vdos's picture
Training in progress, step 50, checkpoint
ddf6a07 verified
{
"best_metric": NaN,
"best_model_checkpoint": "miner_id_24/checkpoint-25",
"epoch": 0.6024096385542169,
"eval_steps": 25,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.012048192771084338,
"grad_norm": 9.409144401550293,
"learning_rate": 5e-05,
"loss": 3.0901,
"step": 1
},
{
"epoch": 0.012048192771084338,
"eval_loss": NaN,
"eval_runtime": 29.0345,
"eval_samples_per_second": 19.253,
"eval_steps_per_second": 2.411,
"step": 1
},
{
"epoch": 0.024096385542168676,
"grad_norm": 10.093705177307129,
"learning_rate": 0.0001,
"loss": 3.807,
"step": 2
},
{
"epoch": 0.03614457831325301,
"grad_norm": 8.005189895629883,
"learning_rate": 9.989294616193017e-05,
"loss": 2.8297,
"step": 3
},
{
"epoch": 0.04819277108433735,
"grad_norm": 4.99748420715332,
"learning_rate": 9.957224306869053e-05,
"loss": 1.8837,
"step": 4
},
{
"epoch": 0.060240963855421686,
"grad_norm": 3.583455801010132,
"learning_rate": 9.903926402016153e-05,
"loss": 1.4351,
"step": 5
},
{
"epoch": 0.07228915662650602,
"grad_norm": 3.411550998687744,
"learning_rate": 9.829629131445342e-05,
"loss": 1.0929,
"step": 6
},
{
"epoch": 0.08433734939759036,
"grad_norm": 3.292454719543457,
"learning_rate": 9.73465064747553e-05,
"loss": 0.9362,
"step": 7
},
{
"epoch": 0.0963855421686747,
"grad_norm": 2.2479372024536133,
"learning_rate": 9.619397662556435e-05,
"loss": 0.8141,
"step": 8
},
{
"epoch": 0.10843373493975904,
"grad_norm": 1.9870944023132324,
"learning_rate": 9.484363707663442e-05,
"loss": 0.735,
"step": 9
},
{
"epoch": 0.12048192771084337,
"grad_norm": 2.4433305263519287,
"learning_rate": 9.330127018922194e-05,
"loss": 0.7608,
"step": 10
},
{
"epoch": 0.13253012048192772,
"grad_norm": 6.735482215881348,
"learning_rate": 9.157348061512727e-05,
"loss": 1.0888,
"step": 11
},
{
"epoch": 0.14457831325301204,
"grad_norm": 6.955150604248047,
"learning_rate": 8.966766701456177e-05,
"loss": 0.8006,
"step": 12
},
{
"epoch": 0.1566265060240964,
"grad_norm": 16.755565643310547,
"learning_rate": 8.759199037394887e-05,
"loss": 1.8009,
"step": 13
},
{
"epoch": 0.1686746987951807,
"grad_norm": 12.474592208862305,
"learning_rate": 8.535533905932738e-05,
"loss": 1.6742,
"step": 14
},
{
"epoch": 0.18072289156626506,
"grad_norm": 8.603779792785645,
"learning_rate": 8.296729075500344e-05,
"loss": 1.3056,
"step": 15
},
{
"epoch": 0.1927710843373494,
"grad_norm": 6.018673419952393,
"learning_rate": 8.043807145043604e-05,
"loss": 1.0944,
"step": 16
},
{
"epoch": 0.20481927710843373,
"grad_norm": 3.591155767440796,
"learning_rate": 7.777851165098012e-05,
"loss": 0.9134,
"step": 17
},
{
"epoch": 0.21686746987951808,
"grad_norm": 1.9267585277557373,
"learning_rate": 7.500000000000001e-05,
"loss": 0.7401,
"step": 18
},
{
"epoch": 0.2289156626506024,
"grad_norm": 1.4407517910003662,
"learning_rate": 7.211443451095007e-05,
"loss": 0.7104,
"step": 19
},
{
"epoch": 0.24096385542168675,
"grad_norm": 1.3494484424591064,
"learning_rate": 6.91341716182545e-05,
"loss": 0.6164,
"step": 20
},
{
"epoch": 0.25301204819277107,
"grad_norm": 1.386812448501587,
"learning_rate": 6.607197326515808e-05,
"loss": 0.5907,
"step": 21
},
{
"epoch": 0.26506024096385544,
"grad_norm": 1.629399061203003,
"learning_rate": 6.294095225512603e-05,
"loss": 0.6211,
"step": 22
},
{
"epoch": 0.27710843373493976,
"grad_norm": 4.120821952819824,
"learning_rate": 5.9754516100806423e-05,
"loss": 0.9784,
"step": 23
},
{
"epoch": 0.2891566265060241,
"grad_norm": 7.409047603607178,
"learning_rate": 5.6526309611002594e-05,
"loss": 1.0961,
"step": 24
},
{
"epoch": 0.30120481927710846,
"grad_norm": 8.13595962524414,
"learning_rate": 5.327015646150716e-05,
"loss": 1.0316,
"step": 25
},
{
"epoch": 0.30120481927710846,
"eval_loss": NaN,
"eval_runtime": 29.0045,
"eval_samples_per_second": 19.273,
"eval_steps_per_second": 2.413,
"step": 25
},
{
"epoch": 0.3132530120481928,
"grad_norm": 3.152708053588867,
"learning_rate": 5e-05,
"loss": 0.7848,
"step": 26
},
{
"epoch": 0.3253012048192771,
"grad_norm": 2.7546987533569336,
"learning_rate": 4.6729843538492847e-05,
"loss": 0.7297,
"step": 27
},
{
"epoch": 0.3373493975903614,
"grad_norm": 2.0920066833496094,
"learning_rate": 4.347369038899744e-05,
"loss": 0.684,
"step": 28
},
{
"epoch": 0.3493975903614458,
"grad_norm": 1.7343143224716187,
"learning_rate": 4.0245483899193595e-05,
"loss": 0.6215,
"step": 29
},
{
"epoch": 0.3614457831325301,
"grad_norm": 1.3055154085159302,
"learning_rate": 3.705904774487396e-05,
"loss": 0.5681,
"step": 30
},
{
"epoch": 0.37349397590361444,
"grad_norm": 1.1538188457489014,
"learning_rate": 3.392802673484193e-05,
"loss": 0.56,
"step": 31
},
{
"epoch": 0.3855421686746988,
"grad_norm": 1.1171661615371704,
"learning_rate": 3.086582838174551e-05,
"loss": 0.507,
"step": 32
},
{
"epoch": 0.39759036144578314,
"grad_norm": 1.0048093795776367,
"learning_rate": 2.7885565489049946e-05,
"loss": 0.5333,
"step": 33
},
{
"epoch": 0.40963855421686746,
"grad_norm": 1.101479411125183,
"learning_rate": 2.500000000000001e-05,
"loss": 0.4368,
"step": 34
},
{
"epoch": 0.42168674698795183,
"grad_norm": 1.8432483673095703,
"learning_rate": 2.2221488349019903e-05,
"loss": 0.6584,
"step": 35
},
{
"epoch": 0.43373493975903615,
"grad_norm": 4.359800815582275,
"learning_rate": 1.9561928549563968e-05,
"loss": 0.9682,
"step": 36
},
{
"epoch": 0.4457831325301205,
"grad_norm": 6.246274948120117,
"learning_rate": 1.703270924499656e-05,
"loss": 0.9427,
"step": 37
},
{
"epoch": 0.4578313253012048,
"grad_norm": 1.7437212467193604,
"learning_rate": 1.4644660940672627e-05,
"loss": 0.616,
"step": 38
},
{
"epoch": 0.46987951807228917,
"grad_norm": 1.6649398803710938,
"learning_rate": 1.2408009626051137e-05,
"loss": 0.6178,
"step": 39
},
{
"epoch": 0.4819277108433735,
"grad_norm": 1.0606410503387451,
"learning_rate": 1.0332332985438248e-05,
"loss": 0.5314,
"step": 40
},
{
"epoch": 0.4939759036144578,
"grad_norm": 0.9042273759841919,
"learning_rate": 8.426519384872733e-06,
"loss": 0.4602,
"step": 41
},
{
"epoch": 0.5060240963855421,
"grad_norm": 0.9936925768852234,
"learning_rate": 6.698729810778065e-06,
"loss": 0.5254,
"step": 42
},
{
"epoch": 0.5180722891566265,
"grad_norm": 0.9050127267837524,
"learning_rate": 5.156362923365588e-06,
"loss": 0.4582,
"step": 43
},
{
"epoch": 0.5301204819277109,
"grad_norm": 0.9362257122993469,
"learning_rate": 3.8060233744356633e-06,
"loss": 0.5136,
"step": 44
},
{
"epoch": 0.5421686746987951,
"grad_norm": 0.8541726469993591,
"learning_rate": 2.653493525244721e-06,
"loss": 0.4507,
"step": 45
},
{
"epoch": 0.5542168674698795,
"grad_norm": 0.9145020842552185,
"learning_rate": 1.70370868554659e-06,
"loss": 0.4414,
"step": 46
},
{
"epoch": 0.5662650602409639,
"grad_norm": 1.0000098943710327,
"learning_rate": 9.607359798384785e-07,
"loss": 0.4481,
"step": 47
},
{
"epoch": 0.5783132530120482,
"grad_norm": 2.1228537559509277,
"learning_rate": 4.277569313094809e-07,
"loss": 0.748,
"step": 48
},
{
"epoch": 0.5903614457831325,
"grad_norm": 3.8465590476989746,
"learning_rate": 1.0705383806982606e-07,
"loss": 0.7558,
"step": 49
},
{
"epoch": 0.6024096385542169,
"grad_norm": 5.656222343444824,
"learning_rate": 0.0,
"loss": 0.797,
"step": 50
},
{
"epoch": 0.6024096385542169,
"eval_loss": NaN,
"eval_runtime": 28.9964,
"eval_samples_per_second": 19.278,
"eval_steps_per_second": 2.414,
"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": 1
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 6.631381723512832e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}