sanchit-gandhi's picture
End of training
88b2eb1 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": 1.603933334350586,
"learning_rate": 2.9687500000000003e-05,
"loss": 10.1011,
"step": 1
},
{
"epoch": 0.32989690721649484,
"grad_norm": 1.7179864645004272,
"learning_rate": 2.9375e-05,
"loss": 10.2338,
"step": 2
},
{
"epoch": 0.4948453608247423,
"grad_norm": 1.748449444770813,
"learning_rate": 2.90625e-05,
"loss": 10.0554,
"step": 3
},
{
"epoch": 0.6597938144329897,
"grad_norm": 2.172945976257324,
"learning_rate": 2.875e-05,
"loss": 10.1999,
"step": 4
},
{
"epoch": 0.8247422680412371,
"grad_norm": 2.121206045150757,
"learning_rate": 2.84375e-05,
"loss": 10.0634,
"step": 5
},
{
"epoch": 0.9896907216494846,
"grad_norm": 2.179741144180298,
"learning_rate": 2.8125e-05,
"loss": 9.8454,
"step": 6
},
{
"epoch": 1.1546391752577319,
"grad_norm": 2.3345258235931396,
"learning_rate": 2.7812500000000002e-05,
"loss": 9.9095,
"step": 7
},
{
"epoch": 1.3195876288659794,
"grad_norm": 2.664315938949585,
"learning_rate": 2.75e-05,
"loss": 9.8296,
"step": 8
},
{
"epoch": 1.4845360824742269,
"grad_norm": 3.1426305770874023,
"learning_rate": 2.71875e-05,
"loss": 9.8446,
"step": 9
},
{
"epoch": 1.6494845360824741,
"grad_norm": 2.7230398654937744,
"learning_rate": 2.6875000000000003e-05,
"loss": 9.5769,
"step": 10
},
{
"epoch": 1.8144329896907216,
"grad_norm": 2.9412689208984375,
"learning_rate": 2.65625e-05,
"loss": 9.5563,
"step": 11
},
{
"epoch": 1.9793814432989691,
"grad_norm": 3.2456820011138916,
"learning_rate": 2.625e-05,
"loss": 9.4496,
"step": 12
},
{
"epoch": 2.1443298969072164,
"grad_norm": 3.3948400020599365,
"learning_rate": 2.59375e-05,
"loss": 9.3822,
"step": 13
},
{
"epoch": 2.3092783505154637,
"grad_norm": 3.4787559509277344,
"learning_rate": 2.5625e-05,
"loss": 9.2314,
"step": 14
},
{
"epoch": 2.4742268041237114,
"grad_norm": 3.8148350715637207,
"learning_rate": 2.5312500000000002e-05,
"loss": 9.1825,
"step": 15
},
{
"epoch": 2.6391752577319587,
"grad_norm": 3.7718141078948975,
"learning_rate": 2.5e-05,
"loss": 9.0502,
"step": 16
},
{
"epoch": 2.804123711340206,
"grad_norm": 3.9114766120910645,
"learning_rate": 2.46875e-05,
"loss": 8.9091,
"step": 17
},
{
"epoch": 2.9690721649484537,
"grad_norm": 3.791461706161499,
"learning_rate": 2.4375e-05,
"loss": 8.825,
"step": 18
},
{
"epoch": 3.134020618556701,
"grad_norm": 3.532548666000366,
"learning_rate": 2.4062500000000002e-05,
"loss": 8.6953,
"step": 19
},
{
"epoch": 3.2989690721649483,
"grad_norm": 3.9233362674713135,
"learning_rate": 2.3749999999999998e-05,
"loss": 8.5862,
"step": 20
},
{
"epoch": 3.463917525773196,
"grad_norm": 3.315626621246338,
"learning_rate": 2.34375e-05,
"loss": 8.5552,
"step": 21
},
{
"epoch": 3.6288659793814433,
"grad_norm": 3.417571783065796,
"learning_rate": 2.3125000000000003e-05,
"loss": 8.358,
"step": 22
},
{
"epoch": 3.7938144329896906,
"grad_norm": 3.810925006866455,
"learning_rate": 2.28125e-05,
"loss": 8.2089,
"step": 23
},
{
"epoch": 3.9587628865979383,
"grad_norm": 3.319178819656372,
"learning_rate": 2.25e-05,
"loss": 8.1787,
"step": 24
},
{
"epoch": 4.123711340206185,
"grad_norm": 3.318840742111206,
"learning_rate": 2.21875e-05,
"loss": 8.1272,
"step": 25
},
{
"epoch": 4.288659793814433,
"grad_norm": 3.684760332107544,
"learning_rate": 2.1875e-05,
"loss": 8.1127,
"step": 26
},
{
"epoch": 4.453608247422681,
"grad_norm": 3.7684097290039062,
"learning_rate": 2.15625e-05,
"loss": 7.9922,
"step": 27
},
{
"epoch": 4.618556701030927,
"grad_norm": 4.493322849273682,
"learning_rate": 2.125e-05,
"loss": 7.8609,
"step": 28
},
{
"epoch": 4.783505154639175,
"grad_norm": 4.545900821685791,
"learning_rate": 2.09375e-05,
"loss": 7.9192,
"step": 29
},
{
"epoch": 4.948453608247423,
"grad_norm": 3.999005079269409,
"learning_rate": 2.0625e-05,
"loss": 7.7976,
"step": 30
},
{
"epoch": 4.948453608247423,
"eval_audio_cosine_sim": 0.5581808090209961,
"eval_loss": 3.100252151489258,
"eval_runtime": 2102.121,
"eval_samples_per_second": 0.006,
"eval_steps_per_second": 0.006,
"eval_text_cosine_sim": 0.3910459578037262,
"step": 30
},
{
"epoch": 5.11340206185567,
"grad_norm": 4.612445831298828,
"learning_rate": 2.0312500000000002e-05,
"loss": 7.7548,
"step": 31
},
{
"epoch": 5.278350515463917,
"grad_norm": 4.341012001037598,
"learning_rate": 1.9999999999999998e-05,
"loss": 7.5716,
"step": 32
},
{
"epoch": 5.443298969072165,
"grad_norm": 4.00157356262207,
"learning_rate": 1.96875e-05,
"loss": 7.8051,
"step": 33
},
{
"epoch": 5.608247422680412,
"grad_norm": 3.777099847793579,
"learning_rate": 1.9375e-05,
"loss": 7.5696,
"step": 34
},
{
"epoch": 5.77319587628866,
"grad_norm": 3.256072998046875,
"learning_rate": 1.90625e-05,
"loss": 7.5285,
"step": 35
},
{
"epoch": 5.938144329896907,
"grad_norm": 3.026923179626465,
"learning_rate": 1.8750000000000002e-05,
"loss": 7.3408,
"step": 36
},
{
"epoch": 6.103092783505154,
"grad_norm": 2.9930453300476074,
"learning_rate": 1.84375e-05,
"loss": 7.2398,
"step": 37
},
{
"epoch": 6.268041237113402,
"grad_norm": 2.6270177364349365,
"learning_rate": 1.8125e-05,
"loss": 7.2282,
"step": 38
},
{
"epoch": 6.43298969072165,
"grad_norm": 2.583390474319458,
"learning_rate": 1.78125e-05,
"loss": 7.2528,
"step": 39
},
{
"epoch": 6.597938144329897,
"grad_norm": 2.5908498764038086,
"learning_rate": 1.7500000000000002e-05,
"loss": 7.2968,
"step": 40
},
{
"epoch": 6.762886597938144,
"grad_norm": 2.725625991821289,
"learning_rate": 1.7187499999999998e-05,
"loss": 7.4103,
"step": 41
},
{
"epoch": 6.927835051546392,
"grad_norm": 2.520019769668579,
"learning_rate": 1.6875e-05,
"loss": 7.2359,
"step": 42
},
{
"epoch": 7.092783505154639,
"grad_norm": 2.3079137802124023,
"learning_rate": 1.6562500000000003e-05,
"loss": 7.0846,
"step": 43
},
{
"epoch": 7.257731958762887,
"grad_norm": 2.5834431648254395,
"learning_rate": 1.625e-05,
"loss": 6.9889,
"step": 44
},
{
"epoch": 7.422680412371134,
"grad_norm": 2.618986129760742,
"learning_rate": 1.59375e-05,
"loss": 7.2963,
"step": 45
},
{
"epoch": 7.587628865979381,
"grad_norm": 2.395085573196411,
"learning_rate": 1.5625e-05,
"loss": 7.1191,
"step": 46
},
{
"epoch": 7.752577319587629,
"grad_norm": 2.3549487590789795,
"learning_rate": 1.53125e-05,
"loss": 6.7482,
"step": 47
},
{
"epoch": 7.917525773195877,
"grad_norm": 2.349393606185913,
"learning_rate": 1.5e-05,
"loss": 7.0581,
"step": 48
},
{
"epoch": 8.082474226804123,
"grad_norm": 2.2347733974456787,
"learning_rate": 1.46875e-05,
"loss": 6.8346,
"step": 49
},
{
"epoch": 8.24742268041237,
"grad_norm": 2.182095527648926,
"learning_rate": 1.4375e-05,
"loss": 7.0005,
"step": 50
},
{
"epoch": 8.412371134020619,
"grad_norm": 2.0510122776031494,
"learning_rate": 1.40625e-05,
"loss": 6.8881,
"step": 51
},
{
"epoch": 8.577319587628866,
"grad_norm": 1.9954293966293335,
"learning_rate": 1.375e-05,
"loss": 6.7836,
"step": 52
},
{
"epoch": 8.742268041237114,
"grad_norm": 1.9961941242218018,
"learning_rate": 1.3437500000000001e-05,
"loss": 6.6969,
"step": 53
},
{
"epoch": 8.907216494845361,
"grad_norm": 2.0308048725128174,
"learning_rate": 1.3125e-05,
"loss": 6.9004,
"step": 54
},
{
"epoch": 9.072164948453608,
"grad_norm": 1.9732003211975098,
"learning_rate": 1.28125e-05,
"loss": 6.7652,
"step": 55
},
{
"epoch": 9.237113402061855,
"grad_norm": 1.6928937435150146,
"learning_rate": 1.25e-05,
"loss": 6.6789,
"step": 56
},
{
"epoch": 9.402061855670103,
"grad_norm": 1.9639475345611572,
"learning_rate": 1.21875e-05,
"loss": 6.6207,
"step": 57
},
{
"epoch": 9.56701030927835,
"grad_norm": 1.6721774339675903,
"learning_rate": 1.1874999999999999e-05,
"loss": 6.7687,
"step": 58
},
{
"epoch": 9.731958762886597,
"grad_norm": 1.8212580680847168,
"learning_rate": 1.1562500000000002e-05,
"loss": 6.5534,
"step": 59
},
{
"epoch": 9.896907216494846,
"grad_norm": 1.6735063791275024,
"learning_rate": 1.125e-05,
"loss": 6.6807,
"step": 60
},
{
"epoch": 9.896907216494846,
"eval_audio_cosine_sim": 0.6627817749977112,
"eval_loss": 3.2311527729034424,
"eval_runtime": 2160.5029,
"eval_samples_per_second": 0.006,
"eval_steps_per_second": 0.006,
"eval_text_cosine_sim": 0.35343077778816223,
"step": 60
},
{
"epoch": 10.061855670103093,
"grad_norm": 1.6333427429199219,
"learning_rate": 1.09375e-05,
"loss": 6.6808,
"step": 61
},
{
"epoch": 10.22680412371134,
"grad_norm": 1.5876134634017944,
"learning_rate": 1.0625e-05,
"loss": 6.6629,
"step": 62
},
{
"epoch": 10.391752577319588,
"grad_norm": 1.4894704818725586,
"learning_rate": 1.03125e-05,
"loss": 6.6252,
"step": 63
},
{
"epoch": 10.556701030927835,
"grad_norm": 1.5303210020065308,
"learning_rate": 9.999999999999999e-06,
"loss": 6.4798,
"step": 64
},
{
"epoch": 10.721649484536082,
"grad_norm": 1.3529256582260132,
"learning_rate": 9.6875e-06,
"loss": 6.5828,
"step": 65
},
{
"epoch": 10.88659793814433,
"grad_norm": 1.4958022832870483,
"learning_rate": 9.375000000000001e-06,
"loss": 6.5528,
"step": 66
},
{
"epoch": 11.051546391752577,
"grad_norm": 1.6871830224990845,
"learning_rate": 9.0625e-06,
"loss": 6.3689,
"step": 67
},
{
"epoch": 11.216494845360824,
"grad_norm": 1.5430212020874023,
"learning_rate": 8.750000000000001e-06,
"loss": 6.3102,
"step": 68
},
{
"epoch": 11.381443298969073,
"grad_norm": 1.4388532638549805,
"learning_rate": 8.4375e-06,
"loss": 6.4015,
"step": 69
},
{
"epoch": 11.54639175257732,
"grad_norm": 1.7323001623153687,
"learning_rate": 8.125e-06,
"loss": 6.7011,
"step": 70
},
{
"epoch": 11.711340206185566,
"grad_norm": 1.4382604360580444,
"learning_rate": 7.8125e-06,
"loss": 6.3432,
"step": 71
},
{
"epoch": 11.876288659793815,
"grad_norm": 1.323933720588684,
"learning_rate": 7.5e-06,
"loss": 6.5288,
"step": 72
},
{
"epoch": 12.041237113402062,
"grad_norm": 1.3559449911117554,
"learning_rate": 7.1875e-06,
"loss": 6.557,
"step": 73
},
{
"epoch": 12.206185567010309,
"grad_norm": 1.1633719205856323,
"learning_rate": 6.875e-06,
"loss": 6.386,
"step": 74
},
{
"epoch": 12.371134020618557,
"grad_norm": 1.5071109533309937,
"learning_rate": 6.5625e-06,
"loss": 6.5438,
"step": 75
},
{
"epoch": 12.536082474226804,
"grad_norm": 1.1435602903366089,
"learning_rate": 6.25e-06,
"loss": 6.4093,
"step": 76
},
{
"epoch": 12.70103092783505,
"grad_norm": 1.4174741506576538,
"learning_rate": 5.9374999999999995e-06,
"loss": 6.3887,
"step": 77
},
{
"epoch": 12.8659793814433,
"grad_norm": 1.423195481300354,
"learning_rate": 5.625e-06,
"loss": 6.5358,
"step": 78
},
{
"epoch": 13.030927835051546,
"grad_norm": 1.2214951515197754,
"learning_rate": 5.3125e-06,
"loss": 6.4307,
"step": 79
},
{
"epoch": 13.195876288659793,
"grad_norm": 1.1089880466461182,
"learning_rate": 4.9999999999999996e-06,
"loss": 6.4179,
"step": 80
},
{
"epoch": 13.360824742268042,
"grad_norm": 1.0945030450820923,
"learning_rate": 4.6875000000000004e-06,
"loss": 6.3149,
"step": 81
},
{
"epoch": 13.525773195876289,
"grad_norm": 1.4375858306884766,
"learning_rate": 4.3750000000000005e-06,
"loss": 6.6049,
"step": 82
},
{
"epoch": 13.690721649484535,
"grad_norm": 1.2144274711608887,
"learning_rate": 4.0625e-06,
"loss": 6.3501,
"step": 83
},
{
"epoch": 13.855670103092784,
"grad_norm": 1.091561198234558,
"learning_rate": 3.75e-06,
"loss": 6.4059,
"step": 84
},
{
"epoch": 14.02061855670103,
"grad_norm": 1.4113916158676147,
"learning_rate": 3.4375e-06,
"loss": 6.1656,
"step": 85
},
{
"epoch": 14.185567010309278,
"grad_norm": 1.275417685508728,
"learning_rate": 3.125e-06,
"loss": 6.2095,
"step": 86
},
{
"epoch": 14.350515463917526,
"grad_norm": 1.1587275266647339,
"learning_rate": 2.8125e-06,
"loss": 6.4885,
"step": 87
},
{
"epoch": 14.515463917525773,
"grad_norm": 1.0685796737670898,
"learning_rate": 2.4999999999999998e-06,
"loss": 6.3991,
"step": 88
},
{
"epoch": 14.68041237113402,
"grad_norm": 1.102617859840393,
"learning_rate": 2.1875000000000002e-06,
"loss": 6.4392,
"step": 89
},
{
"epoch": 14.845360824742269,
"grad_norm": 1.0771472454071045,
"learning_rate": 1.875e-06,
"loss": 6.4444,
"step": 90
},
{
"epoch": 14.845360824742269,
"eval_audio_cosine_sim": 0.5883382558822632,
"eval_loss": 3.6276748180389404,
"eval_runtime": 1072.6246,
"eval_samples_per_second": 0.011,
"eval_steps_per_second": 0.011,
"eval_text_cosine_sim": 0.24592441320419312,
"step": 90
},
{
"epoch": 15.010309278350515,
"grad_norm": 1.2423425912857056,
"learning_rate": 1.5625e-06,
"loss": 6.2888,
"step": 91
},
{
"epoch": 15.175257731958762,
"grad_norm": 1.1462149620056152,
"learning_rate": 1.2499999999999999e-06,
"loss": 6.4798,
"step": 92
},
{
"epoch": 15.34020618556701,
"grad_norm": 1.1183279752731323,
"learning_rate": 9.375e-07,
"loss": 6.254,
"step": 93
},
{
"epoch": 15.505154639175258,
"grad_norm": 1.1249028444290161,
"learning_rate": 6.249999999999999e-07,
"loss": 6.3222,
"step": 94
},
{
"epoch": 15.670103092783505,
"grad_norm": 1.2026952505111694,
"learning_rate": 3.1249999999999997e-07,
"loss": 6.2651,
"step": 95
},
{
"epoch": 15.835051546391753,
"grad_norm": 1.283206820487976,
"learning_rate": 0.0,
"loss": 6.553,
"step": 96
},
{
"epoch": 15.835051546391753,
"step": 96,
"total_flos": 1925884422732936.0,
"train_loss": 7.460742597778638,
"train_runtime": 13364.4334,
"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
}