| { |
| "best_global_step": 124, |
| "best_metric": 0.19054362177848816, |
| "best_model_checkpoint": "./llava_qlora_weather_model/checkpoint-124", |
| "epoch": 3.0, |
| "eval_steps": 31, |
| "global_step": 147, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.20565552699228792, |
| "grad_norm": 2.331812620162964, |
| "learning_rate": 0.0001999744599547812, |
| "loss": 1.3843, |
| "mean_token_accuracy": 0.6790310423821211, |
| "num_tokens": 1290816.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.41131105398457585, |
| "grad_norm": 0.9054113030433655, |
| "learning_rate": 0.00019692540862655585, |
| "loss": 0.2958, |
| "mean_token_accuracy": 0.9068963050842285, |
| "num_tokens": 2581570.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.6169665809768637, |
| "grad_norm": 0.7487640380859375, |
| "learning_rate": 0.00018894622845670283, |
| "loss": 0.2476, |
| "mean_token_accuracy": 0.917157232388854, |
| "num_tokens": 3872000.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6375321336760925, |
| "eval_loss": 0.23227548599243164, |
| "eval_mean_token_accuracy": 0.9232641770289495, |
| "eval_num_tokens": 4001060.0, |
| "eval_runtime": 276.5084, |
| "eval_samples_per_second": 0.705, |
| "eval_steps_per_second": 0.705, |
| "step": 31 |
| }, |
| { |
| "epoch": 0.8226221079691517, |
| "grad_norm": 0.6455361843109131, |
| "learning_rate": 0.0001764427819207624, |
| "loss": 0.2208, |
| "mean_token_accuracy": 0.9249804724007845, |
| "num_tokens": 5162524.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.0205655526992288, |
| "grad_norm": 0.5358895659446716, |
| "learning_rate": 0.00016005105914000507, |
| "loss": 0.2077, |
| "mean_token_accuracy": 0.929297363990313, |
| "num_tokens": 6405024.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.2262210796915167, |
| "grad_norm": 0.5378326177597046, |
| "learning_rate": 0.00014060482812625055, |
| "loss": 0.1782, |
| "mean_token_accuracy": 0.9367100398987531, |
| "num_tokens": 7695653.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.2673521850899743, |
| "eval_loss": 0.20991797745227814, |
| "eval_mean_token_accuracy": 0.9294621323927855, |
| "eval_num_tokens": 7953894.0, |
| "eval_runtime": 276.7899, |
| "eval_samples_per_second": 0.705, |
| "eval_steps_per_second": 0.705, |
| "step": 62 |
| }, |
| { |
| "epoch": 1.4318766066838047, |
| "grad_norm": 0.5556976199150085, |
| "learning_rate": 0.00011909322501769406, |
| "loss": 0.1738, |
| "mean_token_accuracy": 0.9374068416655064, |
| "num_tokens": 8986431.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.6375321336760926, |
| "grad_norm": 0.5379733443260193, |
| "learning_rate": 9.661044148624037e-05, |
| "loss": 0.17, |
| "mean_token_accuracy": 0.9386381138116121, |
| "num_tokens": 10277533.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.8431876606683804, |
| "grad_norm": 0.532435417175293, |
| "learning_rate": 7.430006847567972e-05, |
| "loss": 0.1691, |
| "mean_token_accuracy": 0.9383826002478599, |
| "num_tokens": 11568708.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.904884318766067, |
| "eval_loss": 0.19236187636852264, |
| "eval_mean_token_accuracy": 0.934929045347067, |
| "eval_num_tokens": 11955105.0, |
| "eval_runtime": 277.3902, |
| "eval_samples_per_second": 0.703, |
| "eval_steps_per_second": 0.703, |
| "step": 93 |
| }, |
| { |
| "epoch": 2.0411311053984575, |
| "grad_norm": 0.4640503525733948, |
| "learning_rate": 5.329692723675994e-05, |
| "loss": 0.1587, |
| "mean_token_accuracy": 0.9431061984656693, |
| "num_tokens": 12809961.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 2.2467866323907453, |
| "grad_norm": 0.4997648596763611, |
| "learning_rate": 3.466934643431795e-05, |
| "loss": 0.1347, |
| "mean_token_accuracy": 0.950631283223629, |
| "num_tokens": 14099963.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 2.4524421593830334, |
| "grad_norm": 0.44113558530807495, |
| "learning_rate": 1.9364821412094857e-05, |
| "loss": 0.131, |
| "mean_token_accuracy": 0.9509662713855505, |
| "num_tokens": 15390859.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.5347043701799485, |
| "eval_loss": 0.19054362177848816, |
| "eval_mean_token_accuracy": 0.9369476150243711, |
| "eval_num_tokens": 15906771.0, |
| "eval_runtime": 277.9929, |
| "eval_samples_per_second": 0.701, |
| "eval_steps_per_second": 0.701, |
| "step": 124 |
| }, |
| { |
| "epoch": 2.658097686375321, |
| "grad_norm": 0.44858983159065247, |
| "learning_rate": 8.161819666783888e-06, |
| "loss": 0.1324, |
| "mean_token_accuracy": 0.9513279229402543, |
| "num_tokens": 16681649.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.8637532133676094, |
| "grad_norm": 0.4972115159034729, |
| "learning_rate": 1.6301839547892328e-06, |
| "loss": 0.1297, |
| "mean_token_accuracy": 0.9516113791614771, |
| "num_tokens": 17972810.0, |
| "step": 140 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 147, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 62, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 9.168897575363267e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
