| { | |
| "best_metric": 0.22190242, | |
| "best_model_checkpoint": "/home/patrickbarker/output/qwen2-vl-7b-instruct/v7-20241217-215255/checkpoint-25", | |
| "epoch": 5.0, | |
| "eval_steps": 200, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "acc": 0.81238556, | |
| "epoch": 0.2, | |
| "grad_norm": 65.66679331916022, | |
| "learning_rate": 0.0, | |
| "loss": 1.25573099, | |
| "memory(GiB)": 67.62, | |
| "step": 1, | |
| "train_speed(iter/s)": 0.023979 | |
| }, | |
| { | |
| "acc": 0.82108426, | |
| "epoch": 1.0, | |
| "grad_norm": 23.196112313305246, | |
| "learning_rate": 9.586097901876513e-06, | |
| "loss": 1.10042381, | |
| "memory(GiB)": 76.19, | |
| "step": 5, | |
| "train_speed(iter/s)": 0.030859 | |
| }, | |
| { | |
| "acc": 0.88789063, | |
| "epoch": 2.0, | |
| "grad_norm": 13.59313574447105, | |
| "learning_rate": 7.300595156136896e-06, | |
| "loss": 0.65092702, | |
| "memory(GiB)": 72.38, | |
| "step": 10, | |
| "train_speed(iter/s)": 0.032001 | |
| }, | |
| { | |
| "acc": 0.9200119, | |
| "epoch": 3.0, | |
| "grad_norm": 11.839492832829654, | |
| "learning_rate": 3.9833216627433574e-06, | |
| "loss": 0.44638877, | |
| "memory(GiB)": 72.38, | |
| "step": 15, | |
| "train_speed(iter/s)": 0.032401 | |
| }, | |
| { | |
| "acc": 0.92915792, | |
| "epoch": 4.0, | |
| "grad_norm": 7.035219421662484, | |
| "learning_rate": 1.122331402123254e-06, | |
| "loss": 0.35480633, | |
| "memory(GiB)": 72.38, | |
| "step": 20, | |
| "train_speed(iter/s)": 0.032608 | |
| }, | |
| { | |
| "acc": 0.93988991, | |
| "epoch": 5.0, | |
| "grad_norm": 3.037013055453519, | |
| "learning_rate": 1e-09, | |
| "loss": 0.33668923, | |
| "memory(GiB)": 72.38, | |
| "step": 25, | |
| "train_speed(iter/s)": 0.032732 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_acc": 0.9023861171366594, | |
| "eval_loss": 0.22190241515636444, | |
| "eval_runtime": 14.5521, | |
| "eval_samples_per_second": 1.374, | |
| "eval_steps_per_second": 0.206, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 57677248987136.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |