| { | |
| "best_metric": 4.576775972964242e-05, | |
| "best_model_checkpoint": "llava-1.5-7b-sroie/checkpoint-1400", | |
| "epoch": 11.2, | |
| "eval_steps": 100, | |
| "global_step": 1400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.2929452955722809, | |
| "learning_rate": 1.3776e-05, | |
| "loss": 0.572, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.03328470513224602, | |
| "eval_runtime": 60.4001, | |
| "eval_samples_per_second": 2.086, | |
| "eval_steps_per_second": 2.086, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.01092437468469143, | |
| "learning_rate": 1.3552e-05, | |
| "loss": 0.0046, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.0008874661289155483, | |
| "eval_runtime": 61.352, | |
| "eval_samples_per_second": 2.054, | |
| "eval_steps_per_second": 2.054, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.005791354924440384, | |
| "learning_rate": 1.3327999999999998e-05, | |
| "loss": 0.0007, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.0004409343237057328, | |
| "eval_runtime": 61.1557, | |
| "eval_samples_per_second": 2.06, | |
| "eval_steps_per_second": 2.06, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.002793485065922141, | |
| "learning_rate": 1.3104e-05, | |
| "loss": 0.0004, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 0.00029093530611135066, | |
| "eval_runtime": 60.6893, | |
| "eval_samples_per_second": 2.076, | |
| "eval_steps_per_second": 2.076, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.0015204106457531452, | |
| "learning_rate": 1.288e-05, | |
| "loss": 0.0002, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.0002039903192780912, | |
| "eval_runtime": 61.3698, | |
| "eval_samples_per_second": 2.053, | |
| "eval_steps_per_second": 2.053, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 0.0009014272363856435, | |
| "learning_rate": 1.2656e-05, | |
| "loss": 0.0002, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": 0.00016169117589015514, | |
| "eval_runtime": 60.6043, | |
| "eval_samples_per_second": 2.079, | |
| "eval_steps_per_second": 2.079, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 0.0006670297589153051, | |
| "learning_rate": 1.2432e-05, | |
| "loss": 0.0001, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "eval_loss": 0.0001274394162464887, | |
| "eval_runtime": 60.5564, | |
| "eval_samples_per_second": 2.081, | |
| "eval_steps_per_second": 2.081, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 0.00041761380271054804, | |
| "learning_rate": 1.2207999999999999e-05, | |
| "loss": 0.0001, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_loss": 0.000104410013591405, | |
| "eval_runtime": 60.3212, | |
| "eval_samples_per_second": 2.089, | |
| "eval_steps_per_second": 2.089, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "grad_norm": 0.0005114951636642218, | |
| "learning_rate": 1.1983999999999999e-05, | |
| "loss": 0.0001, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "eval_loss": 8.86739362613298e-05, | |
| "eval_runtime": 60.7862, | |
| "eval_samples_per_second": 2.073, | |
| "eval_steps_per_second": 2.073, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.00042668674723245203, | |
| "learning_rate": 1.176e-05, | |
| "loss": 0.0001, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 7.54967040847987e-05, | |
| "eval_runtime": 60.5531, | |
| "eval_samples_per_second": 2.081, | |
| "eval_steps_per_second": 2.081, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "grad_norm": 0.0005093810614198446, | |
| "learning_rate": 1.1536e-05, | |
| "loss": 0.0001, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "eval_loss": 6.56623815302737e-05, | |
| "eval_runtime": 60.7572, | |
| "eval_samples_per_second": 2.074, | |
| "eval_steps_per_second": 2.074, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "grad_norm": 0.00028865510830655694, | |
| "learning_rate": 1.1312000000000001e-05, | |
| "loss": 0.0001, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "eval_loss": 5.7752622524276376e-05, | |
| "eval_runtime": 61.9729, | |
| "eval_samples_per_second": 2.033, | |
| "eval_steps_per_second": 2.033, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "grad_norm": 0.00028205872513353825, | |
| "learning_rate": 1.1088e-05, | |
| "loss": 0.0001, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "eval_loss": 5.1226033974671736e-05, | |
| "eval_runtime": 60.9193, | |
| "eval_samples_per_second": 2.068, | |
| "eval_steps_per_second": 2.068, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "grad_norm": 0.00026295348652638495, | |
| "learning_rate": 1.0864e-05, | |
| "loss": 0.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "eval_loss": 4.576775972964242e-05, | |
| "eval_runtime": 61.4909, | |
| "eval_samples_per_second": 2.049, | |
| "eval_steps_per_second": 2.049, | |
| "step": 1400 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 6250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.025909547008e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |