| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.354838709677416, | |
| "global_step": 18400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 7.1936, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 3.278684139251709, | |
| "eval_runtime": 123.6406, | |
| "eval_samples_per_second": 12.31, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0002950276243093923, | |
| "loss": 3.1985, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_loss": 2.8315000534057617, | |
| "eval_runtime": 121.9578, | |
| "eval_samples_per_second": 12.48, | |
| "eval_wer": 0.9966690219037044, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.00028839779005524857, | |
| "loss": 1.1566, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "eval_loss": 0.7793326377868652, | |
| "eval_runtime": 113.9019, | |
| "eval_samples_per_second": 13.362, | |
| "eval_wer": 0.7725850408801858, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.00028176795580110497, | |
| "loss": 0.5297, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "eval_loss": 0.6474416255950928, | |
| "eval_runtime": 122.7887, | |
| "eval_samples_per_second": 12.395, | |
| "eval_wer": 0.6751791662460886, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 0.0002751381215469613, | |
| "loss": 0.3725, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_loss": 0.5674485564231873, | |
| "eval_runtime": 114.3739, | |
| "eval_samples_per_second": 13.307, | |
| "eval_wer": 0.6073483395578884, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 0.00026850828729281767, | |
| "loss": 0.2877, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "eval_loss": 0.5630490183830261, | |
| "eval_runtime": 114.404, | |
| "eval_samples_per_second": 13.304, | |
| "eval_wer": 0.5621277884324215, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 0.000261878453038674, | |
| "loss": 0.2342, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "eval_loss": 0.5479915142059326, | |
| "eval_runtime": 123.0266, | |
| "eval_samples_per_second": 12.371, | |
| "eval_wer": 0.5403250227112143, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 0.00025524861878453036, | |
| "loss": 0.1967, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "eval_loss": 0.6276335120201111, | |
| "eval_runtime": 114.8683, | |
| "eval_samples_per_second": 13.25, | |
| "eval_wer": 0.5577874230342182, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 0.0002486187845303867, | |
| "loss": 0.1729, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "eval_loss": 0.536430835723877, | |
| "eval_runtime": 114.6009, | |
| "eval_samples_per_second": 13.281, | |
| "eval_wer": 0.5332593115978601, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 0.00024198895027624309, | |
| "loss": 0.1607, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_loss": 0.5107040405273438, | |
| "eval_runtime": 115.3852, | |
| "eval_samples_per_second": 13.191, | |
| "eval_wer": 0.495306349046129, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 0.00023535911602209943, | |
| "loss": 0.1381, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "eval_loss": 0.5929794907569885, | |
| "eval_runtime": 123.4194, | |
| "eval_samples_per_second": 12.332, | |
| "eval_wer": 0.5220551125466841, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 0.0002287292817679558, | |
| "loss": 0.1261, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "eval_loss": 0.6632552742958069, | |
| "eval_runtime": 124.1096, | |
| "eval_samples_per_second": 12.263, | |
| "eval_wer": 0.5305339658827092, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 0.00022209944751381213, | |
| "loss": 0.1216, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "eval_loss": 0.6114311218261719, | |
| "eval_runtime": 116.8858, | |
| "eval_samples_per_second": 13.021, | |
| "eval_wer": 0.5084283839709296, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 0.0002154696132596685, | |
| "loss": 0.1119, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "eval_loss": 0.6091165542602539, | |
| "eval_runtime": 123.4244, | |
| "eval_samples_per_second": 12.331, | |
| "eval_wer": 0.5041889573029171, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 0.00020883977900552485, | |
| "loss": 0.107, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "eval_loss": 0.5811592936515808, | |
| "eval_runtime": 131.8655, | |
| "eval_samples_per_second": 11.542, | |
| "eval_wer": 0.49581104269708287, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 0.00020220994475138123, | |
| "loss": 0.0985, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "eval_loss": 0.6538776755332947, | |
| "eval_runtime": 132.6579, | |
| "eval_samples_per_second": 11.473, | |
| "eval_wer": 0.5082265065105481, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 0.00019558011049723755, | |
| "loss": 0.0959, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_loss": 0.602001965045929, | |
| "eval_runtime": 132.2249, | |
| "eval_samples_per_second": 11.511, | |
| "eval_wer": 0.4906631674573534, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 0.0001889502762430939, | |
| "loss": 0.0899, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "eval_loss": 0.6006932854652405, | |
| "eval_runtime": 115.0032, | |
| "eval_samples_per_second": 13.234, | |
| "eval_wer": 0.48813969920258404, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 0.00018232044198895027, | |
| "loss": 0.0867, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "eval_loss": 0.5921047329902649, | |
| "eval_runtime": 115.0629, | |
| "eval_samples_per_second": 13.228, | |
| "eval_wer": 0.48763500555163014, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 0.0001756906077348066, | |
| "loss": 0.0825, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "eval_loss": 0.6151732802391052, | |
| "eval_runtime": 115.0579, | |
| "eval_samples_per_second": 13.228, | |
| "eval_wer": 0.4937922680932674, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 0.00016906077348066297, | |
| "loss": 0.0768, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "eval_loss": 0.6089133024215698, | |
| "eval_runtime": 123.8125, | |
| "eval_samples_per_second": 12.293, | |
| "eval_wer": 0.47723831634198044, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 0.00016243093922651931, | |
| "loss": 0.0698, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "eval_loss": 0.6180127263069153, | |
| "eval_runtime": 123.6525, | |
| "eval_samples_per_second": 12.309, | |
| "eval_wer": 0.47330170586454023, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 0.0001558011049723757, | |
| "loss": 0.0753, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "eval_loss": 0.621383547782898, | |
| "eval_runtime": 115.3106, | |
| "eval_samples_per_second": 13.199, | |
| "eval_wer": 0.4866256182497224, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 0.00014917127071823204, | |
| "loss": 0.0674, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_loss": 0.6383547782897949, | |
| "eval_runtime": 123.6323, | |
| "eval_samples_per_second": 12.311, | |
| "eval_wer": 0.4813768042798022, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 0.00014254143646408839, | |
| "loss": 0.0617, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "eval_loss": 0.6684080362319946, | |
| "eval_runtime": 122.6948, | |
| "eval_samples_per_second": 12.405, | |
| "eval_wer": 0.47673362269102654, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "learning_rate": 0.00013591160220994473, | |
| "loss": 0.059, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "eval_loss": 0.6545931696891785, | |
| "eval_runtime": 115.3877, | |
| "eval_samples_per_second": 13.19, | |
| "eval_wer": 0.46734632078328453, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 0.0001292817679558011, | |
| "loss": 0.0598, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "eval_loss": 0.6368861794471741, | |
| "eval_runtime": 115.1488, | |
| "eval_samples_per_second": 13.218, | |
| "eval_wer": 0.4668416271323307, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 0.00012265193370165746, | |
| "loss": 0.0558, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "eval_loss": 0.6463531851768494, | |
| "eval_runtime": 115.1323, | |
| "eval_samples_per_second": 13.22, | |
| "eval_wer": 0.4656303623700414, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 37.42, | |
| "learning_rate": 0.0001160220994475138, | |
| "loss": 0.0568, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 37.42, | |
| "eval_loss": 0.6061355471611023, | |
| "eval_runtime": 115.4845, | |
| "eval_samples_per_second": 13.179, | |
| "eval_wer": 0.46966791157767235, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "learning_rate": 0.00010939226519337017, | |
| "loss": 0.05, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "eval_loss": 0.6069867014884949, | |
| "eval_runtime": 115.3997, | |
| "eval_samples_per_second": 13.189, | |
| "eval_wer": 0.46381346522660744, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 0.00010276243093922651, | |
| "loss": 0.0489, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.6435591578483582, | |
| "eval_runtime": 115.2889, | |
| "eval_samples_per_second": 13.202, | |
| "eval_wer": 0.45493085696981933, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 41.29, | |
| "learning_rate": 9.613259668508287e-05, | |
| "loss": 0.0508, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 41.29, | |
| "eval_loss": 0.6377198696136475, | |
| "eval_runtime": 115.7131, | |
| "eval_samples_per_second": 13.153, | |
| "eval_wer": 0.45493085696981933, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 42.58, | |
| "learning_rate": 8.950276243093922e-05, | |
| "loss": 0.0475, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 42.58, | |
| "eval_loss": 0.6290169954299927, | |
| "eval_runtime": 115.5463, | |
| "eval_samples_per_second": 13.172, | |
| "eval_wer": 0.45836277379630563, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 43.87, | |
| "learning_rate": 8.287292817679558e-05, | |
| "loss": 0.0415, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 43.87, | |
| "eval_loss": 0.6694880723953247, | |
| "eval_runtime": 115.6201, | |
| "eval_samples_per_second": 13.164, | |
| "eval_wer": 0.45291208236600383, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "learning_rate": 7.624309392265193e-05, | |
| "loss": 0.0437, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 45.16, | |
| "eval_loss": 0.5929790139198303, | |
| "eval_runtime": 115.6043, | |
| "eval_samples_per_second": 13.166, | |
| "eval_wer": 0.44594730998284043, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 46.45, | |
| "learning_rate": 6.961325966850828e-05, | |
| "loss": 0.0405, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 46.45, | |
| "eval_loss": 0.5978022813796997, | |
| "eval_runtime": 115.5674, | |
| "eval_samples_per_second": 13.17, | |
| "eval_wer": 0.44857171696780057, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 47.74, | |
| "learning_rate": 6.298342541436463e-05, | |
| "loss": 0.0373, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 47.74, | |
| "eval_loss": 0.6167137622833252, | |
| "eval_runtime": 129.2696, | |
| "eval_samples_per_second": 11.774, | |
| "eval_wer": 0.4487735944281821, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 49.03, | |
| "learning_rate": 5.635359116022099e-05, | |
| "loss": 0.0351, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 49.03, | |
| "eval_loss": 0.6160412430763245, | |
| "eval_runtime": 132.6569, | |
| "eval_samples_per_second": 11.473, | |
| "eval_wer": 0.4417078833148279, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 50.32, | |
| "learning_rate": 4.9723756906077343e-05, | |
| "loss": 0.0373, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 50.32, | |
| "eval_loss": 0.5868551731109619, | |
| "eval_runtime": 116.8754, | |
| "eval_samples_per_second": 13.022, | |
| "eval_wer": 0.4385787826789139, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 51.61, | |
| "learning_rate": 4.30939226519337e-05, | |
| "loss": 0.0346, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 51.61, | |
| "eval_loss": 0.6198846101760864, | |
| "eval_runtime": 124.1832, | |
| "eval_samples_per_second": 12.256, | |
| "eval_wer": 0.4357524982335722, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 52.9, | |
| "learning_rate": 3.646408839779005e-05, | |
| "loss": 0.0328, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 52.9, | |
| "eval_loss": 0.6003885269165039, | |
| "eval_runtime": 124.3773, | |
| "eval_samples_per_second": 12.237, | |
| "eval_wer": 0.43999192490158473, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 54.19, | |
| "learning_rate": 2.9834254143646404e-05, | |
| "loss": 0.0315, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 54.19, | |
| "eval_loss": 0.6082204580307007, | |
| "eval_runtime": 124.1939, | |
| "eval_samples_per_second": 12.255, | |
| "eval_wer": 0.4374684566468154, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 55.48, | |
| "learning_rate": 2.320441988950276e-05, | |
| "loss": 0.0312, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 55.48, | |
| "eval_loss": 0.6157354116439819, | |
| "eval_runtime": 124.1732, | |
| "eval_samples_per_second": 12.257, | |
| "eval_wer": 0.4346421722014737, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 56.77, | |
| "learning_rate": 1.6574585635359113e-05, | |
| "loss": 0.031, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 56.77, | |
| "eval_loss": 0.6112752556800842, | |
| "eval_runtime": 125.7635, | |
| "eval_samples_per_second": 12.102, | |
| "eval_wer": 0.4303018068032704, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "learning_rate": 9.944751381215468e-06, | |
| "loss": 0.0296, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 58.06, | |
| "eval_loss": 0.6031031608581543, | |
| "eval_runtime": 125.68, | |
| "eval_samples_per_second": 12.11, | |
| "eval_wer": 0.4288886645805996, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 59.35, | |
| "learning_rate": 3.3149171270718227e-06, | |
| "loss": 0.0287, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 59.35, | |
| "eval_loss": 0.6061920523643494, | |
| "eval_runtime": 124.5102, | |
| "eval_samples_per_second": 12.224, | |
| "eval_wer": 0.428686787120218, | |
| "step": 18400 | |
| } | |
| ], | |
| "max_steps": 18600, | |
| "num_train_epochs": 60, | |
| "total_flos": 2.8159583015375258e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |