| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.945945945945946, | |
| "eval_steps": 500, | |
| "global_step": 18000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.021621621621621623, | |
| "grad_norm": 43.9290657043457, | |
| "learning_rate": 1.9856576576576577e-05, | |
| "loss": 0.6499, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.043243243243243246, | |
| "grad_norm": 13.308595657348633, | |
| "learning_rate": 1.9712432432432433e-05, | |
| "loss": 0.568, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.06486486486486487, | |
| "grad_norm": 5.098663806915283, | |
| "learning_rate": 1.956828828828829e-05, | |
| "loss": 0.5371, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.08648648648648649, | |
| "grad_norm": 9.126262664794922, | |
| "learning_rate": 1.9424144144144147e-05, | |
| "loss": 0.5181, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.10810810810810811, | |
| "grad_norm": 15.761275291442871, | |
| "learning_rate": 1.9280000000000002e-05, | |
| "loss": 0.504, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12972972972972974, | |
| "grad_norm": 33.255863189697266, | |
| "learning_rate": 1.9135855855855857e-05, | |
| "loss": 0.5017, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.15135135135135136, | |
| "grad_norm": 6.04021692276001, | |
| "learning_rate": 1.8991711711711712e-05, | |
| "loss": 0.468, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.17297297297297298, | |
| "grad_norm": 15.048690795898438, | |
| "learning_rate": 1.8847567567567568e-05, | |
| "loss": 0.5056, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.1945945945945946, | |
| "grad_norm": 22.5860538482666, | |
| "learning_rate": 1.8703423423423426e-05, | |
| "loss": 0.4601, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.21621621621621623, | |
| "grad_norm": 12.32684326171875, | |
| "learning_rate": 1.855927927927928e-05, | |
| "loss": 0.4401, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23783783783783785, | |
| "grad_norm": 9.132744789123535, | |
| "learning_rate": 1.8415135135135137e-05, | |
| "loss": 0.4707, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.2594594594594595, | |
| "grad_norm": 10.921464920043945, | |
| "learning_rate": 1.8270990990990992e-05, | |
| "loss": 0.4676, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.2810810810810811, | |
| "grad_norm": 20.87602424621582, | |
| "learning_rate": 1.8126846846846847e-05, | |
| "loss": 0.469, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.3027027027027027, | |
| "grad_norm": 27.636375427246094, | |
| "learning_rate": 1.7982702702702702e-05, | |
| "loss": 0.4419, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.32432432432432434, | |
| "grad_norm": 8.047364234924316, | |
| "learning_rate": 1.783855855855856e-05, | |
| "loss": 0.4331, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.34594594594594597, | |
| "grad_norm": 14.144438743591309, | |
| "learning_rate": 1.7694414414414416e-05, | |
| "loss": 0.4444, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.3675675675675676, | |
| "grad_norm": 13.833978652954102, | |
| "learning_rate": 1.755027027027027e-05, | |
| "loss": 0.4269, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.3891891891891892, | |
| "grad_norm": 7.048819541931152, | |
| "learning_rate": 1.7406126126126127e-05, | |
| "loss": 0.4518, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.41081081081081083, | |
| "grad_norm": 16.350954055786133, | |
| "learning_rate": 1.7261981981981982e-05, | |
| "loss": 0.4306, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.43243243243243246, | |
| "grad_norm": 36.817256927490234, | |
| "learning_rate": 1.7117837837837837e-05, | |
| "loss": 0.3993, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4540540540540541, | |
| "grad_norm": 6.8075151443481445, | |
| "learning_rate": 1.6973693693693696e-05, | |
| "loss": 0.4624, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.4756756756756757, | |
| "grad_norm": 22.440227508544922, | |
| "learning_rate": 1.682954954954955e-05, | |
| "loss": 0.4062, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.4972972972972973, | |
| "grad_norm": 27.693946838378906, | |
| "learning_rate": 1.6685405405405406e-05, | |
| "loss": 0.3986, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.518918918918919, | |
| "grad_norm": 5.609838485717773, | |
| "learning_rate": 1.6541261261261262e-05, | |
| "loss": 0.418, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 9.340408325195312, | |
| "learning_rate": 1.6397117117117117e-05, | |
| "loss": 0.4292, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5621621621621622, | |
| "grad_norm": 17.670503616333008, | |
| "learning_rate": 1.6252972972972972e-05, | |
| "loss": 0.4358, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.5837837837837838, | |
| "grad_norm": 22.615758895874023, | |
| "learning_rate": 1.610882882882883e-05, | |
| "loss": 0.4324, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.6054054054054054, | |
| "grad_norm": 25.58919906616211, | |
| "learning_rate": 1.5964684684684686e-05, | |
| "loss": 0.4338, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.6270270270270271, | |
| "grad_norm": 32.85356140136719, | |
| "learning_rate": 1.582054054054054e-05, | |
| "loss": 0.4226, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.6486486486486487, | |
| "grad_norm": 9.336725234985352, | |
| "learning_rate": 1.56763963963964e-05, | |
| "loss": 0.4391, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6702702702702703, | |
| "grad_norm": 24.154613494873047, | |
| "learning_rate": 1.5532252252252252e-05, | |
| "loss": 0.3986, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.6918918918918919, | |
| "grad_norm": 18.917322158813477, | |
| "learning_rate": 1.5388108108108107e-05, | |
| "loss": 0.4107, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.7135135135135136, | |
| "grad_norm": 24.948320388793945, | |
| "learning_rate": 1.5243963963963966e-05, | |
| "loss": 0.4165, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.7351351351351352, | |
| "grad_norm": 14.24731731414795, | |
| "learning_rate": 1.5099819819819821e-05, | |
| "loss": 0.4294, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.7567567567567568, | |
| "grad_norm": 10.69514274597168, | |
| "learning_rate": 1.4955675675675676e-05, | |
| "loss": 0.4004, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7783783783783784, | |
| "grad_norm": 5.057804107666016, | |
| "learning_rate": 1.4811531531531533e-05, | |
| "loss": 0.4235, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 19.741506576538086, | |
| "learning_rate": 1.4667387387387388e-05, | |
| "loss": 0.3834, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.8216216216216217, | |
| "grad_norm": 15.608851432800293, | |
| "learning_rate": 1.4523243243243244e-05, | |
| "loss": 0.4089, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.8432432432432433, | |
| "grad_norm": 8.335741996765137, | |
| "learning_rate": 1.43790990990991e-05, | |
| "loss": 0.4025, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.8648648648648649, | |
| "grad_norm": 8.1391019821167, | |
| "learning_rate": 1.4234954954954956e-05, | |
| "loss": 0.444, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8864864864864865, | |
| "grad_norm": 28.06201934814453, | |
| "learning_rate": 1.4090810810810811e-05, | |
| "loss": 0.4113, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.9081081081081082, | |
| "grad_norm": 5.816445827484131, | |
| "learning_rate": 1.3946666666666668e-05, | |
| "loss": 0.4234, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.9297297297297298, | |
| "grad_norm": 22.189376831054688, | |
| "learning_rate": 1.3802522522522523e-05, | |
| "loss": 0.4147, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.9513513513513514, | |
| "grad_norm": 9.486166954040527, | |
| "learning_rate": 1.3658378378378379e-05, | |
| "loss": 0.4047, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.972972972972973, | |
| "grad_norm": 12.465399742126465, | |
| "learning_rate": 1.3514234234234236e-05, | |
| "loss": 0.4234, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9945945945945946, | |
| "grad_norm": 18.71035385131836, | |
| "learning_rate": 1.337009009009009e-05, | |
| "loss": 0.4202, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 1.0162162162162163, | |
| "grad_norm": 18.901094436645508, | |
| "learning_rate": 1.3225945945945946e-05, | |
| "loss": 0.3622, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 1.037837837837838, | |
| "grad_norm": 15.318452835083008, | |
| "learning_rate": 1.3081801801801803e-05, | |
| "loss": 0.3792, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 1.0594594594594595, | |
| "grad_norm": 15.37670612335205, | |
| "learning_rate": 1.2937657657657658e-05, | |
| "loss": 0.3622, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 16.18653106689453, | |
| "learning_rate": 1.2793513513513514e-05, | |
| "loss": 0.3763, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.1027027027027028, | |
| "grad_norm": 40.86188507080078, | |
| "learning_rate": 1.264936936936937e-05, | |
| "loss": 0.3574, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 1.1243243243243244, | |
| "grad_norm": 7.759640693664551, | |
| "learning_rate": 1.2505225225225226e-05, | |
| "loss": 0.3878, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 1.145945945945946, | |
| "grad_norm": 7.456469535827637, | |
| "learning_rate": 1.2361081081081081e-05, | |
| "loss": 0.3968, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 1.1675675675675676, | |
| "grad_norm": 18.799278259277344, | |
| "learning_rate": 1.2216936936936938e-05, | |
| "loss": 0.3725, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 1.1891891891891893, | |
| "grad_norm": 2.651585817337036, | |
| "learning_rate": 1.2072792792792793e-05, | |
| "loss": 0.3766, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.2108108108108109, | |
| "grad_norm": 11.979673385620117, | |
| "learning_rate": 1.1928648648648648e-05, | |
| "loss": 0.3632, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 1.2324324324324325, | |
| "grad_norm": 4.676784515380859, | |
| "learning_rate": 1.1784504504504505e-05, | |
| "loss": 0.3961, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 1.2540540540540541, | |
| "grad_norm": 9.577638626098633, | |
| "learning_rate": 1.164036036036036e-05, | |
| "loss": 0.3602, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 1.2756756756756757, | |
| "grad_norm": 9.80462646484375, | |
| "learning_rate": 1.1496216216216216e-05, | |
| "loss": 0.3479, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 1.2972972972972974, | |
| "grad_norm": 11.522858619689941, | |
| "learning_rate": 1.1352072072072073e-05, | |
| "loss": 0.407, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.318918918918919, | |
| "grad_norm": 3.6061971187591553, | |
| "learning_rate": 1.1207927927927928e-05, | |
| "loss": 0.3918, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 1.3405405405405406, | |
| "grad_norm": 15.432760238647461, | |
| "learning_rate": 1.1063783783783783e-05, | |
| "loss": 0.3726, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 1.3621621621621622, | |
| "grad_norm": 5.143666744232178, | |
| "learning_rate": 1.091963963963964e-05, | |
| "loss": 0.3504, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 1.3837837837837839, | |
| "grad_norm": 8.900772094726562, | |
| "learning_rate": 1.0775495495495495e-05, | |
| "loss": 0.357, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 1.4054054054054055, | |
| "grad_norm": 14.410099983215332, | |
| "learning_rate": 1.0631351351351354e-05, | |
| "loss": 0.3608, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.427027027027027, | |
| "grad_norm": 13.041916847229004, | |
| "learning_rate": 1.0487207207207208e-05, | |
| "loss": 0.3684, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 1.4486486486486487, | |
| "grad_norm": 16.890735626220703, | |
| "learning_rate": 1.0343063063063063e-05, | |
| "loss": 0.3835, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 1.4702702702702704, | |
| "grad_norm": 7.087887763977051, | |
| "learning_rate": 1.0198918918918922e-05, | |
| "loss": 0.4041, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 1.491891891891892, | |
| "grad_norm": 7.32821798324585, | |
| "learning_rate": 1.0054774774774775e-05, | |
| "loss": 0.3563, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 1.5135135135135136, | |
| "grad_norm": 22.392210006713867, | |
| "learning_rate": 9.91063063063063e-06, | |
| "loss": 0.3807, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.535135135135135, | |
| "grad_norm": 14.552515029907227, | |
| "learning_rate": 9.766486486486487e-06, | |
| "loss": 0.3539, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 1.5567567567567568, | |
| "grad_norm": 9.686298370361328, | |
| "learning_rate": 9.622342342342343e-06, | |
| "loss": 0.3971, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 1.5783783783783782, | |
| "grad_norm": 13.134624481201172, | |
| "learning_rate": 9.478198198198198e-06, | |
| "loss": 0.3957, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 35.519474029541016, | |
| "learning_rate": 9.334054054054055e-06, | |
| "loss": 0.3655, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 14.94369125366211, | |
| "learning_rate": 9.189909909909912e-06, | |
| "loss": 0.3503, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.6432432432432433, | |
| "grad_norm": 3.6573474407196045, | |
| "learning_rate": 9.045765765765765e-06, | |
| "loss": 0.3883, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 1.6648648648648647, | |
| "grad_norm": 18.653778076171875, | |
| "learning_rate": 8.901621621621622e-06, | |
| "loss": 0.3425, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 1.6864864864864866, | |
| "grad_norm": 17.726455688476562, | |
| "learning_rate": 8.75747747747748e-06, | |
| "loss": 0.3816, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 1.708108108108108, | |
| "grad_norm": 15.91839599609375, | |
| "learning_rate": 8.613333333333333e-06, | |
| "loss": 0.3573, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 1.7297297297297298, | |
| "grad_norm": 20.901330947875977, | |
| "learning_rate": 8.46918918918919e-06, | |
| "loss": 0.3515, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.7513513513513512, | |
| "grad_norm": 12.939945220947266, | |
| "learning_rate": 8.325045045045047e-06, | |
| "loss": 0.3559, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 1.772972972972973, | |
| "grad_norm": 9.921557426452637, | |
| "learning_rate": 8.180900900900902e-06, | |
| "loss": 0.3844, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 1.7945945945945945, | |
| "grad_norm": 5.254030704498291, | |
| "learning_rate": 8.036756756756757e-06, | |
| "loss": 0.3567, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 1.8162162162162163, | |
| "grad_norm": 29.623857498168945, | |
| "learning_rate": 7.892612612612614e-06, | |
| "loss": 0.3364, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 1.8378378378378377, | |
| "grad_norm": 8.430907249450684, | |
| "learning_rate": 7.74846846846847e-06, | |
| "loss": 0.3742, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.8594594594594596, | |
| "grad_norm": 14.618759155273438, | |
| "learning_rate": 7.604324324324325e-06, | |
| "loss": 0.3624, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 1.881081081081081, | |
| "grad_norm": 20.811887741088867, | |
| "learning_rate": 7.460180180180181e-06, | |
| "loss": 0.3585, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 1.9027027027027028, | |
| "grad_norm": 29.710891723632812, | |
| "learning_rate": 7.316036036036037e-06, | |
| "loss": 0.3648, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 1.9243243243243242, | |
| "grad_norm": 7.511297702789307, | |
| "learning_rate": 7.171891891891892e-06, | |
| "loss": 0.364, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 1.945945945945946, | |
| "grad_norm": 5.749581813812256, | |
| "learning_rate": 7.027747747747748e-06, | |
| "loss": 0.3467, | |
| "step": 18000 | |
| } | |
| ], | |
| "logging_steps": 200, | |
| "max_steps": 27750, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.014716980572979e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |