| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06571741511500548, | |
| "eval_steps": 5, | |
| "global_step": 15, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004381161007667032, | |
| "grad_norm": 0.28300485014915466, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1534, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004381161007667032, | |
| "eval_loss": 0.178242489695549, | |
| "eval_runtime": 10.3376, | |
| "eval_samples_per_second": 9.383, | |
| "eval_steps_per_second": 4.74, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.008762322015334063, | |
| "grad_norm": 0.33447790145874023, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1934, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.013143483023001095, | |
| "grad_norm": 0.27208104729652405, | |
| "learning_rate": 6e-05, | |
| "loss": 0.1603, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.017524644030668127, | |
| "grad_norm": 0.34873825311660767, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1852, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.02190580503833516, | |
| "grad_norm": 0.4304073750972748, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1692, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02190580503833516, | |
| "eval_loss": 0.12932729721069336, | |
| "eval_runtime": 10.3651, | |
| "eval_samples_per_second": 9.358, | |
| "eval_steps_per_second": 4.727, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02628696604600219, | |
| "grad_norm": 0.40466296672821045, | |
| "learning_rate": 0.00012, | |
| "loss": 0.114, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03066812705366922, | |
| "grad_norm": 0.3059878349304199, | |
| "learning_rate": 0.00014, | |
| "loss": 0.1037, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.03504928806133625, | |
| "grad_norm": 0.20368129014968872, | |
| "learning_rate": 0.00016, | |
| "loss": 0.055, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03943044906900329, | |
| "grad_norm": 0.31183746457099915, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0435, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04381161007667032, | |
| "grad_norm": 0.2398935854434967, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0406, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04381161007667032, | |
| "eval_loss": 0.029431650415062904, | |
| "eval_runtime": 10.3988, | |
| "eval_samples_per_second": 9.328, | |
| "eval_steps_per_second": 4.712, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04819277108433735, | |
| "grad_norm": 0.28297099471092224, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.0256, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.05257393209200438, | |
| "grad_norm": 0.13293632864952087, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.0161, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.056955093099671415, | |
| "grad_norm": 0.238490030169487, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.0233, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.06133625410733844, | |
| "grad_norm": 0.1766541600227356, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0222, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.06571741511500548, | |
| "grad_norm": 0.08793684095144272, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.0064, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06571741511500548, | |
| "eval_loss": 0.018376410007476807, | |
| "eval_runtime": 10.3128, | |
| "eval_samples_per_second": 9.406, | |
| "eval_steps_per_second": 4.751, | |
| "step": 15 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.109738548887552e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |