{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.00617970584600173, "eval_steps": 3, "global_step": 25, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.00024718823384006923, "grad_norm": 4.6786603927612305, "learning_rate": 2e-05, "loss": 5.9159, "step": 1 }, { "epoch": 0.00024718823384006923, "eval_loss": 5.787720680236816, "eval_runtime": 53.5592, "eval_samples_per_second": 31.815, "eval_steps_per_second": 15.908, "step": 1 }, { "epoch": 0.0004943764676801385, "grad_norm": 5.279921054840088, "learning_rate": 4e-05, "loss": 5.6142, "step": 2 }, { "epoch": 0.0007415647015202076, "grad_norm": 4.661167144775391, "learning_rate": 6e-05, "loss": 6.9474, "step": 3 }, { "epoch": 0.0007415647015202076, "eval_loss": 5.783358573913574, "eval_runtime": 54.0155, "eval_samples_per_second": 31.547, "eval_steps_per_second": 15.773, "step": 3 }, { "epoch": 0.000988752935360277, "grad_norm": 5.56026029586792, "learning_rate": 8e-05, "loss": 5.8604, "step": 4 }, { "epoch": 0.001235941169200346, "grad_norm": 4.94076681137085, "learning_rate": 0.0001, "loss": 5.6673, "step": 5 }, { "epoch": 0.0014831294030404152, "grad_norm": 3.526470899581909, "learning_rate": 0.00012, "loss": 5.6999, "step": 6 }, { "epoch": 0.0014831294030404152, "eval_loss": 5.737441062927246, "eval_runtime": 54.1511, "eval_samples_per_second": 31.467, "eval_steps_per_second": 15.734, "step": 6 }, { "epoch": 0.0017303176368804844, "grad_norm": 4.711115837097168, "learning_rate": 0.00014, "loss": 6.6336, "step": 7 }, { "epoch": 0.001977505870720554, "grad_norm": 4.793167591094971, "learning_rate": 0.00016, "loss": 4.6339, "step": 8 }, { "epoch": 0.002224694104560623, "grad_norm": 4.787321090698242, "learning_rate": 0.00018, "loss": 5.9321, "step": 9 }, { "epoch": 0.002224694104560623, "eval_loss": 5.613210201263428, "eval_runtime": 54.1901, "eval_samples_per_second": 31.445, "eval_steps_per_second": 15.722, "step": 9 }, { "epoch": 0.002471882338400692, "grad_norm": 5.030083179473877, "learning_rate": 0.0002, "loss": 5.6398, "step": 10 }, { "epoch": 0.0027190705722407614, "grad_norm": 4.728458404541016, "learning_rate": 0.00019781476007338058, "loss": 5.7792, "step": 11 }, { "epoch": 0.0029662588060808304, "grad_norm": 5.103589057922363, "learning_rate": 0.0001913545457642601, "loss": 5.8297, "step": 12 }, { "epoch": 0.0029662588060808304, "eval_loss": 5.452106952667236, "eval_runtime": 54.1757, "eval_samples_per_second": 31.453, "eval_steps_per_second": 15.727, "step": 12 }, { "epoch": 0.0032134470399209, "grad_norm": 4.187473297119141, "learning_rate": 0.00018090169943749476, "loss": 5.2971, "step": 13 }, { "epoch": 0.003460635273760969, "grad_norm": 5.1934123039245605, "learning_rate": 0.00016691306063588583, "loss": 6.0225, "step": 14 }, { "epoch": 0.0037078235076010383, "grad_norm": 4.235335826873779, "learning_rate": 0.00015000000000000001, "loss": 5.04, "step": 15 }, { "epoch": 0.0037078235076010383, "eval_loss": 5.329958915710449, "eval_runtime": 54.1878, "eval_samples_per_second": 31.446, "eval_steps_per_second": 15.723, "step": 15 }, { "epoch": 0.003955011741441108, "grad_norm": 3.6601972579956055, "learning_rate": 0.00013090169943749476, "loss": 5.248, "step": 16 }, { "epoch": 0.004202199975281176, "grad_norm": 4.923957347869873, "learning_rate": 0.00011045284632676536, "loss": 4.7325, "step": 17 }, { "epoch": 0.004449388209121246, "grad_norm": 4.163637638092041, "learning_rate": 8.954715367323468e-05, "loss": 5.3355, "step": 18 }, { "epoch": 0.004449388209121246, "eval_loss": 5.244450092315674, "eval_runtime": 54.2046, "eval_samples_per_second": 31.436, "eval_steps_per_second": 15.718, "step": 18 }, { "epoch": 0.004696576442961315, "grad_norm": 4.613724231719971, "learning_rate": 6.909830056250527e-05, "loss": 5.3436, "step": 19 }, { "epoch": 0.004943764676801384, "grad_norm": 4.689840793609619, "learning_rate": 5.000000000000002e-05, "loss": 6.0068, "step": 20 }, { "epoch": 0.005190952910641453, "grad_norm": 4.097196578979492, "learning_rate": 3.308693936411421e-05, "loss": 4.8949, "step": 21 }, { "epoch": 0.005190952910641453, "eval_loss": 5.19731330871582, "eval_runtime": 54.1886, "eval_samples_per_second": 31.446, "eval_steps_per_second": 15.723, "step": 21 }, { "epoch": 0.005438141144481523, "grad_norm": 3.430635929107666, "learning_rate": 1.9098300562505266e-05, "loss": 4.8494, "step": 22 }, { "epoch": 0.005685329378321592, "grad_norm": 4.0359673500061035, "learning_rate": 8.645454235739903e-06, "loss": 5.0233, "step": 23 }, { "epoch": 0.005932517612161661, "grad_norm": 4.194599151611328, "learning_rate": 2.1852399266194314e-06, "loss": 5.1083, "step": 24 }, { "epoch": 0.005932517612161661, "eval_loss": 5.18394660949707, "eval_runtime": 54.1777, "eval_samples_per_second": 31.452, "eval_steps_per_second": 15.726, "step": 24 }, { "epoch": 0.00617970584600173, "grad_norm": 3.970150947570801, "learning_rate": 0.0, "loss": 5.5501, "step": 25 } ], "logging_steps": 1, "max_steps": 25, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 30, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 6555854438400.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }