| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 50, | |
| "global_step": 625, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 4.338859558105469, | |
| "learning_rate": 2.9528747416929465e-06, | |
| "loss": 0.7159, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.6576318144798279, | |
| "eval_runtime": 12.5345, | |
| "eval_samples_per_second": 159.56, | |
| "eval_steps_per_second": 9.972, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 4.263615131378174, | |
| "learning_rate": 2.814460020065795e-06, | |
| "loss": 0.6551, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.6588318943977356, | |
| "eval_runtime": 12.1193, | |
| "eval_samples_per_second": 165.026, | |
| "eval_steps_per_second": 10.314, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 4.684665679931641, | |
| "learning_rate": 2.5934529411321173e-06, | |
| "loss": 0.6519, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_loss": 0.6581148505210876, | |
| "eval_runtime": 12.4709, | |
| "eval_samples_per_second": 160.373, | |
| "eval_steps_per_second": 10.023, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 4.372674942016602, | |
| "learning_rate": 2.303740192468495e-06, | |
| "loss": 0.6278, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 0.656774640083313, | |
| "eval_runtime": 12.0295, | |
| "eval_samples_per_second": 166.258, | |
| "eval_steps_per_second": 10.391, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 3.940370798110962, | |
| "learning_rate": 1.963525491562421e-06, | |
| "loss": 0.6394, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.6532722115516663, | |
| "eval_runtime": 12.0522, | |
| "eval_samples_per_second": 165.945, | |
| "eval_steps_per_second": 10.372, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 4.178117752075195, | |
| "learning_rate": 1.5941857792939703e-06, | |
| "loss": 0.6528, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.6502550840377808, | |
| "eval_runtime": 12.0395, | |
| "eval_samples_per_second": 166.12, | |
| "eval_steps_per_second": 10.383, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 3.7875773906707764, | |
| "learning_rate": 1.2189280281214128e-06, | |
| "loss": 0.6382, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_loss": 0.6453887820243835, | |
| "eval_runtime": 12.102, | |
| "eval_samples_per_second": 165.261, | |
| "eval_steps_per_second": 10.329, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 3.9947192668914795, | |
| "learning_rate": 8.613310626523911e-07, | |
| "loss": 0.638, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.6425909399986267, | |
| "eval_runtime": 12.0649, | |
| "eval_samples_per_second": 165.77, | |
| "eval_steps_per_second": 10.361, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 3.9734668731689453, | |
| "learning_rate": 5.438640153769653e-07, | |
| "loss": 0.618, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_loss": 0.6400230526924133, | |
| "eval_runtime": 12.0503, | |
| "eval_samples_per_second": 165.97, | |
| "eval_steps_per_second": 10.373, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.733959436416626, | |
| "learning_rate": 2.86474508437579e-07, | |
| "loss": 0.6378, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.6379128694534302, | |
| "eval_runtime": 12.1078, | |
| "eval_samples_per_second": 165.183, | |
| "eval_steps_per_second": 10.324, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 3.8779754638671875, | |
| "learning_rate": 1.0533527116762298e-07, | |
| "loss": 0.6338, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 0.636811375617981, | |
| "eval_runtime": 13.2199, | |
| "eval_samples_per_second": 151.287, | |
| "eval_steps_per_second": 9.455, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.924581527709961, | |
| "learning_rate": 1.1827948028283353e-08, | |
| "loss": 0.6284, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.6364374160766602, | |
| "eval_runtime": 14.3025, | |
| "eval_samples_per_second": 139.836, | |
| "eval_steps_per_second": 8.74, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 625, | |
| "total_flos": 9.391098276138189e+16, | |
| "train_loss": 0.6481949188232422, | |
| "train_runtime": 2478.7348, | |
| "train_samples_per_second": 4.034, | |
| "train_steps_per_second": 0.252 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 625, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.391098276138189e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |