| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.027496133356246778, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001374806667812339, | |
| "grad_norm": 0.42937517166137695, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7403, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001374806667812339, | |
| "eval_loss": 1.7371172904968262, | |
| "eval_runtime": 76.9114, | |
| "eval_samples_per_second": 15.94, | |
| "eval_steps_per_second": 7.97, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002749613335624678, | |
| "grad_norm": 0.45928120613098145, | |
| "learning_rate": 2e-05, | |
| "loss": 1.702, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0041244200034370165, | |
| "grad_norm": 0.3917050063610077, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6859, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005499226671249356, | |
| "grad_norm": 0.4303048849105835, | |
| "learning_rate": 4e-05, | |
| "loss": 1.7193, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.006874033339061694, | |
| "grad_norm": 0.4062155485153198, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6722, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006874033339061694, | |
| "eval_loss": 1.672951102256775, | |
| "eval_runtime": 28.0469, | |
| "eval_samples_per_second": 43.712, | |
| "eval_steps_per_second": 21.856, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008248840006874033, | |
| "grad_norm": 0.4282960295677185, | |
| "learning_rate": 6e-05, | |
| "loss": 1.6042, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.009623646674686372, | |
| "grad_norm": 0.4546167552471161, | |
| "learning_rate": 7e-05, | |
| "loss": 1.6189, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.010998453342498712, | |
| "grad_norm": 0.48203930258750916, | |
| "learning_rate": 8e-05, | |
| "loss": 1.6342, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.01237326001031105, | |
| "grad_norm": 0.5538018941879272, | |
| "learning_rate": 9e-05, | |
| "loss": 1.6007, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013748066678123389, | |
| "grad_norm": 0.511499285697937, | |
| "learning_rate": 0.0001, | |
| "loss": 1.5262, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.013748066678123389, | |
| "eval_loss": 1.472701072692871, | |
| "eval_runtime": 58.4822, | |
| "eval_samples_per_second": 20.964, | |
| "eval_steps_per_second": 10.482, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.015122873345935728, | |
| "grad_norm": 0.5674175024032593, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.4697, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.016497680013748066, | |
| "grad_norm": 0.5313400030136108, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.3832, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.017872486681560405, | |
| "grad_norm": 0.5378307700157166, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.3486, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.019247293349372745, | |
| "grad_norm": 0.48813918232917786, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.2926, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.020622100017185084, | |
| "grad_norm": 0.4850353002548218, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2852, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.020622100017185084, | |
| "eval_loss": 1.2362544536590576, | |
| "eval_runtime": 59.0914, | |
| "eval_samples_per_second": 20.748, | |
| "eval_steps_per_second": 10.374, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.021996906684997423, | |
| "grad_norm": 0.46873438358306885, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.2155, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02337171335280976, | |
| "grad_norm": 0.42953404784202576, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.1754, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0247465200206221, | |
| "grad_norm": 0.41407155990600586, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.2063, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.026121326688434438, | |
| "grad_norm": 0.42715543508529663, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.1701, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.027496133356246778, | |
| "grad_norm": 0.4192153811454773, | |
| "learning_rate": 0.0, | |
| "loss": 1.1276, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.027496133356246778, | |
| "eval_loss": 1.1842869520187378, | |
| "eval_runtime": 61.791, | |
| "eval_samples_per_second": 19.841, | |
| "eval_steps_per_second": 9.921, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.302986508730368e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |