| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.07543611504007544, | |
| "eval_steps": 5, | |
| "global_step": 40, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001885902876001886, | |
| "eval_loss": 1.5560035705566406, | |
| "eval_runtime": 68.851, | |
| "eval_samples_per_second": 12.97, | |
| "eval_steps_per_second": 1.627, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005657708628005658, | |
| "grad_norm": 0.48519009351730347, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5268, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00942951438000943, | |
| "eval_loss": 1.493416428565979, | |
| "eval_runtime": 68.726, | |
| "eval_samples_per_second": 12.994, | |
| "eval_steps_per_second": 1.63, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.011315417256011316, | |
| "grad_norm": 0.44942334294319153, | |
| "learning_rate": 6e-05, | |
| "loss": 1.6204, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.016973125884016973, | |
| "grad_norm": 0.3278067708015442, | |
| "learning_rate": 9e-05, | |
| "loss": 1.53, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01885902876001886, | |
| "eval_loss": 1.3736556768417358, | |
| "eval_runtime": 68.8401, | |
| "eval_samples_per_second": 12.972, | |
| "eval_steps_per_second": 1.627, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02263083451202263, | |
| "grad_norm": 0.3598170578479767, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.4709, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.028288543140028287, | |
| "grad_norm": 0.2836269438266754, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.2911, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.028288543140028287, | |
| "eval_loss": 1.324967384338379, | |
| "eval_runtime": 68.8059, | |
| "eval_samples_per_second": 12.979, | |
| "eval_steps_per_second": 1.628, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.033946251768033946, | |
| "grad_norm": 0.29951444268226624, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.2599, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.03771805752003772, | |
| "eval_loss": 1.3078007698059082, | |
| "eval_runtime": 68.8442, | |
| "eval_samples_per_second": 12.971, | |
| "eval_steps_per_second": 1.627, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.039603960396039604, | |
| "grad_norm": 0.2421295940876007, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 1.2654, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.04526166902404526, | |
| "grad_norm": 0.25890016555786133, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.2665, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.04714757190004715, | |
| "eval_loss": 1.2976175546646118, | |
| "eval_runtime": 68.8074, | |
| "eval_samples_per_second": 12.978, | |
| "eval_steps_per_second": 1.628, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05091937765205092, | |
| "grad_norm": 0.24166753888130188, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 1.248, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.056577086280056574, | |
| "grad_norm": 0.2408292293548584, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3139, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.056577086280056574, | |
| "eval_loss": 1.2924392223358154, | |
| "eval_runtime": 68.7889, | |
| "eval_samples_per_second": 12.982, | |
| "eval_steps_per_second": 1.628, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06223479490806223, | |
| "grad_norm": 0.24388636648654938, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 1.1885, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.066006600660066, | |
| "eval_loss": 1.288648009300232, | |
| "eval_runtime": 68.7447, | |
| "eval_samples_per_second": 12.99, | |
| "eval_steps_per_second": 1.629, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06789250353606789, | |
| "grad_norm": 0.2198489010334015, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.3563, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.07355021216407355, | |
| "grad_norm": 0.23230212926864624, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.2303, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.07543611504007544, | |
| "eval_loss": 1.2861415147781372, | |
| "eval_runtime": 68.8353, | |
| "eval_samples_per_second": 12.973, | |
| "eval_steps_per_second": 1.627, | |
| "step": 40 | |
| } | |
| ], | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.86529082703872e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |