| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.29069767441860467, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005813953488372093, | |
| "eval_loss": 0.3217775821685791, | |
| "eval_runtime": 22.5243, | |
| "eval_samples_per_second": 12.875, | |
| "eval_steps_per_second": 1.643, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01744186046511628, | |
| "grad_norm": 0.7888044714927673, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2985, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.029069767441860465, | |
| "eval_loss": 0.21819913387298584, | |
| "eval_runtime": 22.2887, | |
| "eval_samples_per_second": 13.011, | |
| "eval_steps_per_second": 1.66, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03488372093023256, | |
| "grad_norm": 0.64023357629776, | |
| "learning_rate": 6e-05, | |
| "loss": 0.2397, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05232558139534884, | |
| "grad_norm": 0.6746029853820801, | |
| "learning_rate": 9e-05, | |
| "loss": 0.1323, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05813953488372093, | |
| "eval_loss": 0.06604965776205063, | |
| "eval_runtime": 22.362, | |
| "eval_samples_per_second": 12.968, | |
| "eval_steps_per_second": 1.655, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06976744186046512, | |
| "grad_norm": 0.4298788905143738, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.0767, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0872093023255814, | |
| "grad_norm": 0.7074999213218689, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0782, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0872093023255814, | |
| "eval_loss": 0.056295592337846756, | |
| "eval_runtime": 22.4039, | |
| "eval_samples_per_second": 12.944, | |
| "eval_steps_per_second": 1.651, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10465116279069768, | |
| "grad_norm": 0.45047903060913086, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0643, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.11627906976744186, | |
| "eval_loss": 0.05145295709371567, | |
| "eval_runtime": 22.3971, | |
| "eval_samples_per_second": 12.948, | |
| "eval_steps_per_second": 1.652, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12209302325581395, | |
| "grad_norm": 0.49891409277915955, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.0479, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13953488372093023, | |
| "grad_norm": 0.3716846704483032, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.0426, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.14534883720930233, | |
| "eval_loss": 0.04720764607191086, | |
| "eval_runtime": 22.394, | |
| "eval_samples_per_second": 12.95, | |
| "eval_steps_per_second": 1.652, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1569767441860465, | |
| "grad_norm": 0.487210214138031, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.0556, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1744186046511628, | |
| "grad_norm": 0.3575952351093292, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0631, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1744186046511628, | |
| "eval_loss": 0.044543541967868805, | |
| "eval_runtime": 22.4045, | |
| "eval_samples_per_second": 12.944, | |
| "eval_steps_per_second": 1.651, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19186046511627908, | |
| "grad_norm": 0.2726097106933594, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.0447, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.20348837209302326, | |
| "eval_loss": 0.04344777390360832, | |
| "eval_runtime": 22.4094, | |
| "eval_samples_per_second": 12.941, | |
| "eval_steps_per_second": 1.651, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.20930232558139536, | |
| "grad_norm": 0.3613501787185669, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0494, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.22674418604651161, | |
| "grad_norm": 0.2460230439901352, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.0322, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.23255813953488372, | |
| "eval_loss": 0.04378252848982811, | |
| "eval_runtime": 22.4104, | |
| "eval_samples_per_second": 12.94, | |
| "eval_steps_per_second": 1.651, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2441860465116279, | |
| "grad_norm": 0.20030193030834198, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0484, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2616279069767442, | |
| "grad_norm": 0.31143292784690857, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0406, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2616279069767442, | |
| "eval_loss": 0.04343704134225845, | |
| "eval_runtime": 22.3766, | |
| "eval_samples_per_second": 12.96, | |
| "eval_steps_per_second": 1.654, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.27906976744186046, | |
| "grad_norm": 0.31634950637817383, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.0343, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.29069767441860467, | |
| "eval_loss": 0.04321952164173126, | |
| "eval_runtime": 22.4042, | |
| "eval_samples_per_second": 12.944, | |
| "eval_steps_per_second": 1.651, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.65570641213522e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |