| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3246753246753247, | |
| "eval_steps": 4, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.012987012987012988, | |
| "grad_norm": 1.7707418203353882, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 3.9226, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.012987012987012988, | |
| "eval_loss": 3.3614439964294434, | |
| "eval_runtime": 2.1416, | |
| "eval_samples_per_second": 15.409, | |
| "eval_steps_per_second": 4.202, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.025974025974025976, | |
| "grad_norm": 1.3376400470733643, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 3.7688, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.03896103896103896, | |
| "grad_norm": 0.8218762874603271, | |
| "learning_rate": 3e-06, | |
| "loss": 2.3745, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.05194805194805195, | |
| "grad_norm": 1.0264558792114258, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 3.6438, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.05194805194805195, | |
| "eval_loss": 3.3627524375915527, | |
| "eval_runtime": 1.1019, | |
| "eval_samples_per_second": 29.948, | |
| "eval_steps_per_second": 8.168, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06493506493506493, | |
| "grad_norm": 1.4011552333831787, | |
| "learning_rate": 5e-06, | |
| "loss": 3.3352, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07792207792207792, | |
| "grad_norm": 0.9959847331047058, | |
| "learning_rate": 6e-06, | |
| "loss": 2.8443, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 1.132798194885254, | |
| "learning_rate": 7e-06, | |
| "loss": 2.8373, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1038961038961039, | |
| "grad_norm": 0.8775717616081238, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 3.1148, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1038961038961039, | |
| "eval_loss": 3.360588788986206, | |
| "eval_runtime": 1.1041, | |
| "eval_samples_per_second": 29.888, | |
| "eval_steps_per_second": 8.151, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.11688311688311688, | |
| "grad_norm": 1.255873441696167, | |
| "learning_rate": 9e-06, | |
| "loss": 3.6851, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.12987012987012986, | |
| "grad_norm": 1.1891772747039795, | |
| "learning_rate": 1e-05, | |
| "loss": 3.358, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 0.936803936958313, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 2.6444, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.15584415584415584, | |
| "grad_norm": 0.9471662044525146, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 2.7137, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.15584415584415584, | |
| "eval_loss": 3.3596856594085693, | |
| "eval_runtime": 1.1064, | |
| "eval_samples_per_second": 29.827, | |
| "eval_steps_per_second": 8.135, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.16883116883116883, | |
| "grad_norm": 1.4470901489257812, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 3.2627, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 1.4950265884399414, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 3.6887, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.19480519480519481, | |
| "grad_norm": 1.1622928380966187, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 3.0483, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2077922077922078, | |
| "grad_norm": 1.2167164087295532, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 3.3572, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.2077922077922078, | |
| "eval_loss": 3.3558952808380127, | |
| "eval_runtime": 1.1045, | |
| "eval_samples_per_second": 29.877, | |
| "eval_steps_per_second": 8.148, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.22077922077922077, | |
| "grad_norm": 0.882570743560791, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 2.9154, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.23376623376623376, | |
| "grad_norm": 1.3990870714187622, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 2.6733, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.24675324675324675, | |
| "grad_norm": 0.8659160137176514, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 2.9225, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2597402597402597, | |
| "grad_norm": 1.0206493139266968, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 2.9704, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2597402597402597, | |
| "eval_loss": 3.353020668029785, | |
| "eval_runtime": 1.1115, | |
| "eval_samples_per_second": 29.689, | |
| "eval_steps_per_second": 8.097, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2727272727272727, | |
| "grad_norm": 1.4770469665527344, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 3.2062, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 1.4567232131958008, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 3.6146, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.2987012987012987, | |
| "grad_norm": 1.4736906290054321, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 3.479, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.3116883116883117, | |
| "grad_norm": 1.1980702877044678, | |
| "learning_rate": 1.0926199633097156e-07, | |
| "loss": 2.5612, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.3116883116883117, | |
| "eval_loss": 3.353377342224121, | |
| "eval_runtime": 1.124, | |
| "eval_samples_per_second": 29.36, | |
| "eval_steps_per_second": 8.007, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.3246753246753247, | |
| "grad_norm": 1.2400470972061157, | |
| "learning_rate": 0.0, | |
| "loss": 3.1659, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1286710965043200.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |