| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.4149274849663955, | |
| "eval_steps": 1000, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017686593562079942, | |
| "grad_norm": 28.95281982421875, | |
| "learning_rate": 4.6000000000000004e-07, | |
| "loss": 1.849, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.035373187124159884, | |
| "grad_norm": 21.0062255859375, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 1.7562, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05305978068623983, | |
| "grad_norm": 21.639610290527344, | |
| "learning_rate": 1.46e-06, | |
| "loss": 1.6167, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.07074637424831977, | |
| "grad_norm": 19.797197341918945, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 1.4966, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08843296781039972, | |
| "grad_norm": 19.28240203857422, | |
| "learning_rate": 2.46e-06, | |
| "loss": 1.2549, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.10611956137247966, | |
| "grad_norm": 19.18793487548828, | |
| "learning_rate": 2.96e-06, | |
| "loss": 1.1787, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.1238061549345596, | |
| "grad_norm": 17.472681045532227, | |
| "learning_rate": 3.46e-06, | |
| "loss": 1.074, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.14149274849663954, | |
| "grad_norm": 17.52532386779785, | |
| "learning_rate": 3.96e-06, | |
| "loss": 1.1118, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1591793420587195, | |
| "grad_norm": 17.355623245239258, | |
| "learning_rate": 4.4600000000000005e-06, | |
| "loss": 1.0218, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.17686593562079944, | |
| "grad_norm": 18.548181533813477, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 1.0106, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.19455252918287938, | |
| "grad_norm": 18.0971736907959, | |
| "learning_rate": 5.460000000000001e-06, | |
| "loss": 0.9914, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.21223912274495932, | |
| "grad_norm": 16.739173889160156, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 0.96, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22992571630703926, | |
| "grad_norm": 17.19851303100586, | |
| "learning_rate": 6.460000000000001e-06, | |
| "loss": 0.9472, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.2476123098691192, | |
| "grad_norm": 18.49357795715332, | |
| "learning_rate": 6.96e-06, | |
| "loss": 0.8836, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.26529890343119916, | |
| "grad_norm": 17.01639175415039, | |
| "learning_rate": 7.4600000000000006e-06, | |
| "loss": 0.8465, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.2829854969932791, | |
| "grad_norm": 16.33963966369629, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 0.8472, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.30067209055535904, | |
| "grad_norm": 16.123769760131836, | |
| "learning_rate": 8.46e-06, | |
| "loss": 0.8425, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.318358684117439, | |
| "grad_norm": 17.656047821044922, | |
| "learning_rate": 8.96e-06, | |
| "loss": 0.8385, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.3360452776795189, | |
| "grad_norm": 17.501123428344727, | |
| "learning_rate": 9.460000000000001e-06, | |
| "loss": 0.8041, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.3537318712415989, | |
| "grad_norm": 15.51560115814209, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 0.7809, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3714184648036788, | |
| "grad_norm": 14.336594581604004, | |
| "learning_rate": 9.846666666666668e-06, | |
| "loss": 0.7417, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.38910505836575876, | |
| "grad_norm": 14.65346908569336, | |
| "learning_rate": 9.68e-06, | |
| "loss": 0.7605, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.40679165192783867, | |
| "grad_norm": 15.94804573059082, | |
| "learning_rate": 9.513333333333334e-06, | |
| "loss": 0.7472, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.42447824548991864, | |
| "grad_norm": 17.208477020263672, | |
| "learning_rate": 9.346666666666666e-06, | |
| "loss": 0.7179, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.4421648390519986, | |
| "grad_norm": 16.141422271728516, | |
| "learning_rate": 9.180000000000002e-06, | |
| "loss": 0.7047, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.4598514326140785, | |
| "grad_norm": 14.764686584472656, | |
| "learning_rate": 9.013333333333334e-06, | |
| "loss": 0.705, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.4775380261761585, | |
| "grad_norm": 15.155953407287598, | |
| "learning_rate": 8.846666666666668e-06, | |
| "loss": 0.6948, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.4952246197382384, | |
| "grad_norm": 15.866467475891113, | |
| "learning_rate": 8.68e-06, | |
| "loss": 0.6712, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.5129112133003184, | |
| "grad_norm": 16.8061466217041, | |
| "learning_rate": 8.513333333333335e-06, | |
| "loss": 0.6726, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.5305978068623983, | |
| "grad_norm": 14.077337265014648, | |
| "learning_rate": 8.346666666666668e-06, | |
| "loss": 0.6666, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.5482844004244782, | |
| "grad_norm": 13.416068077087402, | |
| "learning_rate": 8.18e-06, | |
| "loss": 0.6808, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.5659709939865581, | |
| "grad_norm": 14.89547348022461, | |
| "learning_rate": 8.013333333333333e-06, | |
| "loss": 0.6492, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.5836575875486382, | |
| "grad_norm": 13.067695617675781, | |
| "learning_rate": 7.846666666666667e-06, | |
| "loss": 0.651, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.6013441811107181, | |
| "grad_norm": 14.657025337219238, | |
| "learning_rate": 7.680000000000001e-06, | |
| "loss": 0.6188, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.619030774672798, | |
| "grad_norm": 16.345060348510742, | |
| "learning_rate": 7.513333333333334e-06, | |
| "loss": 0.6519, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.636717368234878, | |
| "grad_norm": 17.713294982910156, | |
| "learning_rate": 7.346666666666668e-06, | |
| "loss": 0.604, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.6544039617969579, | |
| "grad_norm": 14.218528747558594, | |
| "learning_rate": 7.180000000000001e-06, | |
| "loss": 0.6272, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.6720905553590378, | |
| "grad_norm": 12.941822052001953, | |
| "learning_rate": 7.0133333333333345e-06, | |
| "loss": 0.6271, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.6897771489211177, | |
| "grad_norm": 15.65915298461914, | |
| "learning_rate": 6.846666666666667e-06, | |
| "loss": 0.6139, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.7074637424831978, | |
| "grad_norm": 15.025140762329102, | |
| "learning_rate": 6.680000000000001e-06, | |
| "loss": 0.6185, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7074637424831978, | |
| "eval_loss": 0.6446258425712585, | |
| "eval_runtime": 1231.4066, | |
| "eval_samples_per_second": 8.74, | |
| "eval_steps_per_second": 0.547, | |
| "eval_wer": 0.3809160603927091, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7251503360452777, | |
| "grad_norm": 15.671004295349121, | |
| "learning_rate": 6.513333333333333e-06, | |
| "loss": 0.5958, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.7428369296073576, | |
| "grad_norm": 14.718060493469238, | |
| "learning_rate": 6.346666666666668e-06, | |
| "loss": 0.5905, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.7605235231694376, | |
| "grad_norm": 14.305146217346191, | |
| "learning_rate": 6.18e-06, | |
| "loss": 0.6024, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.7782101167315175, | |
| "grad_norm": 14.678905487060547, | |
| "learning_rate": 6.013333333333335e-06, | |
| "loss": 0.6035, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.7958967102935974, | |
| "grad_norm": 13.787155151367188, | |
| "learning_rate": 5.846666666666667e-06, | |
| "loss": 0.5738, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.8135833038556773, | |
| "grad_norm": 13.068892478942871, | |
| "learning_rate": 5.68e-06, | |
| "loss": 0.5771, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.8312698974177574, | |
| "grad_norm": 14.032684326171875, | |
| "learning_rate": 5.513333333333334e-06, | |
| "loss": 0.5921, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.8489564909798373, | |
| "grad_norm": 13.227592468261719, | |
| "learning_rate": 5.346666666666667e-06, | |
| "loss": 0.5848, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.8666430845419172, | |
| "grad_norm": 15.441591262817383, | |
| "learning_rate": 5.18e-06, | |
| "loss": 0.5607, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.8843296781039972, | |
| "grad_norm": 13.670654296875, | |
| "learning_rate": 5.013333333333333e-06, | |
| "loss": 0.5968, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.9020162716660771, | |
| "grad_norm": 14.419194221496582, | |
| "learning_rate": 4.846666666666667e-06, | |
| "loss": 0.533, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.919702865228157, | |
| "grad_norm": 14.973267555236816, | |
| "learning_rate": 4.680000000000001e-06, | |
| "loss": 0.5584, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.937389458790237, | |
| "grad_norm": 13.328692436218262, | |
| "learning_rate": 4.513333333333333e-06, | |
| "loss": 0.5475, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.955076052352317, | |
| "grad_norm": 15.836326599121094, | |
| "learning_rate": 4.346666666666667e-06, | |
| "loss": 0.5422, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.9727626459143969, | |
| "grad_norm": 12.22904109954834, | |
| "learning_rate": 4.18e-06, | |
| "loss": 0.574, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.9904492394764768, | |
| "grad_norm": 16.11627769470215, | |
| "learning_rate": 4.013333333333334e-06, | |
| "loss": 0.5559, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.0081358330385568, | |
| "grad_norm": 12.527290344238281, | |
| "learning_rate": 3.8466666666666665e-06, | |
| "loss": 0.5176, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.0258224266006368, | |
| "grad_norm": 11.869111061096191, | |
| "learning_rate": 3.6800000000000003e-06, | |
| "loss": 0.5112, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.0435090201627166, | |
| "grad_norm": 13.095253944396973, | |
| "learning_rate": 3.5133333333333337e-06, | |
| "loss": 0.483, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 1.0611956137247966, | |
| "grad_norm": 11.897124290466309, | |
| "learning_rate": 3.346666666666667e-06, | |
| "loss": 0.4822, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0788822072868764, | |
| "grad_norm": 15.847147941589355, | |
| "learning_rate": 3.1800000000000005e-06, | |
| "loss": 0.4396, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 1.0965688008489565, | |
| "grad_norm": 12.42253303527832, | |
| "learning_rate": 3.013333333333334e-06, | |
| "loss": 0.4731, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.1142553944110365, | |
| "grad_norm": 13.371386528015137, | |
| "learning_rate": 2.8466666666666672e-06, | |
| "loss": 0.4623, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 1.1319419879731163, | |
| "grad_norm": 13.145184516906738, | |
| "learning_rate": 2.68e-06, | |
| "loss": 0.4499, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.1496285815351963, | |
| "grad_norm": 13.22415542602539, | |
| "learning_rate": 2.5133333333333336e-06, | |
| "loss": 0.4716, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 1.1673151750972763, | |
| "grad_norm": 10.703633308410645, | |
| "learning_rate": 2.346666666666667e-06, | |
| "loss": 0.4536, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.1850017686593561, | |
| "grad_norm": 15.410745620727539, | |
| "learning_rate": 2.1800000000000003e-06, | |
| "loss": 0.4616, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 1.2026883622214362, | |
| "grad_norm": 14.111817359924316, | |
| "learning_rate": 2.0133333333333337e-06, | |
| "loss": 0.4498, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.2203749557835162, | |
| "grad_norm": 12.579985618591309, | |
| "learning_rate": 1.8466666666666668e-06, | |
| "loss": 0.464, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 1.238061549345596, | |
| "grad_norm": 12.897194862365723, | |
| "learning_rate": 1.6800000000000002e-06, | |
| "loss": 0.4678, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.255748142907676, | |
| "grad_norm": 11.709689140319824, | |
| "learning_rate": 1.5133333333333334e-06, | |
| "loss": 0.4221, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 1.2734347364697558, | |
| "grad_norm": 14.162332534790039, | |
| "learning_rate": 1.3466666666666668e-06, | |
| "loss": 0.4625, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.2911213300318358, | |
| "grad_norm": 12.656081199645996, | |
| "learning_rate": 1.1800000000000001e-06, | |
| "loss": 0.4524, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 1.3088079235939158, | |
| "grad_norm": 12.56009292602539, | |
| "learning_rate": 1.0133333333333333e-06, | |
| "loss": 0.4386, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.3264945171559956, | |
| "grad_norm": 11.350845336914062, | |
| "learning_rate": 8.466666666666668e-07, | |
| "loss": 0.4516, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 1.3441811107180757, | |
| "grad_norm": 11.870651245117188, | |
| "learning_rate": 6.800000000000001e-07, | |
| "loss": 0.4226, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.3618677042801557, | |
| "grad_norm": 11.236252784729004, | |
| "learning_rate": 5.133333333333334e-07, | |
| "loss": 0.442, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 1.3795542978422355, | |
| "grad_norm": 12.496944427490234, | |
| "learning_rate": 3.466666666666667e-07, | |
| "loss": 0.4464, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.3972408914043155, | |
| "grad_norm": 11.83437728881836, | |
| "learning_rate": 1.8e-07, | |
| "loss": 0.4552, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 1.4149274849663955, | |
| "grad_norm": 12.044703483581543, | |
| "learning_rate": 1.3333333333333334e-08, | |
| "loss": 0.4497, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.4149274849663955, | |
| "eval_loss": 0.5812480449676514, | |
| "eval_runtime": 1144.5779, | |
| "eval_samples_per_second": 9.403, | |
| "eval_steps_per_second": 0.588, | |
| "eval_wer": 0.3526167566298991, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.4149274849663955, | |
| "step": 2000, | |
| "total_flos": 4.15045601427456e+18, | |
| "train_loss": 0.6969459619522095, | |
| "train_runtime": 9919.8131, | |
| "train_samples_per_second": 6.452, | |
| "train_steps_per_second": 0.202 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.15045601427456e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |