358 lines
7.9 KiB
JSON
358 lines
7.9 KiB
JSON
{
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.896,
|
|
"eval_steps": 500,
|
|
"global_step": 45,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.064,
|
|
"grad_norm": 5.865464631994252,
|
|
"learning_rate": 1.0000000000000002e-06,
|
|
"loss": 0.8384,
|
|
"step": 1
|
|
},
|
|
{
|
|
"epoch": 0.128,
|
|
"grad_norm": 5.795282414900805,
|
|
"learning_rate": 2.0000000000000003e-06,
|
|
"loss": 0.8522,
|
|
"step": 2
|
|
},
|
|
{
|
|
"epoch": 0.192,
|
|
"grad_norm": 6.004920342985478,
|
|
"learning_rate": 3e-06,
|
|
"loss": 0.8846,
|
|
"step": 3
|
|
},
|
|
{
|
|
"epoch": 0.256,
|
|
"grad_norm": 5.2374685746239775,
|
|
"learning_rate": 4.000000000000001e-06,
|
|
"loss": 0.8557,
|
|
"step": 4
|
|
},
|
|
{
|
|
"epoch": 0.32,
|
|
"grad_norm": 4.177667527282572,
|
|
"learning_rate": 5e-06,
|
|
"loss": 0.8018,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 0.384,
|
|
"grad_norm": 2.4660509268420956,
|
|
"learning_rate": 4.992293334332821e-06,
|
|
"loss": 0.8114,
|
|
"step": 6
|
|
},
|
|
{
|
|
"epoch": 0.448,
|
|
"grad_norm": 2.1484055104781894,
|
|
"learning_rate": 4.9692208514878445e-06,
|
|
"loss": 0.7613,
|
|
"step": 7
|
|
},
|
|
{
|
|
"epoch": 0.512,
|
|
"grad_norm": 4.24941370523086,
|
|
"learning_rate": 4.930924800994192e-06,
|
|
"loss": 0.7823,
|
|
"step": 8
|
|
},
|
|
{
|
|
"epoch": 0.576,
|
|
"grad_norm": 4.4742275329617796,
|
|
"learning_rate": 4.8776412907378845e-06,
|
|
"loss": 0.7947,
|
|
"step": 9
|
|
},
|
|
{
|
|
"epoch": 0.64,
|
|
"grad_norm": 4.584150310828626,
|
|
"learning_rate": 4.809698831278217e-06,
|
|
"loss": 0.7633,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 0.704,
|
|
"grad_norm": 4.214193969501364,
|
|
"learning_rate": 4.72751631047092e-06,
|
|
"loss": 0.7346,
|
|
"step": 11
|
|
},
|
|
{
|
|
"epoch": 0.768,
|
|
"grad_norm": 3.991993764075393,
|
|
"learning_rate": 4.631600410885231e-06,
|
|
"loss": 0.746,
|
|
"step": 12
|
|
},
|
|
{
|
|
"epoch": 0.832,
|
|
"grad_norm": 3.1675527640947263,
|
|
"learning_rate": 4.522542485937369e-06,
|
|
"loss": 0.7606,
|
|
"step": 13
|
|
},
|
|
{
|
|
"epoch": 0.896,
|
|
"grad_norm": 2.335842962931273,
|
|
"learning_rate": 4.401014914000078e-06,
|
|
"loss": 0.7106,
|
|
"step": 14
|
|
},
|
|
{
|
|
"epoch": 0.96,
|
|
"grad_norm": 1.7295030845724642,
|
|
"learning_rate": 4.267766952966369e-06,
|
|
"loss": 0.73,
|
|
"step": 15
|
|
},
|
|
{
|
|
"epoch": 1.032,
|
|
"grad_norm": 1.9808122172905522,
|
|
"learning_rate": 4.123620120825459e-06,
|
|
"loss": 1.0288,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 1.096,
|
|
"grad_norm": 1.549926635365851,
|
|
"learning_rate": 3.969463130731183e-06,
|
|
"loss": 0.8017,
|
|
"step": 17
|
|
},
|
|
{
|
|
"epoch": 1.16,
|
|
"grad_norm": 1.1443613272680522,
|
|
"learning_rate": 3.806246411789872e-06,
|
|
"loss": 0.5691,
|
|
"step": 18
|
|
},
|
|
{
|
|
"epoch": 1.224,
|
|
"grad_norm": 1.2766961031761175,
|
|
"learning_rate": 3.634976249348867e-06,
|
|
"loss": 0.6708,
|
|
"step": 19
|
|
},
|
|
{
|
|
"epoch": 1.288,
|
|
"grad_norm": 0.9945115738673014,
|
|
"learning_rate": 3.4567085809127247e-06,
|
|
"loss": 0.5849,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 1.3519999999999999,
|
|
"grad_norm": 0.8883087926133657,
|
|
"learning_rate": 3.272542485937369e-06,
|
|
"loss": 0.6678,
|
|
"step": 21
|
|
},
|
|
{
|
|
"epoch": 1.416,
|
|
"grad_norm": 0.8678409731064457,
|
|
"learning_rate": 3.0836134096397642e-06,
|
|
"loss": 0.6192,
|
|
"step": 22
|
|
},
|
|
{
|
|
"epoch": 1.48,
|
|
"grad_norm": 0.8592932615188835,
|
|
"learning_rate": 2.8910861626005774e-06,
|
|
"loss": 0.6421,
|
|
"step": 23
|
|
},
|
|
{
|
|
"epoch": 1.544,
|
|
"grad_norm": 0.8360777220132355,
|
|
"learning_rate": 2.696147739319613e-06,
|
|
"loss": 0.6162,
|
|
"step": 24
|
|
},
|
|
{
|
|
"epoch": 1.608,
|
|
"grad_norm": 0.7344073667637827,
|
|
"learning_rate": 2.5e-06,
|
|
"loss": 0.6482,
|
|
"step": 25
|
|
},
|
|
{
|
|
"epoch": 1.6720000000000002,
|
|
"grad_norm": 0.7979431332300274,
|
|
"learning_rate": 2.3038522606803882e-06,
|
|
"loss": 0.6589,
|
|
"step": 26
|
|
},
|
|
{
|
|
"epoch": 1.736,
|
|
"grad_norm": 0.6706917123119454,
|
|
"learning_rate": 2.1089138373994226e-06,
|
|
"loss": 0.5894,
|
|
"step": 27
|
|
},
|
|
{
|
|
"epoch": 1.8,
|
|
"grad_norm": 0.5516361015903626,
|
|
"learning_rate": 1.9163865903602374e-06,
|
|
"loss": 0.6103,
|
|
"step": 28
|
|
},
|
|
{
|
|
"epoch": 1.8639999999999999,
|
|
"grad_norm": 0.585020748716762,
|
|
"learning_rate": 1.7274575140626318e-06,
|
|
"loss": 0.5794,
|
|
"step": 29
|
|
},
|
|
{
|
|
"epoch": 1.928,
|
|
"grad_norm": 0.6557857522167144,
|
|
"learning_rate": 1.5432914190872757e-06,
|
|
"loss": 0.6615,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 1.992,
|
|
"grad_norm": 1.0124709491728257,
|
|
"learning_rate": 1.3650237506511333e-06,
|
|
"loss": 0.8842,
|
|
"step": 31
|
|
},
|
|
{
|
|
"epoch": 2.064,
|
|
"grad_norm": 0.6368456188165972,
|
|
"learning_rate": 1.193753588210128e-06,
|
|
"loss": 0.6189,
|
|
"step": 32
|
|
},
|
|
{
|
|
"epoch": 2.128,
|
|
"grad_norm": 0.5133965301911704,
|
|
"learning_rate": 1.0305368692688175e-06,
|
|
"loss": 0.5628,
|
|
"step": 33
|
|
},
|
|
{
|
|
"epoch": 2.192,
|
|
"grad_norm": 0.5303466898294212,
|
|
"learning_rate": 8.763798791745413e-07,
|
|
"loss": 0.5932,
|
|
"step": 34
|
|
},
|
|
{
|
|
"epoch": 2.2560000000000002,
|
|
"grad_norm": 0.6189893333744867,
|
|
"learning_rate": 7.322330470336314e-07,
|
|
"loss": 0.5959,
|
|
"step": 35
|
|
},
|
|
{
|
|
"epoch": 2.32,
|
|
"grad_norm": 0.49962903266405023,
|
|
"learning_rate": 5.989850859999227e-07,
|
|
"loss": 0.6223,
|
|
"step": 36
|
|
},
|
|
{
|
|
"epoch": 2.384,
|
|
"grad_norm": 0.4805054080903501,
|
|
"learning_rate": 4.774575140626317e-07,
|
|
"loss": 0.5717,
|
|
"step": 37
|
|
},
|
|
{
|
|
"epoch": 2.448,
|
|
"grad_norm": 0.5109754242918464,
|
|
"learning_rate": 3.683995891147696e-07,
|
|
"loss": 0.5771,
|
|
"step": 38
|
|
},
|
|
{
|
|
"epoch": 2.512,
|
|
"grad_norm": 0.44837516000440286,
|
|
"learning_rate": 2.7248368952908055e-07,
|
|
"loss": 0.5931,
|
|
"step": 39
|
|
},
|
|
{
|
|
"epoch": 2.576,
|
|
"grad_norm": 0.4870172383822943,
|
|
"learning_rate": 1.9030116872178317e-07,
|
|
"loss": 0.6105,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 2.64,
|
|
"grad_norm": 0.4254891486184633,
|
|
"learning_rate": 1.223587092621162e-07,
|
|
"loss": 0.5796,
|
|
"step": 41
|
|
},
|
|
{
|
|
"epoch": 2.7039999999999997,
|
|
"grad_norm": 0.5097241752627314,
|
|
"learning_rate": 6.907519900580862e-08,
|
|
"loss": 0.6076,
|
|
"step": 42
|
|
},
|
|
{
|
|
"epoch": 2.768,
|
|
"grad_norm": 0.4670809129547463,
|
|
"learning_rate": 3.077914851215585e-08,
|
|
"loss": 0.5931,
|
|
"step": 43
|
|
},
|
|
{
|
|
"epoch": 2.832,
|
|
"grad_norm": 0.4245230557723331,
|
|
"learning_rate": 7.70666566718009e-09,
|
|
"loss": 0.554,
|
|
"step": 44
|
|
},
|
|
{
|
|
"epoch": 2.896,
|
|
"grad_norm": 0.4713555028487441,
|
|
"learning_rate": 0.0,
|
|
"loss": 0.6079,
|
|
"step": 45
|
|
},
|
|
{
|
|
"epoch": 2.896,
|
|
"step": 45,
|
|
"total_flos": 5.682962857616998e+16,
|
|
"train_loss": 0.687734149561988,
|
|
"train_runtime": 1863.975,
|
|
"train_samples_per_second": 1.603,
|
|
"train_steps_per_second": 0.024
|
|
}
|
|
],
|
|
"logging_steps": 1.0,
|
|
"max_steps": 45,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 3,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 5.682962857616998e+16,
|
|
"train_batch_size": 1,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|