| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 50, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004, | |
| "eval_loss": 1.9942963123321533, | |
| "eval_runtime": 2.9922, | |
| "eval_samples_per_second": 52.136, | |
| "eval_steps_per_second": 2.674, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 3.944833517074585, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.4744, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.8111104965209961, | |
| "eval_runtime": 2.9551, | |
| "eval_samples_per_second": 52.79, | |
| "eval_steps_per_second": 2.707, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 2.267634868621826, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.8438, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.7407088875770569, | |
| "eval_runtime": 3.0287, | |
| "eval_samples_per_second": 51.507, | |
| "eval_steps_per_second": 2.641, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.4580960273742676, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7954, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.7344938516616821, | |
| "eval_runtime": 3.0361, | |
| "eval_samples_per_second": 51.381, | |
| "eval_steps_per_second": 2.635, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.8795665502548218, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.8062, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.7369781732559204, | |
| "eval_runtime": 2.9655, | |
| "eval_samples_per_second": 52.604, | |
| "eval_steps_per_second": 2.698, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.9113938808441162, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.8094, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.7419852614402771, | |
| "eval_runtime": 2.9722, | |
| "eval_samples_per_second": 52.487, | |
| "eval_steps_per_second": 2.692, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 2.6403286457061768, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7765, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 0.7483660578727722, | |
| "eval_runtime": 2.9618, | |
| "eval_samples_per_second": 52.671, | |
| "eval_steps_per_second": 2.701, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.6574498414993286, | |
| "learning_rate": 1.9983081582712684e-05, | |
| "loss": 0.8277, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.7558473348617554, | |
| "eval_runtime": 2.9758, | |
| "eval_samples_per_second": 52.423, | |
| "eval_steps_per_second": 2.688, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.647865653038025, | |
| "learning_rate": 1.9932383577419432e-05, | |
| "loss": 0.8109, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.7527696490287781, | |
| "eval_runtime": 2.9606, | |
| "eval_samples_per_second": 52.692, | |
| "eval_steps_per_second": 2.702, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 6.825328350067139, | |
| "learning_rate": 1.9848077530122083e-05, | |
| "loss": 0.8396, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_loss": 0.7513787150382996, | |
| "eval_runtime": 2.9613, | |
| "eval_samples_per_second": 52.679, | |
| "eval_steps_per_second": 2.701, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.86250901222229, | |
| "learning_rate": 1.973044870579824e-05, | |
| "loss": 0.814, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7477762699127197, | |
| "eval_runtime": 2.9731, | |
| "eval_samples_per_second": 52.471, | |
| "eval_steps_per_second": 2.691, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 3.209193229675293, | |
| "learning_rate": 1.957989512315489e-05, | |
| "loss": 0.512, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.792364239692688, | |
| "eval_runtime": 2.9678, | |
| "eval_samples_per_second": 52.564, | |
| "eval_steps_per_second": 2.696, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 3.9676766395568848, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.4991, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.7819795608520508, | |
| "eval_runtime": 2.9632, | |
| "eval_samples_per_second": 52.645, | |
| "eval_steps_per_second": 2.7, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 1.8103879690170288, | |
| "learning_rate": 1.9182161068802742e-05, | |
| "loss": 0.5022, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_loss": 0.7795585989952087, | |
| "eval_runtime": 2.9661, | |
| "eval_samples_per_second": 52.594, | |
| "eval_steps_per_second": 2.697, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 1.4793493747711182, | |
| "learning_rate": 1.8936326403234125e-05, | |
| "loss": 0.4964, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_loss": 0.7873778343200684, | |
| "eval_runtime": 2.9886, | |
| "eval_samples_per_second": 52.199, | |
| "eval_steps_per_second": 2.677, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 1.360481858253479, | |
| "learning_rate": 1.866025403784439e-05, | |
| "loss": 0.519, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 0.779276430606842, | |
| "eval_runtime": 3.0307, | |
| "eval_samples_per_second": 51.473, | |
| "eval_steps_per_second": 2.64, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.756487250328064, | |
| "learning_rate": 1.8354878114129368e-05, | |
| "loss": 0.5117, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.7745152115821838, | |
| "eval_runtime": 2.9728, | |
| "eval_samples_per_second": 52.475, | |
| "eval_steps_per_second": 2.691, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 1.5152206420898438, | |
| "learning_rate": 1.802123192755044e-05, | |
| "loss": 0.5411, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 0.771414577960968, | |
| "eval_runtime": 2.9972, | |
| "eval_samples_per_second": 52.048, | |
| "eval_steps_per_second": 2.669, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 2.860755681991577, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.5345, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_loss": 0.7689093351364136, | |
| "eval_runtime": 2.9564, | |
| "eval_samples_per_second": 52.766, | |
| "eval_steps_per_second": 2.706, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 1.5143054723739624, | |
| "learning_rate": 1.7273736415730488e-05, | |
| "loss": 0.534, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_loss": 0.7705763578414917, | |
| "eval_runtime": 2.9633, | |
| "eval_samples_per_second": 52.644, | |
| "eval_steps_per_second": 2.7, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.869483470916748, | |
| "learning_rate": 1.686241637868734e-05, | |
| "loss": 0.5292, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.769786536693573, | |
| "eval_runtime": 2.961, | |
| "eval_samples_per_second": 52.686, | |
| "eval_steps_per_second": 2.702, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 3000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 1.1397181534424269e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |