| { | |
| "best_metric": Infinity, | |
| "best_model_checkpoint": "./checkpoint-1000", | |
| "epoch": 4.999928330824912, | |
| "global_step": 17440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 3.7499999999999997e-06, | |
| "loss": 12.8121, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 7.499999999999999e-06, | |
| "loss": 6.7805, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 1.1249999999999999e-05, | |
| "loss": 4.6147, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.4999999999999999e-05, | |
| "loss": 3.955, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.875e-05, | |
| "loss": 3.5002, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 3.1912, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.6249999999999998e-05, | |
| "loss": 3.0349, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 2.99625e-05, | |
| "loss": 3.0058, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 3.37125e-05, | |
| "loss": 2.9392, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3.7462499999999996e-05, | |
| "loss": 2.9114, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 711.5482, | |
| "eval_samples_per_second": 22.516, | |
| "eval_steps_per_second": 1.408, | |
| "eval_wer": 0.9997048122028068, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.12125e-05, | |
| "loss": 2.8881, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.4962499999999995e-05, | |
| "loss": 2.7567, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.871249999999999e-05, | |
| "loss": 2.2193, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 5.2462499999999994e-05, | |
| "loss": 1.8226, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 5.62125e-05, | |
| "loss": 1.6558, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 5.9962499999999994e-05, | |
| "loss": 1.5479, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 6.37125e-05, | |
| "loss": 1.4422, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 6.746249999999999e-05, | |
| "loss": 1.3521, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 7.121249999999999e-05, | |
| "loss": 1.2988, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 7.49625e-05, | |
| "loss": 1.2436, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 723.1036, | |
| "eval_samples_per_second": 22.156, | |
| "eval_steps_per_second": 1.386, | |
| "eval_wer": 0.43096134964994576, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 7.451910621761658e-05, | |
| "loss": 1.2041, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 7.403335492227979e-05, | |
| "loss": 1.1714, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 7.354760362694299e-05, | |
| "loss": 1.1481, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 7.306185233160621e-05, | |
| "loss": 1.1268, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 7.258095854922278e-05, | |
| "loss": 1.1271, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 7.2095207253886e-05, | |
| "loss": 1.0927, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.160945595854922e-05, | |
| "loss": 1.0879, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 7.112370466321243e-05, | |
| "loss": 1.0699, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 7.063795336787565e-05, | |
| "loss": 1.0669, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 7.015220207253885e-05, | |
| "loss": 1.0552, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 706.7689, | |
| "eval_samples_per_second": 22.668, | |
| "eval_steps_per_second": 1.418, | |
| "eval_wer": 0.3144455923969891, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 6.966645077720206e-05, | |
| "loss": 1.0525, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 6.918069948186528e-05, | |
| "loss": 1.0475, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 6.86949481865285e-05, | |
| "loss": 1.0333, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 6.82091968911917e-05, | |
| "loss": 1.0227, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.772344559585491e-05, | |
| "loss": 1.0154, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 6.723769430051813e-05, | |
| "loss": 1.0084, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 6.675194300518135e-05, | |
| "loss": 1.0153, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 6.626619170984455e-05, | |
| "loss": 1.0059, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 6.578529792746114e-05, | |
| "loss": 1.0069, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 6.529954663212434e-05, | |
| "loss": 1.0044, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 720.8047, | |
| "eval_samples_per_second": 22.227, | |
| "eval_steps_per_second": 1.39, | |
| "eval_wer": 0.2814038104894342, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 6.481379533678756e-05, | |
| "loss": 0.9918, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 6.432804404145077e-05, | |
| "loss": 0.9913, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 6.384229274611399e-05, | |
| "loss": 0.9955, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 6.33565414507772e-05, | |
| "loss": 0.989, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 6.287079015544041e-05, | |
| "loss": 0.9825, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 6.238503886010362e-05, | |
| "loss": 0.9793, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 6.189928756476684e-05, | |
| "loss": 0.9768, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 6.141353626943006e-05, | |
| "loss": 0.9712, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 6.092778497409326e-05, | |
| "loss": 0.9777, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 6.0442033678756475e-05, | |
| "loss": 0.9718, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 706.8206, | |
| "eval_samples_per_second": 22.666, | |
| "eval_steps_per_second": 1.418, | |
| "eval_wer": 0.26581661137243073, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 5.995628238341968e-05, | |
| "loss": 0.967, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 5.947053108808289e-05, | |
| "loss": 0.9688, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 5.898477979274611e-05, | |
| "loss": 0.9677, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 5.8499028497409324e-05, | |
| "loss": 0.961, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 5.8013277202072536e-05, | |
| "loss": 0.9528, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 5.752752590673574e-05, | |
| "loss": 0.959, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 5.7041774611398954e-05, | |
| "loss": 0.9615, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 5.6556023316062174e-05, | |
| "loss": 0.9587, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 5.607512953367875e-05, | |
| "loss": 0.9573, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 5.5589378238341966e-05, | |
| "loss": 0.9502, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 705.7107, | |
| "eval_samples_per_second": 22.702, | |
| "eval_steps_per_second": 1.42, | |
| "eval_wer": 0.2566016183991837, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 5.510362694300518e-05, | |
| "loss": 0.9515, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 5.461787564766839e-05, | |
| "loss": 0.9479, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 5.4132124352331596e-05, | |
| "loss": 0.94, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 5.3646373056994816e-05, | |
| "loss": 0.9363, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 5.316062176165803e-05, | |
| "loss": 0.9419, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 5.267487046632124e-05, | |
| "loss": 0.9435, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 5.218911917098445e-05, | |
| "loss": 0.9406, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 5.170336787564766e-05, | |
| "loss": 0.9457, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 5.121761658031088e-05, | |
| "loss": 0.9343, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 5.073186528497409e-05, | |
| "loss": 0.9418, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 712.8454, | |
| "eval_samples_per_second": 22.475, | |
| "eval_steps_per_second": 1.406, | |
| "eval_wer": 0.24762405908889645, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 5.02461139896373e-05, | |
| "loss": 0.9401, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 4.9760362694300515e-05, | |
| "loss": 0.9375, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 4.927461139896372e-05, | |
| "loss": 0.9362, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 4.878886010362694e-05, | |
| "loss": 0.9249, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 4.830310880829015e-05, | |
| "loss": 0.9283, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 4.782221502590673e-05, | |
| "loss": 0.9371, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 4.7336463730569944e-05, | |
| "loss": 0.9346, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 4.6850712435233157e-05, | |
| "loss": 0.9266, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 4.636496113989637e-05, | |
| "loss": 0.9285, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 4.587920984455958e-05, | |
| "loss": 0.9215, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 699.6113, | |
| "eval_samples_per_second": 22.9, | |
| "eval_steps_per_second": 1.432, | |
| "eval_wer": 0.24195131968196723, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 4.5393458549222794e-05, | |
| "loss": 0.9211, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 4.4907707253886006e-05, | |
| "loss": 0.9198, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 4.442195595854922e-05, | |
| "loss": 0.9257, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 4.393620466321243e-05, | |
| "loss": 0.9279, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 4.345045336787564e-05, | |
| "loss": 0.9111, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 4.296470207253886e-05, | |
| "loss": 0.9176, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 4.247895077720207e-05, | |
| "loss": 0.9169, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 4.199319948186528e-05, | |
| "loss": 0.9103, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 4.150744818652849e-05, | |
| "loss": 0.9102, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 4.102655440414507e-05, | |
| "loss": 0.9236, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 704.4849, | |
| "eval_samples_per_second": 22.741, | |
| "eval_steps_per_second": 1.422, | |
| "eval_wer": 0.23878125942515385, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 4.0540803108808285e-05, | |
| "loss": 0.9129, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 4.0055051813471504e-05, | |
| "loss": 0.9117, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 3.956930051813472e-05, | |
| "loss": 0.912, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 3.908354922279792e-05, | |
| "loss": 0.9013, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 3.8597797927461135e-05, | |
| "loss": 0.9128, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 3.811204663212435e-05, | |
| "loss": 0.9061, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 3.7626295336787566e-05, | |
| "loss": 0.9055, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 3.714054404145078e-05, | |
| "loss": 0.9019, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 3.6654792746113984e-05, | |
| "loss": 0.9, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "learning_rate": 3.61690414507772e-05, | |
| "loss": 0.9014, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 702.0047, | |
| "eval_samples_per_second": 22.822, | |
| "eval_steps_per_second": 1.427, | |
| "eval_wer": 0.2353801826314067, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 3.5683290155440416e-05, | |
| "loss": 0.9045, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 3.519753886010362e-05, | |
| "loss": 0.8981, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 3.471178756476684e-05, | |
| "loss": 0.898, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 3.4226036269430046e-05, | |
| "loss": 0.8926, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 3.374028497409326e-05, | |
| "loss": 0.9, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 3.325453367875648e-05, | |
| "loss": 0.8928, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 3.276878238341968e-05, | |
| "loss": 0.8844, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.22830310880829e-05, | |
| "loss": 0.8844, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 3.179727979274611e-05, | |
| "loss": 0.89, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 3.131152849740932e-05, | |
| "loss": 0.8814, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 712.5543, | |
| "eval_samples_per_second": 22.484, | |
| "eval_steps_per_second": 1.406, | |
| "eval_wer": 0.23118979933646916, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 3.082577720207254e-05, | |
| "loss": 0.8912, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 3.034002590673575e-05, | |
| "loss": 0.8956, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 2.985427461139896e-05, | |
| "loss": 0.8866, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 2.9368523316062173e-05, | |
| "loss": 0.8851, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 2.888277202072539e-05, | |
| "loss": 0.8897, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 2.8397020725388598e-05, | |
| "loss": 0.8832, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 2.791126943005181e-05, | |
| "loss": 0.8887, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 2.7425518134715023e-05, | |
| "loss": 0.8847, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 2.6939766839378235e-05, | |
| "loss": 0.8754, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 2.645401554404145e-05, | |
| "loss": 0.8809, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 721.0245, | |
| "eval_samples_per_second": 22.22, | |
| "eval_steps_per_second": 1.39, | |
| "eval_wer": 0.22846893790147144, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 2.596826424870466e-05, | |
| "loss": 0.8739, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 2.5482512953367873e-05, | |
| "loss": 0.8713, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 2.4996761658031085e-05, | |
| "loss": 0.8833, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 2.4511010362694297e-05, | |
| "loss": 0.8756, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 2.4025259067357513e-05, | |
| "loss": 0.8705, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 2.3539507772020722e-05, | |
| "loss": 0.874, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 2.3053756476683938e-05, | |
| "loss": 0.8681, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 2.2568005181347147e-05, | |
| "loss": 0.8699, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 2.208225388601036e-05, | |
| "loss": 0.8749, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 2.1596502590673575e-05, | |
| "loss": 0.8717, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 704.6479, | |
| "eval_samples_per_second": 22.736, | |
| "eval_steps_per_second": 1.422, | |
| "eval_wer": 0.22629353217867845, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 2.1110751295336784e-05, | |
| "loss": 0.8725, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 2.0625e-05, | |
| "loss": 0.8678, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 2.0139248704663212e-05, | |
| "loss": 0.8618, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 1.9658354922279792e-05, | |
| "loss": 0.8646, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 1.9172603626943005e-05, | |
| "loss": 0.8746, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 1.8686852331606217e-05, | |
| "loss": 0.863, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 1.8205958549222797e-05, | |
| "loss": 0.862, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 1.772020725388601e-05, | |
| "loss": 0.8678, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 1.723445595854922e-05, | |
| "loss": 0.8694, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 1.6748704663212434e-05, | |
| "loss": 0.8787, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 708.1909, | |
| "eval_samples_per_second": 22.622, | |
| "eval_steps_per_second": 1.415, | |
| "eval_wer": 0.2217630412043662, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 1.6262953367875646e-05, | |
| "loss": 0.8657, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 1.577720207253886e-05, | |
| "loss": 0.8631, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 1.529630829015544e-05, | |
| "loss": 0.8658, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 1.4810556994818651e-05, | |
| "loss": 0.855, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 1.4324805699481864e-05, | |
| "loss": 0.8502, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 1.3839054404145076e-05, | |
| "loss": 0.8625, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 1.335330310880829e-05, | |
| "loss": 0.8598, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 1.2867551813471502e-05, | |
| "loss": 0.8574, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 1.2381800518134713e-05, | |
| "loss": 0.8575, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 1.1896049222797926e-05, | |
| "loss": 0.8567, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 705.2463, | |
| "eval_samples_per_second": 22.717, | |
| "eval_steps_per_second": 1.421, | |
| "eval_wer": 0.21933736756656164, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 1.1410297927461138e-05, | |
| "loss": 0.8474, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 1.0924546632124352e-05, | |
| "loss": 0.8519, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 1.0438795336787564e-05, | |
| "loss": 0.8487, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 9.953044041450777e-06, | |
| "loss": 0.8479, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 9.467292746113987e-06, | |
| "loss": 0.8463, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 8.981541450777202e-06, | |
| "loss": 0.8459, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 8.500647668393781e-06, | |
| "loss": 0.853, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 8.014896373056994e-06, | |
| "loss": 0.8456, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 7.529145077720206e-06, | |
| "loss": 0.8501, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 7.0433937823834195e-06, | |
| "loss": 0.8488, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 702.6461, | |
| "eval_samples_per_second": 22.801, | |
| "eval_steps_per_second": 1.426, | |
| "eval_wer": 0.2187084892160197, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 6.557642487046632e-06, | |
| "loss": 0.8491, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 6.0718911917098434e-06, | |
| "loss": 0.8415, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 5.586139896373057e-06, | |
| "loss": 0.8467, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 5.100388601036269e-06, | |
| "loss": 0.8356, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 4.6146373056994814e-06, | |
| "loss": 0.8444, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.128886010362694e-06, | |
| "loss": 0.8485, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 3.6431347150259062e-06, | |
| "loss": 0.8424, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 3.1573834196891186e-06, | |
| "loss": 0.8435, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 2.6716321243523314e-06, | |
| "loss": 0.8444, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 2.185880829015544e-06, | |
| "loss": 0.8359, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "eval_loss": Infinity, | |
| "eval_runtime": 698.574, | |
| "eval_samples_per_second": 22.934, | |
| "eval_steps_per_second": 1.434, | |
| "eval_wer": 0.2171683789697946, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 1.7001295336787564e-06, | |
| "loss": 0.8419, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 1.2143782383419686e-06, | |
| "loss": 0.8392, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 7.286269430051813e-07, | |
| "loss": 0.8369, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 2.4287564766839375e-07, | |
| "loss": 0.8428, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 17440, | |
| "total_flos": 3.487279250614594e+20, | |
| "train_loss": 1.187884036554109, | |
| "train_runtime": 143109.8834, | |
| "train_samples_per_second": 15.599, | |
| "train_steps_per_second": 0.122 | |
| } | |
| ], | |
| "max_steps": 17440, | |
| "num_train_epochs": 5, | |
| "total_flos": 3.487279250614594e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |