| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.571428571428571, | |
| "eval_steps": 20, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "eval_loss": 0.269596129655838, | |
| "eval_runtime": 41.7008, | |
| "eval_samples_per_second": 3.285, | |
| "eval_steps_per_second": 0.432, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3253, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 0.2160598337650299, | |
| "eval_runtime": 41.6766, | |
| "eval_samples_per_second": 3.287, | |
| "eval_steps_per_second": 0.432, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.25, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_loss": 0.2049967646598816, | |
| "eval_runtime": 41.643, | |
| "eval_samples_per_second": 3.29, | |
| "eval_steps_per_second": 0.432, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2212, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 0.20088960230350494, | |
| "eval_runtime": 41.6399, | |
| "eval_samples_per_second": 3.29, | |
| "eval_steps_per_second": 0.432, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2168, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.19830842316150665, | |
| "eval_runtime": 41.6606, | |
| "eval_samples_per_second": 3.288, | |
| "eval_steps_per_second": 0.432, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": 0.1965203881263733, | |
| "eval_runtime": 41.6343, | |
| "eval_samples_per_second": 3.291, | |
| "eval_steps_per_second": 0.432, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.211, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.19506876170635223, | |
| "eval_runtime": 41.6134, | |
| "eval_samples_per_second": 3.292, | |
| "eval_steps_per_second": 0.433, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2198, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 0.19399328529834747, | |
| "eval_runtime": 41.7184, | |
| "eval_samples_per_second": 3.284, | |
| "eval_steps_per_second": 0.431, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2018, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 0.19290637969970703, | |
| "eval_runtime": 41.6764, | |
| "eval_samples_per_second": 3.287, | |
| "eval_steps_per_second": 0.432, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2161, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_loss": 0.19207116961479187, | |
| "eval_runtime": 41.9073, | |
| "eval_samples_per_second": 3.269, | |
| "eval_steps_per_second": 0.43, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "eval_loss": 0.19164356589317322, | |
| "eval_runtime": 41.7357, | |
| "eval_samples_per_second": 3.283, | |
| "eval_steps_per_second": 0.431, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1967, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "eval_loss": 0.19071181118488312, | |
| "eval_runtime": 41.6774, | |
| "eval_samples_per_second": 3.287, | |
| "eval_steps_per_second": 0.432, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2225, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "eval_loss": 0.1900840550661087, | |
| "eval_runtime": 41.6631, | |
| "eval_samples_per_second": 3.288, | |
| "eval_steps_per_second": 0.432, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2038, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.18933609127998352, | |
| "eval_runtime": 41.6745, | |
| "eval_samples_per_second": 3.287, | |
| "eval_steps_per_second": 0.432, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2065, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_loss": 0.18885642290115356, | |
| "eval_runtime": 41.7299, | |
| "eval_samples_per_second": 3.283, | |
| "eval_steps_per_second": 0.431, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_loss": 0.18856477737426758, | |
| "eval_runtime": 41.7641, | |
| "eval_samples_per_second": 3.28, | |
| "eval_steps_per_second": 0.431, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1962, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "eval_loss": 0.1882474571466446, | |
| "eval_runtime": 41.7165, | |
| "eval_samples_per_second": 3.284, | |
| "eval_steps_per_second": 0.431, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2058, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_loss": 0.1876191794872284, | |
| "eval_runtime": 41.6862, | |
| "eval_samples_per_second": 3.286, | |
| "eval_steps_per_second": 0.432, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1935, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "eval_loss": 0.18731944262981415, | |
| "eval_runtime": 41.649, | |
| "eval_samples_per_second": 3.289, | |
| "eval_steps_per_second": 0.432, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2191, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "eval_loss": 0.18684075772762299, | |
| "eval_runtime": 41.636, | |
| "eval_samples_per_second": 3.29, | |
| "eval_steps_per_second": 0.432, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.18620511889457703, | |
| "eval_runtime": 41.6932, | |
| "eval_samples_per_second": 3.286, | |
| "eval_steps_per_second": 0.432, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1848, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_loss": 0.1859755963087082, | |
| "eval_runtime": 41.7348, | |
| "eval_samples_per_second": 3.283, | |
| "eval_steps_per_second": 0.431, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1797, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "eval_loss": 0.1856825351715088, | |
| "eval_runtime": 41.6956, | |
| "eval_samples_per_second": 3.286, | |
| "eval_steps_per_second": 0.432, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2054, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "eval_loss": 0.18539157509803772, | |
| "eval_runtime": 41.683, | |
| "eval_samples_per_second": 3.287, | |
| "eval_steps_per_second": 0.432, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.206, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "eval_loss": 0.18517018854618073, | |
| "eval_runtime": 41.665, | |
| "eval_samples_per_second": 3.288, | |
| "eval_steps_per_second": 0.432, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 25, | |
| "total_flos": 1.7463220022909338e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |