 1.2828, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.888932014465352e-05, | |
| "loss": 1.2563, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 1.2320669889450073, | |
| "eval_runtime": 24.9408, | |
| "eval_samples_per_second": 5.172, | |
| "eval_steps_per_second": 1.724, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.827184371610511e-05, | |
| "loss": 1.2308, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.752422169756048e-05, | |
| "loss": 1.2289, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 1.2012428045272827, | |
| "eval_runtime": 24.9502, | |
| "eval_samples_per_second": 5.17, | |
| "eval_steps_per_second": 1.723, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.665063509461097e-05, | |
| "loss": 1.2051, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.5655969357899874e-05, | |
| "loss": 1.1837, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 1.1688246726989746, | |
| "eval_runtime": 24.9605, | |
| "eval_samples_per_second": 5.168, | |
| "eval_steps_per_second": 1.723, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 4.454578706170075e-05, | |
| "loss": 1.1831, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.332629679574566e-05, | |
| "loss": 1.1534, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 1.1306308507919312, | |
| "eval_runtime": 24.9566, | |
| "eval_samples_per_second": 5.169, | |
| "eval_steps_per_second": 1.723, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 4.2004318444272985e-05, | |
| "loss": 1.1413, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 4.058724504646834e-05, | |
| "loss": 1.1254, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 1.1036521196365356, | |
| "eval_runtime": 24.9587, | |
| "eval_samples_per_second": 5.169, | |
| "eval_steps_per_second": 1.723, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 3.908300145159055e-05, | |
| "loss": 1.1167, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.1011, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "eval_loss": 1.0882474184036255, | |
| "eval_runtime": 24.8955, | |
| "eval_samples_per_second": 5.182, | |
| "eval_steps_per_second": 1.727, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3.5847093477938956e-05, | |
| "loss": 1.0891, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 3.413352560915988e-05, | |
| "loss": 1.0825, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 1.074791669845581, | |
| "eval_runtime": 24.8917, | |
| "eval_samples_per_second": 5.182, | |
| "eval_steps_per_second": 1.727, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.2368879360272606e-05, | |
| "loss": 1.0778, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 3.056302334890786e-05, | |
| "loss": 1.0876, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_loss": 1.063535451889038, | |
| "eval_runtime": 24.8907, | |
| "eval_samples_per_second": 5.183, | |
| "eval_steps_per_second": 1.728, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 2.872605665440436e-05, | |
| "loss": 1.0835, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 2.686825233966061e-05, | |
| "loss": 1.0716, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_loss": 1.0540255308151245, | |
| "eval_runtime": 24.8912, | |
| "eval_samples_per_second": 5.183, | |
| "eval_steps_per_second": 1.728, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.0767, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 2.3131747660339394e-05, | |
| "loss": 1.0517, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "eval_loss": 1.0458558797836304, | |
| "eval_runtime": 24.9528, | |
| "eval_samples_per_second": 5.17, | |
| "eval_steps_per_second": 1.723, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 2.1273943345595637e-05, | |
| "loss": 1.0564, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.9436976651092144e-05, | |
| "loss": 1.0289, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_loss": 1.0388729572296143, | |
| "eval_runtime": 24.9226, | |
| "eval_samples_per_second": 5.176, | |
| "eval_steps_per_second": 1.725, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 1.7631120639727393e-05, | |
| "loss": 1.0318, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 1.5866474390840125e-05, | |
| "loss": 1.0564, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "eval_loss": 1.0331891775131226, | |
| "eval_runtime": 24.9512, | |
| "eval_samples_per_second": 5.17, | |
| "eval_steps_per_second": 1.723, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 1.4152906522061048e-05, | |
| "loss": 1.0499, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 1.2500000000000006e-05, | |
| "loss": 1.034, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "eval_loss": 1.0287576913833618, | |
| "eval_runtime": 24.9512, | |
| "eval_samples_per_second": 5.17, | |
| "eval_steps_per_second": 1.723, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 1.0916998548409449e-05, | |
| "loss": 1.0387, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 9.412754953531663e-06, | |
| "loss": 1.0337, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "eval_loss": 1.0253461599349976, | |
| "eval_runtime": 24.8842, | |
| "eval_samples_per_second": 5.184, | |
| "eval_steps_per_second": 1.728, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 7.99568155572701e-06, | |
| "loss": 1.0375, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 6.673703204254347e-06, | |
| "loss": 1.033, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "eval_loss": 1.0230636596679688, | |
| "eval_runtime": 24.9446, | |
| "eval_samples_per_second": 5.171, | |
| "eval_steps_per_second": 1.724, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 5.454212938299255e-06, | |
| "loss": 1.0292, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 4.344030642100133e-06, | |
| "loss": 1.0312, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "eval_loss": 1.021319031715393, | |
| "eval_runtime": 24.9503, | |
| "eval_samples_per_second": 5.17, | |
| "eval_steps_per_second": 1.723, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 3.3493649053890326e-06, | |
| "loss": 1.0343, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 2.475778302439524e-06, | |
| "loss": 1.0207, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "eval_loss": 1.0203514099121094, | |
| "eval_runtime": 24.9557, | |
| "eval_samples_per_second": 5.169, | |
| "eval_steps_per_second": 1.723, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 1.7281562838948966e-06, | |
| "loss": 1.0413, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 1.1106798553464804e-06, | |
| "loss": 1.0271, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "eval_loss": 1.019814133644104, | |
| "eval_runtime": 24.9781, | |
| "eval_samples_per_second": 5.165, | |
| "eval_steps_per_second": 1.722, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 6.268021954544096e-07, | |
| "loss": 1.0233, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 2.7922934437178695e-07, | |
| "loss": 1.0351, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "eval_loss": 1.0197088718414307, | |
| "eval_runtime": 24.9625, | |
| "eval_samples_per_second": 5.168, | |
| "eval_steps_per_second": 1.723, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 6.990507047049676e-08, | |
| "loss": 1.015, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.0, | |
| "loss": 1.0339, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "eval_loss": 1.0196079015731812, | |
| "eval_runtime": 24.9619, | |
| "eval_samples_per_second": 5.168, | |
| "eval_steps_per_second": 1.723, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "step": 210, | |
| "total_flos": 3.221775454641848e+17, | |
| "train_loss": 1.1020877304531278, | |
| "train_runtime": 16139.1818, | |
| "train_samples_per_second": 1.575, | |
| "train_steps_per_second": 0.013 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 210, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 20, | |
| "total_flos": 3.221775454641848e+17, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |