sanchit-gandhi's picture
End of training
7fdabdb verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 15.835051546391753,
"eval_steps": 30,
"global_step": 96,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.16494845360824742,
"grad_norm": 1.6045022010803223,
"learning_rate": 2.9687500000000003e-05,
"loss": 10.1016,
"step": 1
},
{
"epoch": 0.32989690721649484,
"grad_norm": 1.7116177082061768,
"learning_rate": 2.9375e-05,
"loss": 10.2357,
"step": 2
},
{
"epoch": 0.4948453608247423,
"grad_norm": 1.7453874349594116,
"learning_rate": 2.90625e-05,
"loss": 10.0525,
"step": 3
},
{
"epoch": 0.6597938144329897,
"grad_norm": 2.173015594482422,
"learning_rate": 2.875e-05,
"loss": 10.2,
"step": 4
},
{
"epoch": 0.8247422680412371,
"grad_norm": 2.119373083114624,
"learning_rate": 2.84375e-05,
"loss": 10.0632,
"step": 5
},
{
"epoch": 0.9896907216494846,
"grad_norm": 2.180950403213501,
"learning_rate": 2.8125e-05,
"loss": 9.8459,
"step": 6
},
{
"epoch": 1.1546391752577319,
"grad_norm": 2.3363752365112305,
"learning_rate": 2.7812500000000002e-05,
"loss": 9.9087,
"step": 7
},
{
"epoch": 1.3195876288659794,
"grad_norm": 2.6667237281799316,
"learning_rate": 2.75e-05,
"loss": 9.8297,
"step": 8
},
{
"epoch": 1.4845360824742269,
"grad_norm": 3.121326446533203,
"learning_rate": 2.71875e-05,
"loss": 9.8469,
"step": 9
},
{
"epoch": 1.6494845360824741,
"grad_norm": 2.7169275283813477,
"learning_rate": 2.6875000000000003e-05,
"loss": 9.5758,
"step": 10
},
{
"epoch": 1.8144329896907216,
"grad_norm": 2.9351377487182617,
"learning_rate": 2.65625e-05,
"loss": 9.5564,
"step": 11
},
{
"epoch": 1.9793814432989691,
"grad_norm": 3.241955041885376,
"learning_rate": 2.625e-05,
"loss": 9.4493,
"step": 12
},
{
"epoch": 2.1443298969072164,
"grad_norm": 3.388002872467041,
"learning_rate": 2.59375e-05,
"loss": 9.3817,
"step": 13
},
{
"epoch": 2.3092783505154637,
"grad_norm": 3.465426445007324,
"learning_rate": 2.5625e-05,
"loss": 9.2322,
"step": 14
},
{
"epoch": 2.4742268041237114,
"grad_norm": 3.8056743144989014,
"learning_rate": 2.5312500000000002e-05,
"loss": 9.1842,
"step": 15
},
{
"epoch": 2.6391752577319587,
"grad_norm": 3.7609691619873047,
"learning_rate": 2.5e-05,
"loss": 9.0523,
"step": 16
},
{
"epoch": 2.804123711340206,
"grad_norm": 3.9078407287597656,
"learning_rate": 2.46875e-05,
"loss": 8.9138,
"step": 17
},
{
"epoch": 2.9690721649484537,
"grad_norm": 3.792387008666992,
"learning_rate": 2.4375e-05,
"loss": 8.829,
"step": 18
},
{
"epoch": 3.134020618556701,
"grad_norm": 3.535268783569336,
"learning_rate": 2.4062500000000002e-05,
"loss": 8.6985,
"step": 19
},
{
"epoch": 3.2989690721649483,
"grad_norm": 3.926476240158081,
"learning_rate": 2.3749999999999998e-05,
"loss": 8.5909,
"step": 20
},
{
"epoch": 3.463917525773196,
"grad_norm": 3.3404812812805176,
"learning_rate": 2.34375e-05,
"loss": 8.5582,
"step": 21
},
{
"epoch": 3.6288659793814433,
"grad_norm": 3.4405341148376465,
"learning_rate": 2.3125000000000003e-05,
"loss": 8.3595,
"step": 22
},
{
"epoch": 3.7938144329896906,
"grad_norm": 3.84586238861084,
"learning_rate": 2.28125e-05,
"loss": 8.2115,
"step": 23
},
{
"epoch": 3.9587628865979383,
"grad_norm": 3.3119468688964844,
"learning_rate": 2.25e-05,
"loss": 8.1774,
"step": 24
},
{
"epoch": 4.123711340206185,
"grad_norm": 3.2873129844665527,
"learning_rate": 2.21875e-05,
"loss": 8.1239,
"step": 25
},
{
"epoch": 4.288659793814433,
"grad_norm": 3.5876059532165527,
"learning_rate": 2.1875e-05,
"loss": 8.1053,
"step": 26
},
{
"epoch": 4.453608247422681,
"grad_norm": 3.688255786895752,
"learning_rate": 2.15625e-05,
"loss": 7.9861,
"step": 27
},
{
"epoch": 4.618556701030927,
"grad_norm": 4.39879846572876,
"learning_rate": 2.125e-05,
"loss": 7.8541,
"step": 28
},
{
"epoch": 4.783505154639175,
"grad_norm": 4.439304828643799,
"learning_rate": 2.09375e-05,
"loss": 7.9102,
"step": 29
},
{
"epoch": 4.948453608247423,
"grad_norm": 3.9502387046813965,
"learning_rate": 2.0625e-05,
"loss": 7.791,
"step": 30
},
{
"epoch": 4.948453608247423,
"eval_audio_cosine_sim": 0.5583739280700684,
"eval_loss": 3.1003806591033936,
"eval_runtime": 2123.8603,
"eval_samples_per_second": 0.006,
"eval_steps_per_second": 0.006,
"eval_text_cosine_sim": 0.376191645860672,
"step": 30
},
{
"epoch": 5.11340206185567,
"grad_norm": 4.569061756134033,
"learning_rate": 2.0312500000000002e-05,
"loss": 7.746,
"step": 31
},
{
"epoch": 5.278350515463917,
"grad_norm": 4.3288702964782715,
"learning_rate": 1.9999999999999998e-05,
"loss": 7.5649,
"step": 32
},
{
"epoch": 5.443298969072165,
"grad_norm": 4.021690368652344,
"learning_rate": 1.96875e-05,
"loss": 7.7969,
"step": 33
},
{
"epoch": 5.608247422680412,
"grad_norm": 3.7971837520599365,
"learning_rate": 1.9375e-05,
"loss": 7.5618,
"step": 34
},
{
"epoch": 5.77319587628866,
"grad_norm": 3.3054466247558594,
"learning_rate": 1.90625e-05,
"loss": 7.5207,
"step": 35
},
{
"epoch": 5.938144329896907,
"grad_norm": 3.0602829456329346,
"learning_rate": 1.8750000000000002e-05,
"loss": 7.3319,
"step": 36
},
{
"epoch": 6.103092783505154,
"grad_norm": 3.018862009048462,
"learning_rate": 1.84375e-05,
"loss": 7.2311,
"step": 37
},
{
"epoch": 6.268041237113402,
"grad_norm": 2.6457912921905518,
"learning_rate": 1.8125e-05,
"loss": 7.2177,
"step": 38
},
{
"epoch": 6.43298969072165,
"grad_norm": 2.6215384006500244,
"learning_rate": 1.78125e-05,
"loss": 7.2411,
"step": 39
},
{
"epoch": 6.597938144329897,
"grad_norm": 2.6344871520996094,
"learning_rate": 1.7500000000000002e-05,
"loss": 7.2866,
"step": 40
},
{
"epoch": 6.762886597938144,
"grad_norm": 2.753688335418701,
"learning_rate": 1.7187499999999998e-05,
"loss": 7.3974,
"step": 41
},
{
"epoch": 6.927835051546392,
"grad_norm": 2.540292501449585,
"learning_rate": 1.6875e-05,
"loss": 7.2237,
"step": 42
},
{
"epoch": 7.092783505154639,
"grad_norm": 2.3271636962890625,
"learning_rate": 1.6562500000000003e-05,
"loss": 7.0722,
"step": 43
},
{
"epoch": 7.257731958762887,
"grad_norm": 2.6378014087677,
"learning_rate": 1.625e-05,
"loss": 6.9758,
"step": 44
},
{
"epoch": 7.422680412371134,
"grad_norm": 2.6299326419830322,
"learning_rate": 1.59375e-05,
"loss": 7.2835,
"step": 45
},
{
"epoch": 7.587628865979381,
"grad_norm": 2.400949716567993,
"learning_rate": 1.5625e-05,
"loss": 7.1064,
"step": 46
},
{
"epoch": 7.752577319587629,
"grad_norm": 2.3601369857788086,
"learning_rate": 1.53125e-05,
"loss": 6.7368,
"step": 47
},
{
"epoch": 7.917525773195877,
"grad_norm": 2.3395187854766846,
"learning_rate": 1.5e-05,
"loss": 7.045,
"step": 48
},
{
"epoch": 8.082474226804123,
"grad_norm": 2.2277605533599854,
"learning_rate": 1.46875e-05,
"loss": 6.8241,
"step": 49
},
{
"epoch": 8.24742268041237,
"grad_norm": 2.162241220474243,
"learning_rate": 1.4375e-05,
"loss": 6.9879,
"step": 50
},
{
"epoch": 8.412371134020619,
"grad_norm": 2.038295030593872,
"learning_rate": 1.40625e-05,
"loss": 6.8768,
"step": 51
},
{
"epoch": 8.577319587628866,
"grad_norm": 1.980531930923462,
"learning_rate": 1.375e-05,
"loss": 6.7726,
"step": 52
},
{
"epoch": 8.742268041237114,
"grad_norm": 1.9837162494659424,
"learning_rate": 1.3437500000000001e-05,
"loss": 6.6858,
"step": 53
},
{
"epoch": 8.907216494845361,
"grad_norm": 2.0047826766967773,
"learning_rate": 1.3125e-05,
"loss": 6.8887,
"step": 54
},
{
"epoch": 9.072164948453608,
"grad_norm": 1.956314206123352,
"learning_rate": 1.28125e-05,
"loss": 6.7538,
"step": 55
},
{
"epoch": 9.237113402061855,
"grad_norm": 1.6746240854263306,
"learning_rate": 1.25e-05,
"loss": 6.6686,
"step": 56
},
{
"epoch": 9.402061855670103,
"grad_norm": 1.9487406015396118,
"learning_rate": 1.21875e-05,
"loss": 6.61,
"step": 57
},
{
"epoch": 9.56701030927835,
"grad_norm": 1.6522157192230225,
"learning_rate": 1.1874999999999999e-05,
"loss": 6.7586,
"step": 58
},
{
"epoch": 9.731958762886597,
"grad_norm": 1.807323694229126,
"learning_rate": 1.1562500000000002e-05,
"loss": 6.5441,
"step": 59
},
{
"epoch": 9.896907216494846,
"grad_norm": 1.6479687690734863,
"learning_rate": 1.125e-05,
"loss": 6.6699,
"step": 60
},
{
"epoch": 9.896907216494846,
"eval_audio_cosine_sim": 0.6575908064842224,
"eval_loss": 3.235611915588379,
"eval_runtime": 2160.2372,
"eval_samples_per_second": 0.006,
"eval_steps_per_second": 0.006,
"eval_text_cosine_sim": 0.3653325140476227,
"step": 60
},
{
"epoch": 10.061855670103093,
"grad_norm": 1.6155036687850952,
"learning_rate": 1.09375e-05,
"loss": 6.6719,
"step": 61
},
{
"epoch": 10.22680412371134,
"grad_norm": 1.5656815767288208,
"learning_rate": 1.0625e-05,
"loss": 6.6531,
"step": 62
},
{
"epoch": 10.391752577319588,
"grad_norm": 1.4706956148147583,
"learning_rate": 1.03125e-05,
"loss": 6.6168,
"step": 63
},
{
"epoch": 10.556701030927835,
"grad_norm": 1.5122315883636475,
"learning_rate": 9.999999999999999e-06,
"loss": 6.4704,
"step": 64
},
{
"epoch": 10.721649484536082,
"grad_norm": 1.3326685428619385,
"learning_rate": 9.6875e-06,
"loss": 6.5744,
"step": 65
},
{
"epoch": 10.88659793814433,
"grad_norm": 1.4799164533615112,
"learning_rate": 9.375000000000001e-06,
"loss": 6.5442,
"step": 66
},
{
"epoch": 11.051546391752577,
"grad_norm": 1.6707019805908203,
"learning_rate": 9.0625e-06,
"loss": 6.3603,
"step": 67
},
{
"epoch": 11.216494845360824,
"grad_norm": 1.5303987264633179,
"learning_rate": 8.750000000000001e-06,
"loss": 6.3026,
"step": 68
},
{
"epoch": 11.381443298969073,
"grad_norm": 1.4235777854919434,
"learning_rate": 8.4375e-06,
"loss": 6.3932,
"step": 69
},
{
"epoch": 11.54639175257732,
"grad_norm": 1.7168058156967163,
"learning_rate": 8.125e-06,
"loss": 6.6932,
"step": 70
},
{
"epoch": 11.711340206185566,
"grad_norm": 1.4212720394134521,
"learning_rate": 7.8125e-06,
"loss": 6.3352,
"step": 71
},
{
"epoch": 11.876288659793815,
"grad_norm": 1.3087238073349,
"learning_rate": 7.5e-06,
"loss": 6.5214,
"step": 72
},
{
"epoch": 12.041237113402062,
"grad_norm": 1.343590259552002,
"learning_rate": 7.1875e-06,
"loss": 6.5502,
"step": 73
},
{
"epoch": 12.206185567010309,
"grad_norm": 1.1480090618133545,
"learning_rate": 6.875e-06,
"loss": 6.3793,
"step": 74
},
{
"epoch": 12.371134020618557,
"grad_norm": 1.4936941862106323,
"learning_rate": 6.5625e-06,
"loss": 6.5373,
"step": 75
},
{
"epoch": 12.536082474226804,
"grad_norm": 1.130237102508545,
"learning_rate": 6.25e-06,
"loss": 6.4033,
"step": 76
},
{
"epoch": 12.70103092783505,
"grad_norm": 1.4036816358566284,
"learning_rate": 5.9374999999999995e-06,
"loss": 6.382,
"step": 77
},
{
"epoch": 12.8659793814433,
"grad_norm": 1.4101227521896362,
"learning_rate": 5.625e-06,
"loss": 6.5297,
"step": 78
},
{
"epoch": 13.030927835051546,
"grad_norm": 1.2091238498687744,
"learning_rate": 5.3125e-06,
"loss": 6.4248,
"step": 79
},
{
"epoch": 13.195876288659793,
"grad_norm": 1.0961215496063232,
"learning_rate": 4.9999999999999996e-06,
"loss": 6.4123,
"step": 80
},
{
"epoch": 13.360824742268042,
"grad_norm": 1.08103609085083,
"learning_rate": 4.6875000000000004e-06,
"loss": 6.3088,
"step": 81
},
{
"epoch": 13.525773195876289,
"grad_norm": 1.4263334274291992,
"learning_rate": 4.3750000000000005e-06,
"loss": 6.5992,
"step": 82
},
{
"epoch": 13.690721649484535,
"grad_norm": 1.2003765106201172,
"learning_rate": 4.0625e-06,
"loss": 6.3439,
"step": 83
},
{
"epoch": 13.855670103092784,
"grad_norm": 1.0816315412521362,
"learning_rate": 3.75e-06,
"loss": 6.4004,
"step": 84
},
{
"epoch": 14.02061855670103,
"grad_norm": 1.4043329954147339,
"learning_rate": 3.4375e-06,
"loss": 6.1602,
"step": 85
},
{
"epoch": 14.185567010309278,
"grad_norm": 1.262394666671753,
"learning_rate": 3.125e-06,
"loss": 6.2039,
"step": 86
},
{
"epoch": 14.350515463917526,
"grad_norm": 1.147353172302246,
"learning_rate": 2.8125e-06,
"loss": 6.4834,
"step": 87
},
{
"epoch": 14.515463917525773,
"grad_norm": 1.0577898025512695,
"learning_rate": 2.4999999999999998e-06,
"loss": 6.3941,
"step": 88
},
{
"epoch": 14.68041237113402,
"grad_norm": 1.0920854806900024,
"learning_rate": 2.1875000000000002e-06,
"loss": 6.4342,
"step": 89
},
{
"epoch": 14.845360824742269,
"grad_norm": 1.0742192268371582,
"learning_rate": 1.875e-06,
"loss": 6.4394,
"step": 90
},
{
"epoch": 14.845360824742269,
"eval_audio_cosine_sim": 0.5900315642356873,
"eval_loss": 3.637291669845581,
"eval_runtime": 1070.3012,
"eval_samples_per_second": 0.011,
"eval_steps_per_second": 0.011,
"eval_text_cosine_sim": 0.26104021072387695,
"step": 90
},
{
"epoch": 15.010309278350515,
"grad_norm": 1.2300424575805664,
"learning_rate": 1.5625e-06,
"loss": 6.2831,
"step": 91
},
{
"epoch": 15.175257731958762,
"grad_norm": 1.1345258951187134,
"learning_rate": 1.2499999999999999e-06,
"loss": 6.4745,
"step": 92
},
{
"epoch": 15.34020618556701,
"grad_norm": 1.109592318534851,
"learning_rate": 9.375e-07,
"loss": 6.249,
"step": 93
},
{
"epoch": 15.505154639175258,
"grad_norm": 1.1140637397766113,
"learning_rate": 6.249999999999999e-07,
"loss": 6.3171,
"step": 94
},
{
"epoch": 15.670103092783505,
"grad_norm": 1.1929951906204224,
"learning_rate": 3.1249999999999997e-07,
"loss": 6.2597,
"step": 95
},
{
"epoch": 15.835051546391753,
"grad_norm": 1.2717913389205933,
"learning_rate": 0.0,
"loss": 6.548,
"step": 96
},
{
"epoch": 15.835051546391753,
"step": 96,
"total_flos": 1925884422732936.0,
"train_loss": 7.454789996147156,
"train_runtime": 13331.2439,
"train_samples_per_second": 0.116,
"train_steps_per_second": 0.007
}
],
"logging_steps": 1.0,
"max_steps": 96,
"num_input_tokens_seen": 0,
"num_train_epochs": 16,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": false,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1925884422732936.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}