vdos's picture
Training in progress, step 50, checkpoint
1b50d2d verified
{
"best_metric": 3.673874616622925,
"best_model_checkpoint": "miner_id_24/checkpoint-50",
"epoch": 0.21534320323014805,
"eval_steps": 25,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.004306864064602961,
"grad_norm": 8.360955238342285,
"learning_rate": 5e-05,
"loss": 5.9843,
"step": 1
},
{
"epoch": 0.004306864064602961,
"eval_loss": 6.015687465667725,
"eval_runtime": 51.9556,
"eval_samples_per_second": 30.122,
"eval_steps_per_second": 3.772,
"step": 1
},
{
"epoch": 0.008613728129205921,
"grad_norm": 8.56803035736084,
"learning_rate": 0.0001,
"loss": 6.0379,
"step": 2
},
{
"epoch": 0.012920592193808884,
"grad_norm": 6.061903476715088,
"learning_rate": 9.989294616193017e-05,
"loss": 5.3358,
"step": 3
},
{
"epoch": 0.017227456258411843,
"grad_norm": 6.505764961242676,
"learning_rate": 9.957224306869053e-05,
"loss": 4.7219,
"step": 4
},
{
"epoch": 0.021534320323014805,
"grad_norm": 6.317793846130371,
"learning_rate": 9.903926402016153e-05,
"loss": 4.7709,
"step": 5
},
{
"epoch": 0.025841184387617767,
"grad_norm": 3.1196484565734863,
"learning_rate": 9.829629131445342e-05,
"loss": 4.4309,
"step": 6
},
{
"epoch": 0.030148048452220726,
"grad_norm": 2.059274435043335,
"learning_rate": 9.73465064747553e-05,
"loss": 4.2994,
"step": 7
},
{
"epoch": 0.034454912516823685,
"grad_norm": 1.7302476167678833,
"learning_rate": 9.619397662556435e-05,
"loss": 4.266,
"step": 8
},
{
"epoch": 0.03876177658142665,
"grad_norm": 2.33834171295166,
"learning_rate": 9.484363707663442e-05,
"loss": 4.1851,
"step": 9
},
{
"epoch": 0.04306864064602961,
"grad_norm": 1.8448387384414673,
"learning_rate": 9.330127018922194e-05,
"loss": 4.1023,
"step": 10
},
{
"epoch": 0.04737550471063257,
"grad_norm": 2.575143575668335,
"learning_rate": 9.157348061512727e-05,
"loss": 4.0886,
"step": 11
},
{
"epoch": 0.051682368775235535,
"grad_norm": 3.2730138301849365,
"learning_rate": 8.966766701456177e-05,
"loss": 4.2101,
"step": 12
},
{
"epoch": 0.05598923283983849,
"grad_norm": 1.706132411956787,
"learning_rate": 8.759199037394887e-05,
"loss": 4.0739,
"step": 13
},
{
"epoch": 0.06029609690444145,
"grad_norm": 2.7032978534698486,
"learning_rate": 8.535533905932738e-05,
"loss": 4.0411,
"step": 14
},
{
"epoch": 0.06460296096904442,
"grad_norm": 1.7852596044540405,
"learning_rate": 8.296729075500344e-05,
"loss": 3.9913,
"step": 15
},
{
"epoch": 0.06890982503364737,
"grad_norm": 1.4892394542694092,
"learning_rate": 8.043807145043604e-05,
"loss": 3.8523,
"step": 16
},
{
"epoch": 0.07321668909825034,
"grad_norm": 1.778761625289917,
"learning_rate": 7.777851165098012e-05,
"loss": 3.9131,
"step": 17
},
{
"epoch": 0.0775235531628533,
"grad_norm": 1.0222779512405396,
"learning_rate": 7.500000000000001e-05,
"loss": 3.7881,
"step": 18
},
{
"epoch": 0.08183041722745626,
"grad_norm": 1.5180976390838623,
"learning_rate": 7.211443451095007e-05,
"loss": 3.7586,
"step": 19
},
{
"epoch": 0.08613728129205922,
"grad_norm": 1.0901626348495483,
"learning_rate": 6.91341716182545e-05,
"loss": 3.7536,
"step": 20
},
{
"epoch": 0.09044414535666218,
"grad_norm": 1.2611289024353027,
"learning_rate": 6.607197326515808e-05,
"loss": 3.7946,
"step": 21
},
{
"epoch": 0.09475100942126514,
"grad_norm": 1.0649021863937378,
"learning_rate": 6.294095225512603e-05,
"loss": 3.7712,
"step": 22
},
{
"epoch": 0.0990578734858681,
"grad_norm": 1.4094829559326172,
"learning_rate": 5.9754516100806423e-05,
"loss": 3.8244,
"step": 23
},
{
"epoch": 0.10336473755047107,
"grad_norm": 1.4050860404968262,
"learning_rate": 5.6526309611002594e-05,
"loss": 3.829,
"step": 24
},
{
"epoch": 0.10767160161507403,
"grad_norm": 1.3968360424041748,
"learning_rate": 5.327015646150716e-05,
"loss": 4.0147,
"step": 25
},
{
"epoch": 0.10767160161507403,
"eval_loss": 3.7597262859344482,
"eval_runtime": 52.1327,
"eval_samples_per_second": 30.02,
"eval_steps_per_second": 3.76,
"step": 25
},
{
"epoch": 0.11197846567967698,
"grad_norm": 1.8688342571258545,
"learning_rate": 5e-05,
"loss": 3.8387,
"step": 26
},
{
"epoch": 0.11628532974427995,
"grad_norm": 1.3678120374679565,
"learning_rate": 4.6729843538492847e-05,
"loss": 3.7591,
"step": 27
},
{
"epoch": 0.1205921938088829,
"grad_norm": 1.2320159673690796,
"learning_rate": 4.347369038899744e-05,
"loss": 3.8064,
"step": 28
},
{
"epoch": 0.12489905787348587,
"grad_norm": 1.46751070022583,
"learning_rate": 4.0245483899193595e-05,
"loss": 3.7541,
"step": 29
},
{
"epoch": 0.12920592193808883,
"grad_norm": 1.0058436393737793,
"learning_rate": 3.705904774487396e-05,
"loss": 3.673,
"step": 30
},
{
"epoch": 0.13351278600269179,
"grad_norm": 1.1568927764892578,
"learning_rate": 3.392802673484193e-05,
"loss": 3.6349,
"step": 31
},
{
"epoch": 0.13781965006729474,
"grad_norm": 1.2745643854141235,
"learning_rate": 3.086582838174551e-05,
"loss": 3.669,
"step": 32
},
{
"epoch": 0.14212651413189772,
"grad_norm": 1.0898311138153076,
"learning_rate": 2.7885565489049946e-05,
"loss": 3.6504,
"step": 33
},
{
"epoch": 0.14643337819650068,
"grad_norm": 0.9707027077674866,
"learning_rate": 2.500000000000001e-05,
"loss": 3.6602,
"step": 34
},
{
"epoch": 0.15074024226110364,
"grad_norm": 1.464558482170105,
"learning_rate": 2.2221488349019903e-05,
"loss": 3.6902,
"step": 35
},
{
"epoch": 0.1550471063257066,
"grad_norm": 1.6716910600662231,
"learning_rate": 1.9561928549563968e-05,
"loss": 3.6462,
"step": 36
},
{
"epoch": 0.15935397039030955,
"grad_norm": 1.391884207725525,
"learning_rate": 1.703270924499656e-05,
"loss": 3.7975,
"step": 37
},
{
"epoch": 0.16366083445491253,
"grad_norm": 0.967711329460144,
"learning_rate": 1.4644660940672627e-05,
"loss": 3.7192,
"step": 38
},
{
"epoch": 0.16796769851951548,
"grad_norm": 1.2279342412948608,
"learning_rate": 1.2408009626051137e-05,
"loss": 3.7489,
"step": 39
},
{
"epoch": 0.17227456258411844,
"grad_norm": 1.1709775924682617,
"learning_rate": 1.0332332985438248e-05,
"loss": 3.6942,
"step": 40
},
{
"epoch": 0.1765814266487214,
"grad_norm": 1.0664342641830444,
"learning_rate": 8.426519384872733e-06,
"loss": 3.7223,
"step": 41
},
{
"epoch": 0.18088829071332435,
"grad_norm": 0.8159927129745483,
"learning_rate": 6.698729810778065e-06,
"loss": 3.6485,
"step": 42
},
{
"epoch": 0.18519515477792733,
"grad_norm": 0.8708414435386658,
"learning_rate": 5.156362923365588e-06,
"loss": 3.6189,
"step": 43
},
{
"epoch": 0.1895020188425303,
"grad_norm": 0.8581545948982239,
"learning_rate": 3.8060233744356633e-06,
"loss": 3.6428,
"step": 44
},
{
"epoch": 0.19380888290713325,
"grad_norm": 0.8968494534492493,
"learning_rate": 2.653493525244721e-06,
"loss": 3.6532,
"step": 45
},
{
"epoch": 0.1981157469717362,
"grad_norm": 0.9328946471214294,
"learning_rate": 1.70370868554659e-06,
"loss": 3.6412,
"step": 46
},
{
"epoch": 0.20242261103633916,
"grad_norm": 1.0675426721572876,
"learning_rate": 9.607359798384785e-07,
"loss": 3.546,
"step": 47
},
{
"epoch": 0.20672947510094214,
"grad_norm": 1.1364132165908813,
"learning_rate": 4.277569313094809e-07,
"loss": 3.6322,
"step": 48
},
{
"epoch": 0.2110363391655451,
"grad_norm": 1.1251806020736694,
"learning_rate": 1.0705383806982606e-07,
"loss": 3.6173,
"step": 49
},
{
"epoch": 0.21534320323014805,
"grad_norm": 1.5030646324157715,
"learning_rate": 0.0,
"loss": 3.8635,
"step": 50
},
{
"epoch": 0.21534320323014805,
"eval_loss": 3.673874616622925,
"eval_runtime": 52.0871,
"eval_samples_per_second": 30.046,
"eval_steps_per_second": 3.763,
"step": 50
}
],
"logging_steps": 1,
"max_steps": 50,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 25,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 1,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 6.174479344467968e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}