| { | |
| "best_metric": 0.26493218541145325, | |
| "best_model_checkpoint": "./outputs/google/flan-t5-small_02112024_16_44_29/checkpoint-10275", | |
| "epoch": 8.0, | |
| "global_step": 16440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.9513381995133826e-05, | |
| "loss": 0.6278, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.9026763990267643e-05, | |
| "loss": 0.4278, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.854014598540147e-05, | |
| "loss": 0.3906, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 4.8053527980535284e-05, | |
| "loss": 0.3726, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.3402346555086388, | |
| "eval_loss": 0.31867411732673645, | |
| "eval_precision": 0.3932500949967925, | |
| "eval_recall": 0.34399184859663345, | |
| "eval_runtime": 24.9698, | |
| "eval_samples_per_second": 36.564, | |
| "eval_steps_per_second": 4.606, | |
| "step": 2055 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.75669099756691e-05, | |
| "loss": 0.3505, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 4.708029197080292e-05, | |
| "loss": 0.3385, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 4.659367396593674e-05, | |
| "loss": 0.3299, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 4.610705596107056e-05, | |
| "loss": 0.316, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.44100378584011557, | |
| "eval_loss": 0.29325029253959656, | |
| "eval_precision": 0.5725633373653386, | |
| "eval_recall": 0.41354031034704636, | |
| "eval_runtime": 30.7804, | |
| "eval_samples_per_second": 29.662, | |
| "eval_steps_per_second": 3.736, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 4.5620437956204383e-05, | |
| "loss": 0.3077, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 4.51338199513382e-05, | |
| "loss": 0.2965, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 4.4647201946472024e-05, | |
| "loss": 0.3038, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 4.416058394160584e-05, | |
| "loss": 0.2948, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.49142134605574894, | |
| "eval_loss": 0.27041032910346985, | |
| "eval_precision": 0.5482614717698139, | |
| "eval_recall": 0.4672472050791782, | |
| "eval_runtime": 32.1755, | |
| "eval_samples_per_second": 28.376, | |
| "eval_steps_per_second": 3.574, | |
| "step": 6165 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 4.3673965936739665e-05, | |
| "loss": 0.286, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 4.318734793187348e-05, | |
| "loss": 0.2794, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 4.27007299270073e-05, | |
| "loss": 0.2684, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 4.221411192214112e-05, | |
| "loss": 0.2759, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.5152580961177389, | |
| "eval_loss": 0.2656283974647522, | |
| "eval_precision": 0.561024311684815, | |
| "eval_recall": 0.4928332834870085, | |
| "eval_runtime": 32.9996, | |
| "eval_samples_per_second": 27.667, | |
| "eval_steps_per_second": 3.485, | |
| "step": 8220 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 4.172749391727494e-05, | |
| "loss": 0.2628, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.38, | |
| "learning_rate": 4.124087591240876e-05, | |
| "loss": 0.2583, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 4.075425790754258e-05, | |
| "loss": 0.2567, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 4.02676399026764e-05, | |
| "loss": 0.2563, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1": 0.5254077620147211, | |
| "eval_loss": 0.26493218541145325, | |
| "eval_precision": 0.5869446986868645, | |
| "eval_recall": 0.49228704075934504, | |
| "eval_runtime": 31.0527, | |
| "eval_samples_per_second": 29.402, | |
| "eval_steps_per_second": 3.703, | |
| "step": 10275 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 3.978102189781022e-05, | |
| "loss": 0.2553, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 3.929440389294404e-05, | |
| "loss": 0.2482, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 3.8807785888077864e-05, | |
| "loss": 0.2412, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.84, | |
| "learning_rate": 3.832116788321168e-05, | |
| "loss": 0.2376, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_f1": 0.5293819723661756, | |
| "eval_loss": 0.2736237049102783, | |
| "eval_precision": 0.5698156189027613, | |
| "eval_recall": 0.5158894882436182, | |
| "eval_runtime": 33.8492, | |
| "eval_samples_per_second": 26.973, | |
| "eval_steps_per_second": 3.397, | |
| "step": 12330 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 3.78345498783455e-05, | |
| "loss": 0.234, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 3.734793187347932e-05, | |
| "loss": 0.2259, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 3.686131386861314e-05, | |
| "loss": 0.2312, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 3.637469586374696e-05, | |
| "loss": 0.2306, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_f1": 0.5377831918627073, | |
| "eval_loss": 0.27052244544029236, | |
| "eval_precision": 0.5662163692648797, | |
| "eval_recall": 0.5227524158025398, | |
| "eval_runtime": 33.9607, | |
| "eval_samples_per_second": 26.884, | |
| "eval_steps_per_second": 3.386, | |
| "step": 14385 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 3.588807785888078e-05, | |
| "loss": 0.2355, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 7.3, | |
| "learning_rate": 3.5401459854014604e-05, | |
| "loss": 0.2126, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 3.491484184914842e-05, | |
| "loss": 0.2236, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 3.4428223844282245e-05, | |
| "loss": 0.2186, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_f1": 0.5354115040070326, | |
| "eval_loss": 0.28732189536094666, | |
| "eval_precision": 0.5835637777117146, | |
| "eval_recall": 0.5252217659081543, | |
| "eval_runtime": 35.8593, | |
| "eval_samples_per_second": 25.461, | |
| "eval_steps_per_second": 3.207, | |
| "step": 16440 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 16440, | |
| "total_flos": 5028888562409472.0, | |
| "train_loss": 0.288517827129132, | |
| "train_runtime": 1921.0513, | |
| "train_samples_per_second": 106.934, | |
| "train_steps_per_second": 26.743 | |
| } | |
| ], | |
| "max_steps": 51375, | |
| "num_train_epochs": 25, | |
| "total_flos": 5028888562409472.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |