| { | |
| "best_metric": 2.9007203578948975, | |
| "best_model_checkpoint": "/content/drive/MyDrive/W210 Capstone - Lyric Generation with Melody/loaf/models/lyrlen/bart/bart-finetuned-lyrlen-128-special_tokens/checkpoint-2000", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 2.8254361152648926, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 3.2828, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 3.0014798641204834, | |
| "eval_runtime": 133.6176, | |
| "eval_samples_per_second": 22.452, | |
| "eval_steps_per_second": 1.407, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 2.1916661262512207, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 3.0513, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 2.9360573291778564, | |
| "eval_runtime": 134.0182, | |
| "eval_samples_per_second": 22.385, | |
| "eval_steps_per_second": 1.403, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.2003419399261475, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 2.9573, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.911083459854126, | |
| "eval_runtime": 135.0831, | |
| "eval_samples_per_second": 22.209, | |
| "eval_steps_per_second": 1.392, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 2.2140896320343018, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.8841, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 2.9007203578948975, | |
| "eval_runtime": 133.4383, | |
| "eval_samples_per_second": 22.482, | |
| "eval_steps_per_second": 1.409, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 2.1458683013916016, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 2.8352, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_loss": 2.9764463901519775, | |
| "eval_runtime": 128.7719, | |
| "eval_samples_per_second": 23.297, | |
| "eval_steps_per_second": 1.46, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.3623976707458496, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.7897, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.9605724811553955, | |
| "eval_runtime": 129.2199, | |
| "eval_samples_per_second": 23.216, | |
| "eval_steps_per_second": 1.455, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "grad_norm": 2.2417197227478027, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 2.7511, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 2.9489927291870117, | |
| "eval_runtime": 128.5445, | |
| "eval_samples_per_second": 23.338, | |
| "eval_steps_per_second": 1.463, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 1.9677695035934448, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.7284, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_loss": 2.945847749710083, | |
| "eval_runtime": 129.0136, | |
| "eval_samples_per_second": 23.253, | |
| "eval_steps_per_second": 1.457, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.6409881114959717, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.7167, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.9470016956329346, | |
| "eval_runtime": 129.7931, | |
| "eval_samples_per_second": 23.114, | |
| "eval_steps_per_second": 1.448, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "grad_norm": 2.425532341003418, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 2.7226, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_loss": 2.941845178604126, | |
| "eval_runtime": 133.0721, | |
| "eval_samples_per_second": 22.544, | |
| "eval_steps_per_second": 1.413, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "grad_norm": 1.9772043228149414, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 2.6823, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "eval_loss": 2.931675672531128, | |
| "eval_runtime": 130.4418, | |
| "eval_samples_per_second": 22.999, | |
| "eval_steps_per_second": 1.441, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.1570191383361816, | |
| "learning_rate": 0.0, | |
| "loss": 2.6445, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.9388811588287354, | |
| "eval_runtime": 131.227, | |
| "eval_samples_per_second": 22.861, | |
| "eval_steps_per_second": 1.433, | |
| "step": 6000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 2.6005255225344e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |