| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.017190991920233798, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00034381983840467596, | |
| "eval_loss": 5.879480838775635, | |
| "eval_runtime": 170.3546, | |
| "eval_samples_per_second": 28.758, | |
| "eval_steps_per_second": 3.598, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0010314595152140279, | |
| "grad_norm": 41.21235275268555, | |
| "learning_rate": 3e-05, | |
| "loss": 5.2239, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0017190991920233798, | |
| "eval_loss": 1.8665575981140137, | |
| "eval_runtime": 171.959, | |
| "eval_samples_per_second": 28.489, | |
| "eval_steps_per_second": 3.565, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0020629190304280558, | |
| "grad_norm": 10.785271644592285, | |
| "learning_rate": 6e-05, | |
| "loss": 2.5669, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0030943785456420837, | |
| "grad_norm": 4.260767459869385, | |
| "learning_rate": 9e-05, | |
| "loss": 1.0414, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0034381983840467596, | |
| "eval_loss": 0.7525524497032166, | |
| "eval_runtime": 171.9965, | |
| "eval_samples_per_second": 28.483, | |
| "eval_steps_per_second": 3.564, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0041258380608561115, | |
| "grad_norm": 4.613126754760742, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.7269, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.005157297576070139, | |
| "grad_norm": 2.771132469177246, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.688, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005157297576070139, | |
| "eval_loss": 0.6774925589561462, | |
| "eval_runtime": 171.919, | |
| "eval_samples_per_second": 28.496, | |
| "eval_steps_per_second": 3.566, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.006188757091284167, | |
| "grad_norm": 2.5343384742736816, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.6198, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.006876396768093519, | |
| "eval_loss": 0.6454147100448608, | |
| "eval_runtime": 171.8957, | |
| "eval_samples_per_second": 28.5, | |
| "eval_steps_per_second": 3.566, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.007220216606498195, | |
| "grad_norm": 2.4850144386291504, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.7168, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.008251676121712223, | |
| "grad_norm": 2.3500659465789795, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.6841, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.008595495960116899, | |
| "eval_loss": 0.6237576007843018, | |
| "eval_runtime": 171.9416, | |
| "eval_samples_per_second": 28.492, | |
| "eval_steps_per_second": 3.565, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.009283135636926251, | |
| "grad_norm": 2.28306245803833, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.6412, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.010314595152140279, | |
| "grad_norm": 2.4008116722106934, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6312, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.010314595152140279, | |
| "eval_loss": 0.6085503101348877, | |
| "eval_runtime": 171.9315, | |
| "eval_samples_per_second": 28.494, | |
| "eval_steps_per_second": 3.565, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.011346054667354307, | |
| "grad_norm": 2.578450918197632, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.615, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.012033694344163659, | |
| "eval_loss": 0.5967410802841187, | |
| "eval_runtime": 171.9827, | |
| "eval_samples_per_second": 28.485, | |
| "eval_steps_per_second": 3.564, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.012377514182568335, | |
| "grad_norm": 2.038388729095459, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.5948, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.013408973697782363, | |
| "grad_norm": 2.125666379928589, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.6002, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.013752793536187038, | |
| "eval_loss": 0.5893370509147644, | |
| "eval_runtime": 171.9189, | |
| "eval_samples_per_second": 28.496, | |
| "eval_steps_per_second": 3.566, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01444043321299639, | |
| "grad_norm": 2.277637004852295, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.6284, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.015471892728210418, | |
| "grad_norm": 2.8201003074645996, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.5752, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.015471892728210418, | |
| "eval_loss": 0.5873016119003296, | |
| "eval_runtime": 171.9639, | |
| "eval_samples_per_second": 28.489, | |
| "eval_steps_per_second": 3.565, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.016503352243424446, | |
| "grad_norm": 2.28126859664917, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.5964, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.017190991920233798, | |
| "eval_loss": 0.5866773128509521, | |
| "eval_runtime": 171.925, | |
| "eval_samples_per_second": 28.495, | |
| "eval_steps_per_second": 3.566, | |
| "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": 2.02538779410432e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |