Wongkraiwich Chuenchomphu
add model deps
8a5d350
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 75,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"grad_norm": 20.00289492370743,
"learning_rate": 0.0,
"loss": 1.5756,
"step": 1
},
{
"epoch": 0.08,
"grad_norm": 169.81482789708753,
"learning_rate": 5e-05,
"loss": 14.3496,
"step": 2
},
{
"epoch": 0.12,
"grad_norm": 170.94221266882175,
"learning_rate": 7.924812503605781e-05,
"loss": 14.3495,
"step": 3
},
{
"epoch": 0.16,
"grad_norm": 96.07266456886373,
"learning_rate": 0.0001,
"loss": 5.728,
"step": 4
},
{
"epoch": 0.2,
"grad_norm": 74.89495923839695,
"learning_rate": 0.0001,
"loss": 3.198,
"step": 5
},
{
"epoch": 0.24,
"grad_norm": 16.06625851910231,
"learning_rate": 9.859154929577466e-05,
"loss": 1.711,
"step": 6
},
{
"epoch": 0.28,
"grad_norm": 11.432322037094815,
"learning_rate": 9.718309859154931e-05,
"loss": 1.1097,
"step": 7
},
{
"epoch": 0.32,
"grad_norm": 10.90988949596415,
"learning_rate": 9.577464788732394e-05,
"loss": 1.1647,
"step": 8
},
{
"epoch": 0.36,
"grad_norm": 5.166602608719267,
"learning_rate": 9.43661971830986e-05,
"loss": 0.6874,
"step": 9
},
{
"epoch": 0.4,
"grad_norm": 5.063308573446352,
"learning_rate": 9.295774647887325e-05,
"loss": 0.6268,
"step": 10
},
{
"epoch": 0.44,
"grad_norm": 7.227736963113414,
"learning_rate": 9.15492957746479e-05,
"loss": 0.5151,
"step": 11
},
{
"epoch": 0.48,
"grad_norm": 4.259327811156309,
"learning_rate": 9.014084507042254e-05,
"loss": 0.4389,
"step": 12
},
{
"epoch": 0.52,
"grad_norm": 6.156983867098677,
"learning_rate": 8.873239436619719e-05,
"loss": 0.5211,
"step": 13
},
{
"epoch": 0.56,
"grad_norm": 5.241850109217824,
"learning_rate": 8.732394366197182e-05,
"loss": 0.3401,
"step": 14
},
{
"epoch": 0.6,
"grad_norm": 45.75950192245342,
"learning_rate": 8.591549295774647e-05,
"loss": 0.6114,
"step": 15
},
{
"epoch": 0.64,
"grad_norm": 12.747985157681715,
"learning_rate": 8.450704225352113e-05,
"loss": 0.4109,
"step": 16
},
{
"epoch": 0.68,
"grad_norm": 6.204387028637272,
"learning_rate": 8.309859154929578e-05,
"loss": 0.3238,
"step": 17
},
{
"epoch": 0.72,
"grad_norm": 1.784294966675183,
"learning_rate": 8.169014084507043e-05,
"loss": 0.2746,
"step": 18
},
{
"epoch": 0.76,
"grad_norm": 1.0640451461171732,
"learning_rate": 8.028169014084508e-05,
"loss": 0.2216,
"step": 19
},
{
"epoch": 0.8,
"grad_norm": 2.247533777438539,
"learning_rate": 7.887323943661972e-05,
"loss": 0.2169,
"step": 20
},
{
"epoch": 0.84,
"grad_norm": 0.5843446520161867,
"learning_rate": 7.746478873239437e-05,
"loss": 0.2063,
"step": 21
},
{
"epoch": 0.88,
"grad_norm": 0.6140022871361251,
"learning_rate": 7.605633802816902e-05,
"loss": 0.1924,
"step": 22
},
{
"epoch": 0.92,
"grad_norm": 0.6328264593401415,
"learning_rate": 7.464788732394367e-05,
"loss": 0.1687,
"step": 23
},
{
"epoch": 0.96,
"grad_norm": 0.47188590510134204,
"learning_rate": 7.323943661971832e-05,
"loss": 0.1502,
"step": 24
},
{
"epoch": 1.0,
"grad_norm": 0.4569187747764233,
"learning_rate": 7.183098591549297e-05,
"loss": 0.1322,
"step": 25
},
{
"epoch": 1.04,
"grad_norm": 0.5022295511205257,
"learning_rate": 7.042253521126761e-05,
"loss": 0.1283,
"step": 26
},
{
"epoch": 1.08,
"grad_norm": 0.32747245962381083,
"learning_rate": 6.901408450704226e-05,
"loss": 0.105,
"step": 27
},
{
"epoch": 1.12,
"grad_norm": 0.4071591561018996,
"learning_rate": 6.76056338028169e-05,
"loss": 0.1127,
"step": 28
},
{
"epoch": 1.16,
"grad_norm": 0.4886479527811691,
"learning_rate": 6.619718309859155e-05,
"loss": 0.1097,
"step": 29
},
{
"epoch": 1.2,
"grad_norm": 0.34862267892547183,
"learning_rate": 6.47887323943662e-05,
"loss": 0.1007,
"step": 30
},
{
"epoch": 1.24,
"grad_norm": 0.33123650236209157,
"learning_rate": 6.338028169014085e-05,
"loss": 0.0945,
"step": 31
},
{
"epoch": 1.28,
"grad_norm": 0.3246463353672584,
"learning_rate": 6.197183098591549e-05,
"loss": 0.0957,
"step": 32
},
{
"epoch": 1.32,
"grad_norm": 0.3503498413159293,
"learning_rate": 6.056338028169014e-05,
"loss": 0.1007,
"step": 33
},
{
"epoch": 1.3599999999999999,
"grad_norm": 0.2994313557882548,
"learning_rate": 5.915492957746479e-05,
"loss": 0.0961,
"step": 34
},
{
"epoch": 1.4,
"grad_norm": 0.3098559906263102,
"learning_rate": 5.774647887323944e-05,
"loss": 0.0891,
"step": 35
},
{
"epoch": 1.44,
"grad_norm": 0.3320370344139516,
"learning_rate": 5.633802816901409e-05,
"loss": 0.093,
"step": 36
},
{
"epoch": 1.48,
"grad_norm": 0.2975761075734135,
"learning_rate": 5.492957746478874e-05,
"loss": 0.0824,
"step": 37
},
{
"epoch": 1.52,
"grad_norm": 0.2609696051747955,
"learning_rate": 5.352112676056338e-05,
"loss": 0.0849,
"step": 38
},
{
"epoch": 1.56,
"grad_norm": 0.2746218271104525,
"learning_rate": 5.2112676056338026e-05,
"loss": 0.0812,
"step": 39
},
{
"epoch": 1.6,
"grad_norm": 0.27280756684282476,
"learning_rate": 5.070422535211268e-05,
"loss": 0.0857,
"step": 40
},
{
"epoch": 1.6400000000000001,
"grad_norm": 0.3333732319784279,
"learning_rate": 4.929577464788733e-05,
"loss": 0.083,
"step": 41
},
{
"epoch": 1.6800000000000002,
"grad_norm": 0.28187480716143415,
"learning_rate": 4.788732394366197e-05,
"loss": 0.0805,
"step": 42
},
{
"epoch": 1.72,
"grad_norm": 0.25874624247743283,
"learning_rate": 4.647887323943662e-05,
"loss": 0.0806,
"step": 43
},
{
"epoch": 1.76,
"grad_norm": 0.28417848530718226,
"learning_rate": 4.507042253521127e-05,
"loss": 0.0776,
"step": 44
},
{
"epoch": 1.8,
"grad_norm": 0.24698649624452082,
"learning_rate": 4.366197183098591e-05,
"loss": 0.0765,
"step": 45
},
{
"epoch": 1.8399999999999999,
"grad_norm": 0.23537664520973256,
"learning_rate": 4.225352112676056e-05,
"loss": 0.0758,
"step": 46
},
{
"epoch": 1.88,
"grad_norm": 0.2534413438607467,
"learning_rate": 4.0845070422535214e-05,
"loss": 0.0804,
"step": 47
},
{
"epoch": 1.92,
"grad_norm": 0.20702488497520458,
"learning_rate": 3.943661971830986e-05,
"loss": 0.0715,
"step": 48
},
{
"epoch": 1.96,
"grad_norm": 0.2613262372736133,
"learning_rate": 3.802816901408451e-05,
"loss": 0.084,
"step": 49
},
{
"epoch": 2.0,
"grad_norm": 0.22492435603929029,
"learning_rate": 3.661971830985916e-05,
"loss": 0.0556,
"step": 50
},
{
"epoch": 2.04,
"grad_norm": 0.2611962185259755,
"learning_rate": 3.5211267605633805e-05,
"loss": 0.0481,
"step": 51
},
{
"epoch": 2.08,
"grad_norm": 0.15962947759230994,
"learning_rate": 3.380281690140845e-05,
"loss": 0.0449,
"step": 52
},
{
"epoch": 2.12,
"grad_norm": 0.1694677846392973,
"learning_rate": 3.23943661971831e-05,
"loss": 0.0482,
"step": 53
},
{
"epoch": 2.16,
"grad_norm": 0.1438746814861677,
"learning_rate": 3.0985915492957744e-05,
"loss": 0.0425,
"step": 54
},
{
"epoch": 2.2,
"grad_norm": 0.17430208968882352,
"learning_rate": 2.9577464788732395e-05,
"loss": 0.043,
"step": 55
},
{
"epoch": 2.24,
"grad_norm": 0.17207620385513733,
"learning_rate": 2.8169014084507046e-05,
"loss": 0.0449,
"step": 56
},
{
"epoch": 2.2800000000000002,
"grad_norm": 0.18001842523178951,
"learning_rate": 2.676056338028169e-05,
"loss": 0.0407,
"step": 57
},
{
"epoch": 2.32,
"grad_norm": 0.20877963876068653,
"learning_rate": 2.535211267605634e-05,
"loss": 0.0418,
"step": 58
},
{
"epoch": 2.36,
"grad_norm": 0.2086206476860082,
"learning_rate": 2.3943661971830986e-05,
"loss": 0.0432,
"step": 59
},
{
"epoch": 2.4,
"grad_norm": 0.18489125316064906,
"learning_rate": 2.2535211267605634e-05,
"loss": 0.0443,
"step": 60
},
{
"epoch": 2.44,
"grad_norm": 0.18836568321121092,
"learning_rate": 2.112676056338028e-05,
"loss": 0.0416,
"step": 61
},
{
"epoch": 2.48,
"grad_norm": 0.18671273407693323,
"learning_rate": 1.971830985915493e-05,
"loss": 0.0392,
"step": 62
},
{
"epoch": 2.52,
"grad_norm": 0.183141175658178,
"learning_rate": 1.830985915492958e-05,
"loss": 0.0406,
"step": 63
},
{
"epoch": 2.56,
"grad_norm": 0.19023565118928507,
"learning_rate": 1.6901408450704224e-05,
"loss": 0.0392,
"step": 64
},
{
"epoch": 2.6,
"grad_norm": 0.1734836339026565,
"learning_rate": 1.5492957746478872e-05,
"loss": 0.0408,
"step": 65
},
{
"epoch": 2.64,
"grad_norm": 0.2124529691628785,
"learning_rate": 1.4084507042253523e-05,
"loss": 0.0402,
"step": 66
},
{
"epoch": 2.68,
"grad_norm": 0.16929391483607104,
"learning_rate": 1.267605633802817e-05,
"loss": 0.0422,
"step": 67
},
{
"epoch": 2.7199999999999998,
"grad_norm": 0.1865969813928253,
"learning_rate": 1.1267605633802817e-05,
"loss": 0.0423,
"step": 68
},
{
"epoch": 2.76,
"grad_norm": 0.19199511508921327,
"learning_rate": 9.859154929577465e-06,
"loss": 0.0428,
"step": 69
},
{
"epoch": 2.8,
"grad_norm": 0.15471843235506497,
"learning_rate": 8.450704225352112e-06,
"loss": 0.0383,
"step": 70
},
{
"epoch": 2.84,
"grad_norm": 0.17211726924185472,
"learning_rate": 7.042253521126762e-06,
"loss": 0.0422,
"step": 71
},
{
"epoch": 2.88,
"grad_norm": 0.16597891624795819,
"learning_rate": 5.6338028169014084e-06,
"loss": 0.0358,
"step": 72
},
{
"epoch": 2.92,
"grad_norm": 0.16696499746864643,
"learning_rate": 4.225352112676056e-06,
"loss": 0.0384,
"step": 73
},
{
"epoch": 2.96,
"grad_norm": 0.16841918706948866,
"learning_rate": 2.8169014084507042e-06,
"loss": 0.0399,
"step": 74
},
{
"epoch": 3.0,
"grad_norm": 0.12380367991746002,
"learning_rate": 1.4084507042253521e-06,
"loss": 0.0275,
"step": 75
},
{
"epoch": 3.0,
"step": 75,
"total_flos": 175054227243008.0,
"train_loss": 0.6997585766762495,
"train_runtime": 937.3587,
"train_samples_per_second": 103.388,
"train_steps_per_second": 0.08
}
],
"logging_steps": 1.0,
"max_steps": 75,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 175054227243008.0,
"train_batch_size": 12,
"trial_name": null,
"trial_params": null
}