| { | |
| "best_metric": 0.3867419362068176, | |
| "best_model_checkpoint": "finetuned-bangladeshi-traditional-food/checkpoint-200", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 256, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15625, | |
| "grad_norm": 104808.0078125, | |
| "learning_rate": 0.0001921875, | |
| "loss": 2.5683, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 121324.2109375, | |
| "learning_rate": 0.000184375, | |
| "loss": 2.0721, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 112550.9140625, | |
| "learning_rate": 0.00017656250000000002, | |
| "loss": 1.5956, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 90598.2265625, | |
| "learning_rate": 0.00016875, | |
| "loss": 1.3772, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "grad_norm": 100583.53125, | |
| "learning_rate": 0.0001609375, | |
| "loss": 1.1265, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 92339.28125, | |
| "learning_rate": 0.000153125, | |
| "loss": 1.0868, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.09375, | |
| "grad_norm": 94877.703125, | |
| "learning_rate": 0.00014531250000000002, | |
| "loss": 0.8778, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 129495.6640625, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.6839, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.40625, | |
| "grad_norm": 150202.59375, | |
| "learning_rate": 0.0001296875, | |
| "loss": 0.6589, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 128498.6953125, | |
| "learning_rate": 0.00012187500000000001, | |
| "loss": 0.703, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "eval_accuracy": 0.9022346368715084, | |
| "eval_loss": 0.612750232219696, | |
| "eval_runtime": 6.1858, | |
| "eval_samples_per_second": 57.874, | |
| "eval_steps_per_second": 3.718, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.71875, | |
| "grad_norm": 172968.234375, | |
| "learning_rate": 0.0001140625, | |
| "loss": 0.6179, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 193931.03125, | |
| "learning_rate": 0.00010625000000000001, | |
| "loss": 0.5416, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.03125, | |
| "grad_norm": 100722.890625, | |
| "learning_rate": 9.84375e-05, | |
| "loss": 0.4705, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.1875, | |
| "grad_norm": 111848.0234375, | |
| "learning_rate": 9.062500000000001e-05, | |
| "loss": 0.4895, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.34375, | |
| "grad_norm": 102607.46875, | |
| "learning_rate": 8.28125e-05, | |
| "loss": 0.4002, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 228628.328125, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.3947, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.65625, | |
| "grad_norm": 144436.234375, | |
| "learning_rate": 6.71875e-05, | |
| "loss": 0.3387, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.8125, | |
| "grad_norm": 218134.15625, | |
| "learning_rate": 5.9375e-05, | |
| "loss": 0.3805, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.96875, | |
| "grad_norm": 132064.421875, | |
| "learning_rate": 5.15625e-05, | |
| "loss": 0.3356, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 24524.32421875, | |
| "learning_rate": 4.375e-05, | |
| "loss": 0.252, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "eval_accuracy": 0.9189944134078212, | |
| "eval_loss": 0.3867419362068176, | |
| "eval_runtime": 6.1019, | |
| "eval_samples_per_second": 58.67, | |
| "eval_steps_per_second": 3.769, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.28125, | |
| "grad_norm": 37676.6171875, | |
| "learning_rate": 3.59375e-05, | |
| "loss": 0.2721, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.4375, | |
| "grad_norm": 105781.78125, | |
| "learning_rate": 2.8125000000000003e-05, | |
| "loss": 0.2583, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.59375, | |
| "grad_norm": 42383.23828125, | |
| "learning_rate": 2.0312500000000002e-05, | |
| "loss": 0.2163, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 24589.4375, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2903, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.90625, | |
| "grad_norm": 235633.53125, | |
| "learning_rate": 4.6875000000000004e-06, | |
| "loss": 0.2813, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 256, | |
| "total_flos": 6.287051649950024e+17, | |
| "train_loss": 0.7197832907550037, | |
| "train_runtime": 268.6443, | |
| "train_samples_per_second": 30.196, | |
| "train_steps_per_second": 0.953 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 256, | |
| "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": 6.287051649950024e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |