{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.0,
"eval_steps": 28,
"global_step": 140,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02857142857142857,
"grad_norm": 0.9386152625083923,
"learning_rate": 0.0,
"loss": 1.744140625,
"step": 1
},
{
"epoch": 0.05714285714285714,
"grad_norm": 0.852827250957489,
"learning_rate": 2e-05,
"loss": 1.6787109375,
"step": 2
},
{
"epoch": 0.08571428571428572,
"grad_norm": 0.8461427092552185,
"learning_rate": 4e-05,
"loss": 1.671875,
"step": 3
},
{
"epoch": 0.11428571428571428,
"grad_norm": 0.6573083400726318,
"learning_rate": 6e-05,
"loss": 1.5703125,
"step": 4
},
{
"epoch": 0.14285714285714285,
"grad_norm": 0.5101009011268616,
"learning_rate": 8e-05,
"loss": 1.5732421875,
"step": 5
},
{
"epoch": 0.17142857142857143,
"grad_norm": 0.3327346444129944,
"learning_rate": 0.0001,
"loss": 1.5615234375,
"step": 6
},
{
"epoch": 0.2,
"grad_norm": 0.5101948380470276,
"learning_rate": 9.998781585307577e-05,
"loss": 1.41162109375,
"step": 7
},
{
"epoch": 0.22857142857142856,
"grad_norm": 0.6675843000411987,
"learning_rate": 9.99512700102336e-05,
"loss": 1.41015625,
"step": 8
},
{
"epoch": 0.2571428571428571,
"grad_norm": 0.6396194100379944,
"learning_rate": 9.989038226169209e-05,
"loss": 1.474609375,
"step": 9
},
{
"epoch": 0.2857142857142857,
"grad_norm": 0.5296529531478882,
"learning_rate": 9.98051855792412e-05,
"loss": 1.4306640625,
"step": 10
},
{
"epoch": 0.3142857142857143,
"grad_norm": 0.2675076723098755,
"learning_rate": 9.969572609838744e-05,
"loss": 1.2821044921875,
"step": 11
},
{
"epoch": 0.34285714285714286,
"grad_norm": 0.2704640030860901,
"learning_rate": 9.956206309337068e-05,
"loss": 1.4039306640625,
"step": 12
},
{
"epoch": 0.37142857142857144,
"grad_norm": 0.2551879286766052,
"learning_rate": 9.940426894506607e-05,
"loss": 1.32421875,
"step": 13
},
{
"epoch": 0.4,
"grad_norm": 0.35215961933135986,
"learning_rate": 9.92224291017886e-05,
"loss": 1.025390625,
"step": 14
},
{
"epoch": 0.42857142857142855,
"grad_norm": 0.2997990548610687,
"learning_rate": 9.901664203302126e-05,
"loss": 1.3216552734375,
"step": 15
},
{
"epoch": 0.45714285714285713,
"grad_norm": 0.25571194291114807,
"learning_rate": 9.878701917609207e-05,
"loss": 1.3499755859375,
"step": 16
},
{
"epoch": 0.4857142857142857,
"grad_norm": 0.34939518570899963,
"learning_rate": 9.853368487582887e-05,
"loss": 1.2916259765625,
"step": 17
},
{
"epoch": 0.5142857142857142,
"grad_norm": 0.17111709713935852,
"learning_rate": 9.825677631722435e-05,
"loss": 1.322265625,
"step": 18
},
{
"epoch": 0.5428571428571428,
"grad_norm": 0.15961293876171112,
"learning_rate": 9.795644345114795e-05,
"loss": 1.359375,
"step": 19
},
{
"epoch": 0.5714285714285714,
"grad_norm": 0.16286684572696686,
"learning_rate": 9.763284891314482e-05,
"loss": 1.212890625,
"step": 20
},
{
"epoch": 0.6,
"grad_norm": 0.15212465822696686,
"learning_rate": 9.728616793536588e-05,
"loss": 1.3544921875,
"step": 21
},
{
"epoch": 0.6285714285714286,
"grad_norm": 0.15721262991428375,
"learning_rate": 9.69165882516764e-05,
"loss": 1.29296875,
"step": 22
},
{
"epoch": 0.6571428571428571,
"grad_norm": 0.15810737013816833,
"learning_rate": 9.652430999599491e-05,
"loss": 1.2890625,
"step": 23
},
{
"epoch": 0.6857142857142857,
"grad_norm": 0.30460578203201294,
"learning_rate": 9.610954559391703e-05,
"loss": 1.294189453125,
"step": 24
},
{
"epoch": 0.7142857142857143,
"grad_norm": 0.17775993049144745,
"learning_rate": 9.567251964768343e-05,
"loss": 1.2939453125,
"step": 25
},
{
"epoch": 0.7428571428571429,
"grad_norm": 0.1571461409330368,
"learning_rate": 9.521346881455356e-05,
"loss": 1.2939453125,
"step": 26
},
{
"epoch": 0.7714285714285715,
"grad_norm": 0.11447877436876297,
"learning_rate": 9.473264167865173e-05,
"loss": 1.3115234375,
"step": 27
},
{
"epoch": 0.8,
"grad_norm": 0.08568654954433441,
"learning_rate": 9.423029861635431e-05,
"loss": 1.08123779296875,
"step": 28
},
{
"epoch": 0.8,
"eval_loss": 1.21484375,
"eval_runtime": 5.4015,
"eval_samples_per_second": 0.37,
"eval_steps_per_second": 0.185,
"step": 28
},
{
"epoch": 0.8285714285714286,
"grad_norm": 0.13479705154895782,
"learning_rate": 9.370671165529146e-05,
"loss": 1.2197265625,
"step": 29
},
{
"epoch": 0.8571428571428571,
"grad_norm": 0.13089671730995178,
"learning_rate": 9.316216432703917e-05,
"loss": 1.177734375,
"step": 30
},
{
"epoch": 0.8857142857142857,
"grad_norm": 0.16832147538661957,
"learning_rate": 9.259695151358214e-05,
"loss": 1.296875,
"step": 31
},
{
"epoch": 0.9142857142857143,
"grad_norm": 0.1125297099351883,
"learning_rate": 9.201137928762981e-05,
"loss": 1.28515625,
"step": 32
},
{
"epoch": 0.9428571428571428,
"grad_norm": 0.0942939817905426,
"learning_rate": 9.140576474687264e-05,
"loss": 1.24267578125,
"step": 33
},
{
"epoch": 0.9714285714285714,
"grad_norm": 0.11853406578302383,
"learning_rate": 9.078043584226815e-05,
"loss": 1.2470703125,
"step": 34
},
{
"epoch": 1.0,
"grad_norm": 0.8680324554443359,
"learning_rate": 9.013573120044967e-05,
"loss": 1.0714111328125,
"step": 35
},
{
"epoch": 1.0285714285714285,
"grad_norm": 0.0927174836397171,
"learning_rate": 8.947199994035401e-05,
"loss": 1.251953125,
"step": 36
},
{
"epoch": 1.0571428571428572,
"grad_norm": 0.10877764970064163,
"learning_rate": 8.878960148416748e-05,
"loss": 1.2294921875,
"step": 37
},
{
"epoch": 1.0857142857142856,
"grad_norm": 0.09218857437372208,
"learning_rate": 8.80889053626923e-05,
"loss": 1.220703125,
"step": 38
},
{
"epoch": 1.1142857142857143,
"grad_norm": 0.08481802046298981,
"learning_rate": 8.73702910152393e-05,
"loss": 1.19287109375,
"step": 39
},
{
"epoch": 1.1428571428571428,
"grad_norm": 0.08881324529647827,
"learning_rate": 8.663414758415479e-05,
"loss": 1.22021484375,
"step": 40
},
{
"epoch": 1.1714285714285715,
"grad_norm": 0.0994924008846283,
"learning_rate": 8.588087370409303e-05,
"loss": 1.271484375,
"step": 41
},
{
"epoch": 1.2,
"grad_norm": 0.09856198728084564,
"learning_rate": 8.511087728614862e-05,
"loss": 1.1400489807128906,
"step": 42
},
{
"epoch": 1.2285714285714286,
"grad_norm": 0.09289012849330902,
"learning_rate": 8.432457529696549e-05,
"loss": 1.2001953125,
"step": 43
},
{
"epoch": 1.2571428571428571,
"grad_norm": 0.10627222061157227,
"learning_rate": 8.352239353294196e-05,
"loss": 1.198577880859375,
"step": 44
},
{
"epoch": 1.2857142857142856,
"grad_norm": 0.15058334171772003,
"learning_rate": 8.270476638965462e-05,
"loss": 1.192840576171875,
"step": 45
},
{
"epoch": 1.3142857142857143,
"grad_norm": 0.09897656738758087,
"learning_rate": 8.187213662662538e-05,
"loss": 1.0825576782226562,
"step": 46
},
{
"epoch": 1.342857142857143,
"grad_norm": 0.09571877866983414,
"learning_rate": 8.102495512755939e-05,
"loss": 1.2222275733947754,
"step": 47
},
{
"epoch": 1.3714285714285714,
"grad_norm": 0.08556420356035233,
"learning_rate": 8.016368065618361e-05,
"loss": 1.16943359375,
"step": 48
},
{
"epoch": 1.4,
"grad_norm": 0.090641088783741,
"learning_rate": 7.928877960781808e-05,
"loss": 0.8678903579711914,
"step": 49
},
{
"epoch": 1.4285714285714286,
"grad_norm": 0.10227451473474503,
"learning_rate": 7.840072575681468e-05,
"loss": 1.1566047668457031,
"step": 50
},
{
"epoch": 1.457142857142857,
"grad_norm": 0.10105621814727783,
"learning_rate": 7.75e-05,
"loss": 1.198678970336914,
"step": 51
},
{
"epoch": 1.4857142857142858,
"grad_norm": 0.11792707443237305,
"learning_rate": 7.658709009626109e-05,
"loss": 1.1418228149414062,
"step": 52
},
{
"epoch": 1.5142857142857142,
"grad_norm": 0.096110500395298,
"learning_rate": 7.566249040241553e-05,
"loss": 1.1884765625,
"step": 53
},
{
"epoch": 1.5428571428571427,
"grad_norm": 0.10725893080234528,
"learning_rate": 7.472670160550849e-05,
"loss": 1.23095703125,
"step": 54
},
{
"epoch": 1.5714285714285714,
"grad_norm": 0.09755166620016098,
"learning_rate": 7.378023045168181e-05,
"loss": 1.1025390625,
"step": 55
},
{
"epoch": 1.6,
"grad_norm": 0.10026982426643372,
"learning_rate": 7.282358947176207e-05,
"loss": 1.22265625,
"step": 56
},
{
"epoch": 1.6,
"eval_loss": 1.140625,
"eval_runtime": 5.4002,
"eval_samples_per_second": 0.37,
"eval_steps_per_second": 0.185,
"step": 56
},
{
"epoch": 1.6285714285714286,
"grad_norm": 0.1034621149301529,
"learning_rate": 7.185729670371605e-05,
"loss": 1.1748046875,
"step": 57
},
{
"epoch": 1.657142857142857,
"grad_norm": 0.08397577702999115,
"learning_rate": 7.08818754121241e-05,
"loss": 1.1796875,
"step": 58
},
{
"epoch": 1.6857142857142857,
"grad_norm": 0.10789086669683456,
"learning_rate": 6.989785380482312e-05,
"loss": 1.1534786224365234,
"step": 59
},
{
"epoch": 1.7142857142857144,
"grad_norm": 0.10360898822546005,
"learning_rate": 6.890576474687263e-05,
"loss": 1.1875,
"step": 60
},
{
"epoch": 1.7428571428571429,
"grad_norm": 0.09336746484041214,
"learning_rate": 6.790614547199907e-05,
"loss": 1.1865234375,
"step": 61
},
{
"epoch": 1.7714285714285714,
"grad_norm": 0.1032928079366684,
"learning_rate": 6.689953729167411e-05,
"loss": 1.2080078125,
"step": 62
},
{
"epoch": 1.8,
"grad_norm": 0.0901721864938736,
"learning_rate": 6.588648530198504e-05,
"loss": 0.99957275390625,
"step": 63
},
{
"epoch": 1.8285714285714287,
"grad_norm": 0.08693459630012512,
"learning_rate": 6.486753808845564e-05,
"loss": 1.1259765625,
"step": 64
},
{
"epoch": 1.8571428571428572,
"grad_norm": 0.08747872710227966,
"learning_rate": 6.384324742897736e-05,
"loss": 1.09375,
"step": 65
},
{
"epoch": 1.8857142857142857,
"grad_norm": 0.10195285826921463,
"learning_rate": 6.281416799501188e-05,
"loss": 1.1982421875,
"step": 66
},
{
"epoch": 1.9142857142857141,
"grad_norm": 0.09723973274230957,
"learning_rate": 6.178085705122675e-05,
"loss": 1.1904296875,
"step": 67
},
{
"epoch": 1.9428571428571428,
"grad_norm": 0.0876585841178894,
"learning_rate": 6.0743874153726765e-05,
"loss": 1.16015625,
"step": 68
},
{
"epoch": 1.9714285714285715,
"grad_norm": 0.10089046508073807,
"learning_rate": 5.970378084704441e-05,
"loss": 1.16162109375,
"step": 69
},
{
"epoch": 2.0,
"grad_norm": 0.08220144361257553,
"learning_rate": 5.8661140360053624e-05,
"loss": 0.9820556640625,
"step": 70
},
{
"epoch": 2.0285714285714285,
"grad_norm": 0.09639836102724075,
"learning_rate": 5.761651730097142e-05,
"loss": 1.1689453125,
"step": 71
},
{
"epoch": 2.057142857142857,
"grad_norm": 0.08934585005044937,
"learning_rate": 5.657047735161256e-05,
"loss": 1.154296875,
"step": 72
},
{
"epoch": 2.085714285714286,
"grad_norm": 0.09140218794345856,
"learning_rate": 5.552358696106288e-05,
"loss": 1.1484375,
"step": 73
},
{
"epoch": 2.1142857142857143,
"grad_norm": 0.09645097702741623,
"learning_rate": 5.4476413038937144e-05,
"loss": 1.12060546875,
"step": 74
},
{
"epoch": 2.142857142857143,
"grad_norm": 0.09433187544345856,
"learning_rate": 5.342952264838747e-05,
"loss": 1.146484375,
"step": 75
},
{
"epoch": 2.1714285714285713,
"grad_norm": 0.10496731847524643,
"learning_rate": 5.23834826990286e-05,
"loss": 1.2001953125,
"step": 76
},
{
"epoch": 2.2,
"grad_norm": 0.10152551531791687,
"learning_rate": 5.1338859639946394e-05,
"loss": 1.070345163345337,
"step": 77
},
{
"epoch": 2.2285714285714286,
"grad_norm": 0.10176125913858414,
"learning_rate": 5.0296219152955604e-05,
"loss": 1.12890625,
"step": 78
},
{
"epoch": 2.257142857142857,
"grad_norm": 0.10605648159980774,
"learning_rate": 4.9256125846273246e-05,
"loss": 1.1309094429016113,
"step": 79
},
{
"epoch": 2.2857142857142856,
"grad_norm": 0.09611742198467255,
"learning_rate": 4.821914294877327e-05,
"loss": 1.1230978965759277,
"step": 80
},
{
"epoch": 2.314285714285714,
"grad_norm": 0.08913472294807434,
"learning_rate": 4.718583200498814e-05,
"loss": 1.0227570533752441,
"step": 81
},
{
"epoch": 2.342857142857143,
"grad_norm": 0.11219677329063416,
"learning_rate": 4.615675257102264e-05,
"loss": 1.1562614440917969,
"step": 82
},
{
"epoch": 2.3714285714285714,
"grad_norm": 0.10772853344678879,
"learning_rate": 4.513246191154434e-05,
"loss": 1.109375,
"step": 83
},
{
"epoch": 2.4,
"grad_norm": 0.08562427759170532,
"learning_rate": 4.411351469801496e-05,
"loss": 0.8189945220947266,
"step": 84
},
{
"epoch": 2.4,
"eval_loss": 1.109375,
"eval_runtime": 5.3939,
"eval_samples_per_second": 0.371,
"eval_steps_per_second": 0.185,
"step": 84
},
{
"epoch": 2.4285714285714284,
"grad_norm": 0.10952357202768326,
"learning_rate": 4.3100462708325916e-05,
"loss": 1.0928478240966797,
"step": 85
},
{
"epoch": 2.4571428571428573,
"grad_norm": 0.1026853621006012,
"learning_rate": 4.209385452800095e-05,
"loss": 1.1446080207824707,
"step": 86
},
{
"epoch": 2.4857142857142858,
"grad_norm": 0.10352090001106262,
"learning_rate": 4.109423525312738e-05,
"loss": 1.0831298828125,
"step": 87
},
{
"epoch": 2.5142857142857142,
"grad_norm": 0.10942506045103073,
"learning_rate": 4.010214619517689e-05,
"loss": 1.1318359375,
"step": 88
},
{
"epoch": 2.5428571428571427,
"grad_norm": 0.10975760221481323,
"learning_rate": 3.911812458787591e-05,
"loss": 1.177734375,
"step": 89
},
{
"epoch": 2.571428571428571,
"grad_norm": 0.09781485795974731,
"learning_rate": 3.814270329628396e-05,
"loss": 1.05078125,
"step": 90
},
{
"epoch": 2.6,
"grad_norm": 0.10182441025972366,
"learning_rate": 3.717641052823795e-05,
"loss": 1.1669921875,
"step": 91
},
{
"epoch": 2.6285714285714286,
"grad_norm": 0.11254462599754333,
"learning_rate": 3.62197695483182e-05,
"loss": 1.12255859375,
"step": 92
},
{
"epoch": 2.657142857142857,
"grad_norm": 0.09910311549901962,
"learning_rate": 3.5273298394491515e-05,
"loss": 1.1318359375,
"step": 93
},
{
"epoch": 2.685714285714286,
"grad_norm": 0.15467706322669983,
"learning_rate": 3.433750959758447e-05,
"loss": 1.1025604009628296,
"step": 94
},
{
"epoch": 2.7142857142857144,
"grad_norm": 0.09800928831100464,
"learning_rate": 3.3412909903738935e-05,
"loss": 1.1396484375,
"step": 95
},
{
"epoch": 2.742857142857143,
"grad_norm": 0.10627949237823486,
"learning_rate": 3.250000000000001e-05,
"loss": 1.1376953125,
"step": 96
},
{
"epoch": 2.7714285714285714,
"grad_norm": 0.1054811105132103,
"learning_rate": 3.159927424318531e-05,
"loss": 1.1630859375,
"step": 97
},
{
"epoch": 2.8,
"grad_norm": 0.09246952831745148,
"learning_rate": 3.071122039218194e-05,
"loss": 0.96343994140625,
"step": 98
},
{
"epoch": 2.8285714285714287,
"grad_norm": 0.09628032147884369,
"learning_rate": 2.98363193438164e-05,
"loss": 1.08544921875,
"step": 99
},
{
"epoch": 2.857142857142857,
"grad_norm": 0.09441180527210236,
"learning_rate": 2.897504487244061e-05,
"loss": 1.05615234375,
"step": 100
},
{
"epoch": 2.8857142857142857,
"grad_norm": 0.11135559529066086,
"learning_rate": 2.8127863373374635e-05,
"loss": 1.1552734375,
"step": 101
},
{
"epoch": 2.914285714285714,
"grad_norm": 0.1078437864780426,
"learning_rate": 2.729523361034538e-05,
"loss": 1.1484375,
"step": 102
},
{
"epoch": 2.942857142857143,
"grad_norm": 0.10249437391757965,
"learning_rate": 2.6477606467058035e-05,
"loss": 1.12060546875,
"step": 103
},
{
"epoch": 2.9714285714285715,
"grad_norm": 0.10100141167640686,
"learning_rate": 2.5675424703034524e-05,
"loss": 1.12109375,
"step": 104
},
{
"epoch": 3.0,
"grad_norm": 0.08678524196147919,
"learning_rate": 2.4889122713851394e-05,
"loss": 0.94921875,
"step": 105
},
{
"epoch": 3.0285714285714285,
"grad_norm": 0.10091595351696014,
"learning_rate": 2.4119126295906998e-05,
"loss": 1.12744140625,
"step": 106
},
{
"epoch": 3.057142857142857,
"grad_norm": 0.09988492727279663,
"learning_rate": 2.3365852415845223e-05,
"loss": 1.119140625,
"step": 107
},
{
"epoch": 3.085714285714286,
"grad_norm": 0.10282906144857407,
"learning_rate": 2.2629708984760708e-05,
"loss": 1.11279296875,
"step": 108
},
{
"epoch": 3.1142857142857143,
"grad_norm": 0.09999439865350723,
"learning_rate": 2.1911094637307716e-05,
"loss": 1.0869140625,
"step": 109
},
{
"epoch": 3.142857142857143,
"grad_norm": 0.1038166955113411,
"learning_rate": 2.121039851583254e-05,
"loss": 1.115234375,
"step": 110
},
{
"epoch": 3.1714285714285713,
"grad_norm": 0.10540751367807388,
"learning_rate": 2.0528000059645997e-05,
"loss": 1.1650390625,
"step": 111
},
{
"epoch": 3.2,
"grad_norm": 0.13892337679862976,
"learning_rate": 1.9864268799550338e-05,
"loss": 1.039089322090149,
"step": 112
},
{
"epoch": 3.2,
"eval_loss": 1.09765625,
"eval_runtime": 5.4123,
"eval_samples_per_second": 0.37,
"eval_steps_per_second": 0.185,
"step": 112
},
{
"epoch": 3.2285714285714286,
"grad_norm": 0.09683047980070114,
"learning_rate": 1.9219564157731846e-05,
"loss": 1.09814453125,
"step": 113
},
{
"epoch": 3.257142857142857,
"grad_norm": 0.10966203361749649,
"learning_rate": 1.8594235253127375e-05,
"loss": 1.097679853439331,
"step": 114
},
{
"epoch": 3.2857142857142856,
"grad_norm": 0.10792383551597595,
"learning_rate": 1.7988620712370196e-05,
"loss": 1.093778371810913,
"step": 115
},
{
"epoch": 3.314285714285714,
"grad_norm": 0.09936859458684921,
"learning_rate": 1.740304848641787e-05,
"loss": 0.99672532081604,
"step": 116
},
{
"epoch": 3.342857142857143,
"grad_norm": 0.1135304793715477,
"learning_rate": 1.6837835672960835e-05,
"loss": 1.1254997253417969,
"step": 117
},
{
"epoch": 3.3714285714285714,
"grad_norm": 0.10551383346319199,
"learning_rate": 1.6293288344708568e-05,
"loss": 1.0830078125,
"step": 118
},
{
"epoch": 3.4,
"grad_norm": 0.09492458403110504,
"learning_rate": 1.57697013836457e-05,
"loss": 0.7991197109222412,
"step": 119
},
{
"epoch": 3.4285714285714284,
"grad_norm": 0.10766921192407608,
"learning_rate": 1.526735832134829e-05,
"loss": 1.0654666423797607,
"step": 120
},
{
"epoch": 3.4571428571428573,
"grad_norm": 0.1085314080119133,
"learning_rate": 1.4786531185446454e-05,
"loss": 1.1201512813568115,
"step": 121
},
{
"epoch": 3.4857142857142858,
"grad_norm": 0.11182897537946701,
"learning_rate": 1.432748035231658e-05,
"loss": 1.0567700862884521,
"step": 122
},
{
"epoch": 3.5142857142857142,
"grad_norm": 0.10476535558700562,
"learning_rate": 1.389045440608296e-05,
"loss": 1.10693359375,
"step": 123
},
{
"epoch": 3.5428571428571427,
"grad_norm": 0.10022886097431183,
"learning_rate": 1.3475690004005098e-05,
"loss": 1.15234375,
"step": 124
},
{
"epoch": 3.571428571428571,
"grad_norm": 0.09982618689537048,
"learning_rate": 1.3083411748323591e-05,
"loss": 1.029296875,
"step": 125
},
{
"epoch": 3.6,
"grad_norm": 0.10781899839639664,
"learning_rate": 1.2713832064634126e-05,
"loss": 1.14453125,
"step": 126
},
{
"epoch": 3.6285714285714286,
"grad_norm": 0.10565479844808578,
"learning_rate": 1.2367151086855187e-05,
"loss": 1.10302734375,
"step": 127
},
{
"epoch": 3.657142857142857,
"grad_norm": 0.09674814343452454,
"learning_rate": 1.2043556548852065e-05,
"loss": 1.11083984375,
"step": 128
},
{
"epoch": 3.685714285714286,
"grad_norm": 0.10641477257013321,
"learning_rate": 1.174322368277565e-05,
"loss": 1.0820510387420654,
"step": 129
},
{
"epoch": 3.7142857142857144,
"grad_norm": 0.09659207612276077,
"learning_rate": 1.146631512417113e-05,
"loss": 1.11962890625,
"step": 130
},
{
"epoch": 3.742857142857143,
"grad_norm": 0.10608964413404465,
"learning_rate": 1.1212980823907931e-05,
"loss": 1.12060546875,
"step": 131
},
{
"epoch": 3.7714285714285714,
"grad_norm": 0.10524872690439224,
"learning_rate": 1.0983357966978745e-05,
"loss": 1.14306640625,
"step": 132
},
{
"epoch": 3.8,
"grad_norm": 0.09818793088197708,
"learning_rate": 1.0777570898211406e-05,
"loss": 0.94818115234375,
"step": 133
},
{
"epoch": 3.8285714285714287,
"grad_norm": 0.09379758685827255,
"learning_rate": 1.0595731054933936e-05,
"loss": 1.06787109375,
"step": 134
},
{
"epoch": 3.857142857142857,
"grad_norm": 0.09733222424983978,
"learning_rate": 1.0437936906629336e-05,
"loss": 1.0400390625,
"step": 135
},
{
"epoch": 3.8857142857142857,
"grad_norm": 0.10898660868406296,
"learning_rate": 1.0304273901612566e-05,
"loss": 1.13818359375,
"step": 136
},
{
"epoch": 3.914285714285714,
"grad_norm": 0.11681357771158218,
"learning_rate": 1.0194814420758805e-05,
"loss": 1.1318359375,
"step": 137
},
{
"epoch": 3.942857142857143,
"grad_norm": 0.10775525122880936,
"learning_rate": 1.0109617738307912e-05,
"loss": 1.10498046875,
"step": 138
},
{
"epoch": 3.9714285714285715,
"grad_norm": 0.10294745117425919,
"learning_rate": 1.0048729989766395e-05,
"loss": 1.1083984375,
"step": 139
},
{
"epoch": 4.0,
"grad_norm": 0.08989562839269638,
"learning_rate": 1.0012184146924224e-05,
"loss": 0.934478759765625,
"step": 140
},
{
"epoch": 4.0,
"eval_loss": 1.08984375,
"eval_runtime": 5.3982,
"eval_samples_per_second": 0.37,
"eval_steps_per_second": 0.185,
"step": 140
}
],
"logging_steps": 1.0,
"max_steps": 140,
"num_input_tokens_seen": 0,
"num_train_epochs": 4,
"save_steps": 0,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.2160898563718513e+19,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.0,
"eval_steps": 28,
"global_step": 140,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02857142857142857,
"grad_norm": 0.9386152625083923,
"learning_rate": 0.0,
"loss": 1.744140625,
"step": 1
},
{
"epoch": 0.05714285714285714,
"grad_norm": 0.852827250957489,
"learning_rate": 2e-05,
"loss": 1.6787109375,
"step": 2
},
{
"epoch": 0.08571428571428572,
"grad_norm": 0.8461427092552185,
"learning_rate": 4e-05,
"loss": 1.671875,
"step": 3
},
{
"epoch": 0.11428571428571428,
"grad_norm": 0.6573083400726318,
"learning_rate": 6e-05,
"loss": 1.5703125,
"step": 4
},
{
"epoch": 0.14285714285714285,
"grad_norm": 0.5101009011268616,
"learning_rate": 8e-05,
"loss": 1.5732421875,
"step": 5
},
{
"epoch": 0.17142857142857143,
"grad_norm": 0.3327346444129944,
"learning_rate": 0.0001,
"loss": 1.5615234375,
"step": 6
},
{
"epoch": 0.2,
"grad_norm": 0.5101948380470276,
"learning_rate": 9.998781585307577e-05,
"loss": 1.41162109375,
"step": 7
},
{
"epoch": 0.22857142857142856,
"grad_norm": 0.6675843000411987,
"learning_rate": 9.99512700102336e-05,
"loss": 1.41015625,
"step": 8
},
{
"epoch": 0.2571428571428571,
"grad_norm": 0.6396194100379944,
"learning_rate": 9.989038226169209e-05,
"loss": 1.474609375,
"step": 9
},
{
"epoch": 0.2857142857142857,
"grad_norm": 0.5296529531478882,
"learning_rate": 9.98051855792412e-05,
"loss": 1.4306640625,
"step": 10
},
{
"epoch": 0.3142857142857143,
"grad_norm": 0.2675076723098755,
"learning_rate": 9.969572609838744e-05,
"loss": 1.2821044921875,
"step": 11
},
{
"epoch": 0.34285714285714286,
"grad_norm": 0.2704640030860901,
"learning_rate": 9.956206309337068e-05,
"loss": 1.4039306640625,
"step": 12
},
{
"epoch": 0.37142857142857144,
"grad_norm": 0.2551879286766052,
"learning_rate": 9.940426894506607e-05,
"loss": 1.32421875,