| { | |
| "best_metric": 3.1372523307800293, | |
| "best_model_checkpoint": "xtreme_s_xlsr_mls_upd/checkpoint-100", | |
| "epoch": 3.0, | |
| "global_step": 102, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.0, | |
| "loss": 10.1653, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.0003, | |
| "loss": 9.9916, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.000297029702970297, | |
| "loss": 9.9785, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.000294059405940594, | |
| "loss": 9.1255, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.000294059405940594, | |
| "loss": 7.897, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00029108910891089107, | |
| "loss": 7.7678, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00028811881188118806, | |
| "loss": 6.6699, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 0.0002851485148514851, | |
| "loss": 5.6942, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00028217821782178216, | |
| "loss": 5.0037, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002792079207920792, | |
| "loss": 4.5539, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0002762376237623762, | |
| "loss": 4.3802, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00027326732673267325, | |
| "loss": 4.0771, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0002702970297029703, | |
| "loss": 3.9155, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002673267326732673, | |
| "loss": 3.8023, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00026435643564356434, | |
| "loss": 3.7282, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00026138613861386134, | |
| "loss": 3.7088, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0002584158415841584, | |
| "loss": 3.5787, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00025544554455445543, | |
| "loss": 3.5437, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00025247524752475243, | |
| "loss": 3.495, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.0002495049504950495, | |
| "loss": 3.4678, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_cer": 1.0, | |
| "eval_loss": 3.458103895187378, | |
| "eval_runtime": 36.7504, | |
| "eval_samples_per_second": 13.932, | |
| "eval_steps_per_second": 0.871, | |
| "eval_wer": 1.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0002465346534653465, | |
| "loss": 3.4143, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00024356435643564355, | |
| "loss": 3.4058, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 0.0002405940594059406, | |
| "loss": 3.3713, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00023762376237623762, | |
| "loss": 3.3317, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00023465346534653464, | |
| "loss": 3.2929, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00023168316831683166, | |
| "loss": 3.2983, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00022871287128712868, | |
| "loss": 3.2625, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00022574257425742573, | |
| "loss": 3.3042, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00022277227722772275, | |
| "loss": 3.2494, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00021980198019801977, | |
| "loss": 3.231, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.0002168316831683168, | |
| "loss": 3.2249, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00021386138613861385, | |
| "loss": 3.2355, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00021089108910891087, | |
| "loss": 3.1952, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0002079207920792079, | |
| "loss": 3.1892, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00020495049504950494, | |
| "loss": 3.1875, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00020198019801980199, | |
| "loss": 3.1733, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.000199009900990099, | |
| "loss": 3.2136, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00019603960396039603, | |
| "loss": 3.1819, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.00019306930693069305, | |
| "loss": 3.1764, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00019009900990099007, | |
| "loss": 3.1713, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_cer": 1.0, | |
| "eval_loss": 3.1816344261169434, | |
| "eval_runtime": 36.1118, | |
| "eval_samples_per_second": 14.178, | |
| "eval_steps_per_second": 0.886, | |
| "eval_wer": 1.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.00018712871287128712, | |
| "loss": 3.176, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00018415841584158414, | |
| "loss": 3.1669, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.00018118811881188116, | |
| "loss": 3.1446, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00017821782178217819, | |
| "loss": 3.1556, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.0001752475247524752, | |
| "loss": 3.1447, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00017227722772277226, | |
| "loss": 3.1412, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00016930693069306928, | |
| "loss": 3.1521, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.00016633663366336633, | |
| "loss": 3.1493, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00016336633663366338, | |
| "loss": 3.1494, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.0001603960396039604, | |
| "loss": 3.1419, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00015742574257425742, | |
| "loss": 3.1407, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00015445544554455444, | |
| "loss": 3.1456, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00015148514851485146, | |
| "loss": 3.1317, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.0001485148514851485, | |
| "loss": 3.1369, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00014554455445544553, | |
| "loss": 3.1547, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.00014257425742574255, | |
| "loss": 3.1356, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.0001396039603960396, | |
| "loss": 3.1538, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.00013663366336633662, | |
| "loss": 3.1549, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 0.00013366336633663365, | |
| "loss": 3.1275, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00013069306930693067, | |
| "loss": 3.134, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_cer": 1.0, | |
| "eval_loss": 3.153820276260376, | |
| "eval_runtime": 36.2824, | |
| "eval_samples_per_second": 14.112, | |
| "eval_steps_per_second": 0.882, | |
| "eval_wer": 1.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00012772277227722772, | |
| "loss": 3.1667, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00012475247524752474, | |
| "loss": 3.1346, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.00012178217821782177, | |
| "loss": 3.1247, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00011881188118811881, | |
| "loss": 3.1358, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.00011584158415841583, | |
| "loss": 3.1382, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00011287128712871287, | |
| "loss": 3.1455, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.00010990099009900989, | |
| "loss": 3.1397, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00010693069306930692, | |
| "loss": 3.1402, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.00010396039603960394, | |
| "loss": 3.1163, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.00010099009900990099, | |
| "loss": 3.1175, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.801980198019801e-05, | |
| "loss": 3.1169, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 9.504950495049504e-05, | |
| "loss": 3.1242, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 9.207920792079207e-05, | |
| "loss": 3.1326, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 8.910891089108909e-05, | |
| "loss": 3.1394, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 8.613861386138613e-05, | |
| "loss": 3.1369, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 8.316831683168316e-05, | |
| "loss": 3.1284, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 8.01980198019802e-05, | |
| "loss": 3.1203, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 7.722772277227722e-05, | |
| "loss": 3.1245, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 7.425742574257426e-05, | |
| "loss": 3.1243, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 7.128712871287128e-05, | |
| "loss": 3.132, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_cer": 1.0, | |
| "eval_loss": 3.1410903930664062, | |
| "eval_runtime": 36.0623, | |
| "eval_samples_per_second": 14.198, | |
| "eval_steps_per_second": 0.887, | |
| "eval_wer": 1.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 6.831683168316831e-05, | |
| "loss": 3.1249, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 6.534653465346533e-05, | |
| "loss": 3.1271, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 6.237623762376237e-05, | |
| "loss": 3.1245, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 5.9405940594059404e-05, | |
| "loss": 3.1242, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 5.643564356435643e-05, | |
| "loss": 3.126, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 5.346534653465346e-05, | |
| "loss": 3.1327, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 5.0495049504950497e-05, | |
| "loss": 3.1214, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 4.752475247524752e-05, | |
| "loss": 3.1513, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 4.4554455445544547e-05, | |
| "loss": 3.128, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 4.158415841584158e-05, | |
| "loss": 3.1243, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 3.861386138613861e-05, | |
| "loss": 3.1342, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 3.564356435643564e-05, | |
| "loss": 3.1474, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 3.267326732673267e-05, | |
| "loss": 3.1259, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 2.9702970297029702e-05, | |
| "loss": 3.1205, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 2.673267326732673e-05, | |
| "loss": 3.1273, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 2.376237623762376e-05, | |
| "loss": 3.1177, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 2.079207920792079e-05, | |
| "loss": 3.1179, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 1.782178217821782e-05, | |
| "loss": 3.1272, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 1.4851485148514851e-05, | |
| "loss": 3.1404, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1.188118811881188e-05, | |
| "loss": 3.1295, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_cer": 1.0, | |
| "eval_loss": 3.1372523307800293, | |
| "eval_runtime": 36.088, | |
| "eval_samples_per_second": 14.188, | |
| "eval_steps_per_second": 0.887, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 8.91089108910891e-06, | |
| "loss": 3.1441, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 5.94059405940594e-06, | |
| "loss": 3.1193, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 102, | |
| "total_flos": 3.102653790168285e+18, | |
| "train_loss": 3.6673195455588545, | |
| "train_runtime": 1011.7717, | |
| "train_samples_per_second": 6.443, | |
| "train_steps_per_second": 0.101 | |
| } | |
| ], | |
| "max_steps": 102, | |
| "num_train_epochs": 3, | |
| "total_flos": 3.102653790168285e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |