| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0010560289351928243, | |
| "eval_steps": 10, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.2801446759641216e-05, | |
| "grad_norm": 3.0317537784576416, | |
| "learning_rate": 5e-05, | |
| "loss": 6.2462, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.2801446759641216e-05, | |
| "eval_loss": 6.236235618591309, | |
| "eval_runtime": 3669.8181, | |
| "eval_samples_per_second": 4.346, | |
| "eval_steps_per_second": 2.173, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00010560289351928243, | |
| "grad_norm": 2.6571614742279053, | |
| "learning_rate": 0.0001, | |
| "loss": 5.5005, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00015840434027892365, | |
| "grad_norm": 3.0436835289001465, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 5.8817, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00021120578703856486, | |
| "grad_norm": 3.7482376098632812, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 6.1575, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0002640072337982061, | |
| "grad_norm": 3.965822219848633, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 5.212, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0003168086805578473, | |
| "grad_norm": 2.9985687732696533, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 4.4805, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0003696101273174885, | |
| "grad_norm": 3.5063633918762207, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 4.7592, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0004224115740771297, | |
| "grad_norm": 3.624417781829834, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 4.1695, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00047521302083677094, | |
| "grad_norm": 3.540501832962036, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 3.7187, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0005280144675964122, | |
| "grad_norm": 3.432539224624634, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 3.3295, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005280144675964122, | |
| "eval_loss": 3.3748233318328857, | |
| "eval_runtime": 3686.9343, | |
| "eval_samples_per_second": 4.326, | |
| "eval_steps_per_second": 2.163, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005808159143560534, | |
| "grad_norm": 4.083893299102783, | |
| "learning_rate": 5e-05, | |
| "loss": 3.3886, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0006336173611156946, | |
| "grad_norm": 4.254306316375732, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 3.0357, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0006864188078753358, | |
| "grad_norm": 3.4154324531555176, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 2.7945, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.000739220254634977, | |
| "grad_norm": 3.889486074447632, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 2.3507, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0007920217013946182, | |
| "grad_norm": 3.418156862258911, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 2.0169, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0008448231481542594, | |
| "grad_norm": 3.2946226596832275, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 2.4668, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0008976245949139007, | |
| "grad_norm": 4.151547431945801, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 2.4295, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0009504260416735419, | |
| "grad_norm": 3.994763135910034, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 2.4314, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001003227488433183, | |
| "grad_norm": 3.648967742919922, | |
| "learning_rate": 7.596123493895991e-07, | |
| "loss": 2.3779, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0010560289351928243, | |
| "grad_norm": 3.826017379760742, | |
| "learning_rate": 0.0, | |
| "loss": 2.3704, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0010560289351928243, | |
| "eval_loss": 2.4360647201538086, | |
| "eval_runtime": 3688.8851, | |
| "eval_samples_per_second": 4.324, | |
| "eval_steps_per_second": 2.162, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.335368898576384e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |