| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 3000, |
| "global_step": 44343, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 6.76544212164265e-05, |
| "grad_norm": 5.0, |
| "learning_rate": 0.0002999932345578783, |
| "loss": 3.5938, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.20296326364927947, |
| "grad_norm": 0.99609375, |
| "learning_rate": 0.000279703673635072, |
| "loss": 2.6416, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.20296326364927947, |
| "eval_loss": 2.220994234085083, |
| "eval_runtime": 59.2323, |
| "eval_samples_per_second": 1593.809, |
| "eval_steps_per_second": 6.23, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.40592652729855894, |
| "grad_norm": 0.85546875, |
| "learning_rate": 0.0002594073472701441, |
| "loss": 2.5493, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.40592652729855894, |
| "eval_loss": 2.1989541053771973, |
| "eval_runtime": 59.05, |
| "eval_samples_per_second": 1598.731, |
| "eval_steps_per_second": 6.249, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.6088897909478385, |
| "grad_norm": 0.85546875, |
| "learning_rate": 0.0002391110209052161, |
| "loss": 2.5262, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.6088897909478385, |
| "eval_loss": 2.185298204421997, |
| "eval_runtime": 59.1009, |
| "eval_samples_per_second": 1597.354, |
| "eval_steps_per_second": 6.244, |
| "step": 9000 |
| }, |
| { |
| "epoch": 0.8118530545971179, |
| "grad_norm": 1.0546875, |
| "learning_rate": 0.0002188146945402882, |
| "loss": 2.5175, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.8118530545971179, |
| "eval_loss": 2.1791369915008545, |
| "eval_runtime": 59.0761, |
| "eval_samples_per_second": 1598.025, |
| "eval_steps_per_second": 6.246, |
| "step": 12000 |
| }, |
| { |
| "epoch": 1.0148163182463974, |
| "grad_norm": 0.79296875, |
| "learning_rate": 0.00019851836817536025, |
| "loss": 2.5115, |
| "step": 15000 |
| }, |
| { |
| "epoch": 1.0148163182463974, |
| "eval_loss": 2.175283670425415, |
| "eval_runtime": 59.0895, |
| "eval_samples_per_second": 1597.662, |
| "eval_steps_per_second": 6.245, |
| "step": 15000 |
| }, |
| { |
| "epoch": 1.217779581895677, |
| "grad_norm": 0.79296875, |
| "learning_rate": 0.0001782220418104323, |
| "loss": 2.5073, |
| "step": 18000 |
| }, |
| { |
| "epoch": 1.217779581895677, |
| "eval_loss": 2.1722984313964844, |
| "eval_runtime": 59.2348, |
| "eval_samples_per_second": 1593.743, |
| "eval_steps_per_second": 6.229, |
| "step": 18000 |
| }, |
| { |
| "epoch": 1.4207428455449564, |
| "grad_norm": 0.921875, |
| "learning_rate": 0.00015792571544550436, |
| "loss": 2.505, |
| "step": 21000 |
| }, |
| { |
| "epoch": 1.4207428455449564, |
| "eval_loss": 2.1715786457061768, |
| "eval_runtime": 59.0551, |
| "eval_samples_per_second": 1598.592, |
| "eval_steps_per_second": 6.248, |
| "step": 21000 |
| }, |
| { |
| "epoch": 1.6237061091942357, |
| "grad_norm": 0.85546875, |
| "learning_rate": 0.0001376293890805764, |
| "loss": 2.5018, |
| "step": 24000 |
| }, |
| { |
| "epoch": 1.6237061091942357, |
| "eval_loss": 2.173018217086792, |
| "eval_runtime": 59.0498, |
| "eval_samples_per_second": 1598.734, |
| "eval_steps_per_second": 6.249, |
| "step": 24000 |
| }, |
| { |
| "epoch": 1.8266693728435153, |
| "grad_norm": 0.953125, |
| "learning_rate": 0.00011733306271564845, |
| "loss": 2.5013, |
| "step": 27000 |
| }, |
| { |
| "epoch": 1.8266693728435153, |
| "eval_loss": 2.169842481613159, |
| "eval_runtime": 59.0564, |
| "eval_samples_per_second": 1598.558, |
| "eval_steps_per_second": 6.248, |
| "step": 27000 |
| }, |
| { |
| "epoch": 2.029632636492795, |
| "grad_norm": 0.9609375, |
| "learning_rate": 9.703673635072052e-05, |
| "loss": 2.5016, |
| "step": 30000 |
| }, |
| { |
| "epoch": 2.029632636492795, |
| "eval_loss": 2.1703505516052246, |
| "eval_runtime": 59.0576, |
| "eval_samples_per_second": 1598.523, |
| "eval_steps_per_second": 6.248, |
| "step": 30000 |
| }, |
| { |
| "epoch": 2.232595900142074, |
| "grad_norm": 0.796875, |
| "learning_rate": 7.674040998579256e-05, |
| "loss": 2.4996, |
| "step": 33000 |
| }, |
| { |
| "epoch": 2.232595900142074, |
| "eval_loss": 2.1704564094543457, |
| "eval_runtime": 59.0291, |
| "eval_samples_per_second": 1599.296, |
| "eval_steps_per_second": 6.251, |
| "step": 33000 |
| }, |
| { |
| "epoch": 2.435559163791354, |
| "grad_norm": 0.8984375, |
| "learning_rate": 5.644408362086462e-05, |
| "loss": 2.5016, |
| "step": 36000 |
| }, |
| { |
| "epoch": 2.435559163791354, |
| "eval_loss": 2.1700754165649414, |
| "eval_runtime": 59.045, |
| "eval_samples_per_second": 1598.866, |
| "eval_steps_per_second": 6.249, |
| "step": 36000 |
| }, |
| { |
| "epoch": 2.638522427440633, |
| "grad_norm": 0.80078125, |
| "learning_rate": 3.614775725593667e-05, |
| "loss": 2.5006, |
| "step": 39000 |
| }, |
| { |
| "epoch": 2.638522427440633, |
| "eval_loss": 2.169651985168457, |
| "eval_runtime": 58.9122, |
| "eval_samples_per_second": 1602.47, |
| "eval_steps_per_second": 6.264, |
| "step": 39000 |
| }, |
| { |
| "epoch": 2.841485691089913, |
| "grad_norm": 0.83203125, |
| "learning_rate": 1.5851430891008727e-05, |
| "loss": 2.5013, |
| "step": 42000 |
| }, |
| { |
| "epoch": 2.841485691089913, |
| "eval_loss": 2.169821262359619, |
| "eval_runtime": 59.0486, |
| "eval_samples_per_second": 1598.768, |
| "eval_steps_per_second": 6.249, |
| "step": 42000 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 44343, |
| "total_flos": 4.3758285331164365e+17, |
| "train_loss": 2.5180485786933677, |
| "train_runtime": 21885.4497, |
| "train_samples_per_second": 518.676, |
| "train_steps_per_second": 2.026 |
| } |
| ], |
| "logging_steps": 3000, |
| "max_steps": 44343, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 3000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.3758285331164365e+17, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
