| { | |
| "best_metric": 0.5026881098747253, | |
| "best_model_checkpoint": "finetuned-aiimg/checkpoint-100", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 208, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19230769230769232, | |
| "grad_norm": 5.460508346557617, | |
| "learning_rate": 0.00019038461538461538, | |
| "loss": 0.3821, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 3.0704050064086914, | |
| "learning_rate": 0.00018076923076923077, | |
| "loss": 0.3817, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5769230769230769, | |
| "grad_norm": 2.445289134979248, | |
| "learning_rate": 0.00017115384615384616, | |
| "loss": 0.504, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.5516889095306396, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 0.5648, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 1.973536491394043, | |
| "learning_rate": 0.00015192307692307692, | |
| "loss": 0.5931, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 3.3615033626556396, | |
| "learning_rate": 0.0001423076923076923, | |
| "loss": 0.435, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.3461538461538463, | |
| "grad_norm": 2.4279274940490723, | |
| "learning_rate": 0.0001326923076923077, | |
| "loss": 0.2569, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 2.7991394996643066, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 0.425, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.7307692307692308, | |
| "grad_norm": 4.332102298736572, | |
| "learning_rate": 0.00011346153846153846, | |
| "loss": 0.399, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 2.103902816772461, | |
| "learning_rate": 0.00010384615384615386, | |
| "loss": 0.4149, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "eval_accuracy": 0.7808219178082192, | |
| "eval_loss": 0.5026881098747253, | |
| "eval_runtime": 5.4124, | |
| "eval_samples_per_second": 26.975, | |
| "eval_steps_per_second": 3.51, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.1153846153846154, | |
| "grad_norm": 1.4146589040756226, | |
| "learning_rate": 9.423076923076924e-05, | |
| "loss": 0.3235, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 3.8487939834594727, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.2436, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 3.027752161026001, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.2866, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 1.7752703428268433, | |
| "learning_rate": 6.538461538461539e-05, | |
| "loss": 0.2232, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 1.2024428844451904, | |
| "learning_rate": 5.576923076923077e-05, | |
| "loss": 0.2446, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.5080825090408325, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.1841, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.269230769230769, | |
| "grad_norm": 0.30641067028045654, | |
| "learning_rate": 3.653846153846154e-05, | |
| "loss": 0.1284, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 0.8220038414001465, | |
| "learning_rate": 2.6923076923076923e-05, | |
| "loss": 0.0875, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.6538461538461537, | |
| "grad_norm": 5.190620422363281, | |
| "learning_rate": 1.730769230769231e-05, | |
| "loss": 0.1773, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.908566951751709, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1273, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "eval_accuracy": 0.821917808219178, | |
| "eval_loss": 0.5271799564361572, | |
| "eval_runtime": 7.1914, | |
| "eval_samples_per_second": 20.302, | |
| "eval_steps_per_second": 2.642, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 208, | |
| "total_flos": 2.5603353368661197e+17, | |
| "train_loss": 0.3125012345038928, | |
| "train_runtime": 224.4798, | |
| "train_samples_per_second": 14.718, | |
| "train_steps_per_second": 0.927 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 208, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.5603353368661197e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |