| { | |
| "best_metric": 100.0, | |
| "best_model_checkpoint": "/Users/kevinkwok/Desktop/whisper-small-cantonese/checkpoint-1200", | |
| "epoch": 6.83111954459203, | |
| "eval_steps": 400, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7513, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7588, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7222, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7234, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6998, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7416, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7151, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7279, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7443, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7879, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1e-05, | |
| "loss": 0.951, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0695, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2673, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1e-05, | |
| "loss": 1.3683, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1e-05, | |
| "loss": 1.3847, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6241, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 1.6757683753967285, | |
| "eval_runtime": 3755.6247, | |
| "eval_samples_per_second": 1.489, | |
| "eval_steps_per_second": 0.186, | |
| "eval_wer": 102.7664256523106, | |
| "eval_wer_ortho": 99.64387464387464, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6914, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6943, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7515, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8382, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8555, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9791, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8566, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0313, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1e-05, | |
| "loss": 2.019, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9514, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8461, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0464, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0249, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1e-05, | |
| "loss": 1.948, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0429, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0215, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_loss": 2.010716438293457, | |
| "eval_runtime": 3081.2923, | |
| "eval_samples_per_second": 1.815, | |
| "eval_steps_per_second": 0.227, | |
| "eval_wer": 100.17290160326941, | |
| "eval_wer_ortho": 100.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1e-05, | |
| "loss": 2.1012, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9747, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9878, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9765, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 1e-05, | |
| "loss": 2.1184, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9324, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9585, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9719, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0917, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0203, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9908, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9275, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9532, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8964, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 1e-05, | |
| "loss": 1.904, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9827, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_loss": 1.9757949113845825, | |
| "eval_runtime": 2892.4065, | |
| "eval_samples_per_second": 1.934, | |
| "eval_steps_per_second": 0.242, | |
| "eval_wer": 100.0, | |
| "eval_wer_ortho": 100.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0301, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 1e-05, | |
| "loss": 2.1268, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0044, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1e-05, | |
| "loss": 2.1126, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 1e-05, | |
| "loss": 2.02, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9804, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9873, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0043, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 1e-05, | |
| "loss": 1.965, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9411, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0059, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8808, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9022, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9633, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8834, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0119, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_loss": 1.9379667043685913, | |
| "eval_runtime": 9420.2123, | |
| "eval_samples_per_second": 0.594, | |
| "eval_steps_per_second": 0.074, | |
| "eval_wer": 100.0, | |
| "eval_wer_ortho": 100.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9068, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 1e-05, | |
| "loss": 1.841, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9803, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9377, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9931, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9932, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8911, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8134, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9463, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7938, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8559, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9709, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9548, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8475, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 1e-05, | |
| "loss": 1.99, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8693, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "eval_loss": 1.9047924280166626, | |
| "eval_runtime": 2877.2302, | |
| "eval_samples_per_second": 1.944, | |
| "eval_steps_per_second": 0.243, | |
| "eval_wer": 100.0, | |
| "eval_wer_ortho": 100.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 1e-05, | |
| "loss": 1.925, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9491, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9243, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8867, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9149, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0169, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8339, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8081, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9069, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8723, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8821, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8486, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9235, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 1e-05, | |
| "loss": 1.838, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8834, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9805, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "eval_loss": 1.8733848333358765, | |
| "eval_runtime": 2872.4313, | |
| "eval_samples_per_second": 1.947, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 100.01571832756994, | |
| "eval_wer_ortho": 100.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9637, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 1e-05, | |
| "loss": 1.836, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8557, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8696, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8709, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9039, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9168, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7614, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9177, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8272, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8066, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8952, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 1e-05, | |
| "loss": 1.857, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8715, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8985, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8382, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "eval_loss": 1.8439116477966309, | |
| "eval_runtime": 2891.6652, | |
| "eval_samples_per_second": 1.935, | |
| "eval_steps_per_second": 0.242, | |
| "eval_wer": 100.01571832756994, | |
| "eval_wer_ortho": 100.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9262, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9521, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7268, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7545, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.55, | |
| "learning_rate": 1e-05, | |
| "loss": 1.904, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8546, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7786, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7294, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "learning_rate": 1e-05, | |
| "loss": 1.758, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7159, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8201, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8107, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8754, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9456, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8856, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 1e-05, | |
| "loss": 1.818, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "eval_loss": 1.8148053884506226, | |
| "eval_runtime": 2870.6751, | |
| "eval_samples_per_second": 1.949, | |
| "eval_steps_per_second": 0.244, | |
| "eval_wer": 100.0314366551399, | |
| "eval_wer_ortho": 100.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7602, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8412, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7344, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7457, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7492, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8071, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8119, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9135, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8219, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8165, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8399, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8227, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6899, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7632, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7686, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8491, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "eval_loss": 1.7864172458648682, | |
| "eval_runtime": 3734.5543, | |
| "eval_samples_per_second": 1.498, | |
| "eval_steps_per_second": 0.187, | |
| "eval_wer": 100.0, | |
| "eval_wer_ortho": 100.0, | |
| "step": 3600 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 20000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 38, | |
| "save_steps": 400, | |
| "total_flos": 1.661732452122624e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |