331 lines
7.8 KiB
JSON
331 lines
7.8 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 41,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.024844720496894408,
|
||
|
|
"grad_norm": 25.944225311279297,
|
||
|
|
"learning_rate": 0.0,
|
||
|
|
"loss": 2.5711,
|
||
|
|
"step": 1
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.049689440993788817,
|
||
|
|
"grad_norm": 25.800203323364258,
|
||
|
|
"learning_rate": 2.0000000000000003e-06,
|
||
|
|
"loss": 2.6176,
|
||
|
|
"step": 2
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.07453416149068323,
|
||
|
|
"grad_norm": 26.629310607910156,
|
||
|
|
"learning_rate": 4.000000000000001e-06,
|
||
|
|
"loss": 2.5861,
|
||
|
|
"step": 3
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.09937888198757763,
|
||
|
|
"grad_norm": 23.926393508911133,
|
||
|
|
"learning_rate": 6e-06,
|
||
|
|
"loss": 2.449,
|
||
|
|
"step": 4
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.12422360248447205,
|
||
|
|
"grad_norm": 25.569169998168945,
|
||
|
|
"learning_rate": 8.000000000000001e-06,
|
||
|
|
"loss": 2.1501,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.14906832298136646,
|
||
|
|
"grad_norm": 12.590336799621582,
|
||
|
|
"learning_rate": 1e-05,
|
||
|
|
"loss": 1.6997,
|
||
|
|
"step": 6
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.17391304347826086,
|
||
|
|
"grad_norm": 13.70421314239502,
|
||
|
|
"learning_rate": 9.980973490458728e-06,
|
||
|
|
"loss": 1.6168,
|
||
|
|
"step": 7
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.19875776397515527,
|
||
|
|
"grad_norm": 19.300262451171875,
|
||
|
|
"learning_rate": 9.924038765061042e-06,
|
||
|
|
"loss": 1.4418,
|
||
|
|
"step": 8
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2236024844720497,
|
||
|
|
"grad_norm": 9.026307106018066,
|
||
|
|
"learning_rate": 9.829629131445342e-06,
|
||
|
|
"loss": 1.3728,
|
||
|
|
"step": 9
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2484472049689441,
|
||
|
|
"grad_norm": 10.970627784729004,
|
||
|
|
"learning_rate": 9.698463103929542e-06,
|
||
|
|
"loss": 1.1993,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2732919254658385,
|
||
|
|
"grad_norm": 24.517372131347656,
|
||
|
|
"learning_rate": 9.531538935183252e-06,
|
||
|
|
"loss": 1.1649,
|
||
|
|
"step": 11
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2981366459627329,
|
||
|
|
"grad_norm": 8.326200485229492,
|
||
|
|
"learning_rate": 9.330127018922195e-06,
|
||
|
|
"loss": 0.9107,
|
||
|
|
"step": 12
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.32298136645962733,
|
||
|
|
"grad_norm": 7.975063800811768,
|
||
|
|
"learning_rate": 9.09576022144496e-06,
|
||
|
|
"loss": 1.1491,
|
||
|
|
"step": 13
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.34782608695652173,
|
||
|
|
"grad_norm": 8.031881332397461,
|
||
|
|
"learning_rate": 8.83022221559489e-06,
|
||
|
|
"loss": 1.1275,
|
||
|
|
"step": 14
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.37267080745341613,
|
||
|
|
"grad_norm": 7.613478660583496,
|
||
|
|
"learning_rate": 8.535533905932739e-06,
|
||
|
|
"loss": 0.9593,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.39751552795031053,
|
||
|
|
"grad_norm": 6.6011786460876465,
|
||
|
|
"learning_rate": 8.213938048432697e-06,
|
||
|
|
"loss": 0.8073,
|
||
|
|
"step": 16
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.422360248447205,
|
||
|
|
"grad_norm": 7.115065574645996,
|
||
|
|
"learning_rate": 7.86788218175523e-06,
|
||
|
|
"loss": 0.9254,
|
||
|
|
"step": 17
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4472049689440994,
|
||
|
|
"grad_norm": 7.958808422088623,
|
||
|
|
"learning_rate": 7.500000000000001e-06,
|
||
|
|
"loss": 0.7109,
|
||
|
|
"step": 18
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4720496894409938,
|
||
|
|
"grad_norm": 6.170507431030273,
|
||
|
|
"learning_rate": 7.113091308703498e-06,
|
||
|
|
"loss": 0.8236,
|
||
|
|
"step": 19
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4968944099378882,
|
||
|
|
"grad_norm": 24.92851448059082,
|
||
|
|
"learning_rate": 6.710100716628345e-06,
|
||
|
|
"loss": 0.7223,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5217391304347826,
|
||
|
|
"grad_norm": 9.233124732971191,
|
||
|
|
"learning_rate": 6.294095225512604e-06,
|
||
|
|
"loss": 0.6406,
|
||
|
|
"step": 21
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.546583850931677,
|
||
|
|
"grad_norm": 10.176820755004883,
|
||
|
|
"learning_rate": 5.8682408883346535e-06,
|
||
|
|
"loss": 0.7979,
|
||
|
|
"step": 22
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5714285714285714,
|
||
|
|
"grad_norm": 5.572312831878662,
|
||
|
|
"learning_rate": 5.435778713738292e-06,
|
||
|
|
"loss": 0.6265,
|
||
|
|
"step": 23
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5962732919254659,
|
||
|
|
"grad_norm": 6.122556686401367,
|
||
|
|
"learning_rate": 5e-06,
|
||
|
|
"loss": 0.6584,
|
||
|
|
"step": 24
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6211180124223602,
|
||
|
|
"grad_norm": 9.502265930175781,
|
||
|
|
"learning_rate": 4.564221286261709e-06,
|
||
|
|
"loss": 0.7447,
|
||
|
|
"step": 25
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6459627329192547,
|
||
|
|
"grad_norm": 6.603946685791016,
|
||
|
|
"learning_rate": 4.131759111665349e-06,
|
||
|
|
"loss": 0.7641,
|
||
|
|
"step": 26
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6708074534161491,
|
||
|
|
"grad_norm": 5.79754114151001,
|
||
|
|
"learning_rate": 3.705904774487396e-06,
|
||
|
|
"loss": 0.6721,
|
||
|
|
"step": 27
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6956521739130435,
|
||
|
|
"grad_norm": 6.924298286437988,
|
||
|
|
"learning_rate": 3.289899283371657e-06,
|
||
|
|
"loss": 0.7569,
|
||
|
|
"step": 28
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7204968944099379,
|
||
|
|
"grad_norm": 5.339048385620117,
|
||
|
|
"learning_rate": 2.886908691296504e-06,
|
||
|
|
"loss": 0.7,
|
||
|
|
"step": 29
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7453416149068323,
|
||
|
|
"grad_norm": 10.632051467895508,
|
||
|
|
"learning_rate": 2.5000000000000015e-06,
|
||
|
|
"loss": 0.6907,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7701863354037267,
|
||
|
|
"grad_norm": 12.731398582458496,
|
||
|
|
"learning_rate": 2.132117818244771e-06,
|
||
|
|
"loss": 0.5626,
|
||
|
|
"step": 31
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7950310559006211,
|
||
|
|
"grad_norm": 19.54654312133789,
|
||
|
|
"learning_rate": 1.7860619515673034e-06,
|
||
|
|
"loss": 0.7463,
|
||
|
|
"step": 32
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8198757763975155,
|
||
|
|
"grad_norm": 4.928001403808594,
|
||
|
|
"learning_rate": 1.4644660940672628e-06,
|
||
|
|
"loss": 0.6265,
|
||
|
|
"step": 33
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.84472049689441,
|
||
|
|
"grad_norm": 6.796016693115234,
|
||
|
|
"learning_rate": 1.1697777844051105e-06,
|
||
|
|
"loss": 0.7379,
|
||
|
|
"step": 34
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8695652173913043,
|
||
|
|
"grad_norm": 5.8230204582214355,
|
||
|
|
"learning_rate": 9.042397785550405e-07,
|
||
|
|
"loss": 0.5555,
|
||
|
|
"step": 35
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8944099378881988,
|
||
|
|
"grad_norm": 8.123178482055664,
|
||
|
|
"learning_rate": 6.698729810778065e-07,
|
||
|
|
"loss": 0.7681,
|
||
|
|
"step": 36
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9192546583850931,
|
||
|
|
"grad_norm": 7.072976112365723,
|
||
|
|
"learning_rate": 4.6846106481675035e-07,
|
||
|
|
"loss": 0.7249,
|
||
|
|
"step": 37
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9440993788819876,
|
||
|
|
"grad_norm": 5.121836185455322,
|
||
|
|
"learning_rate": 3.015368960704584e-07,
|
||
|
|
"loss": 0.5846,
|
||
|
|
"step": 38
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.968944099378882,
|
||
|
|
"grad_norm": 7.06396484375,
|
||
|
|
"learning_rate": 1.7037086855465902e-07,
|
||
|
|
"loss": 0.7607,
|
||
|
|
"step": 39
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9937888198757764,
|
||
|
|
"grad_norm": 7.021341323852539,
|
||
|
|
"learning_rate": 7.59612349389599e-08,
|
||
|
|
"loss": 0.7828,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"grad_norm": 7.021341323852539,
|
||
|
|
"learning_rate": 1.9026509541272276e-08,
|
||
|
|
"loss": 0.1798,
|
||
|
|
"step": 41
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"step": 41,
|
||
|
|
"total_flos": 9533860949786624.0,
|
||
|
|
"train_loss": 1.0508991407185067,
|
||
|
|
"train_runtime": 222.7751,
|
||
|
|
"train_samples_per_second": 2.891,
|
||
|
|
"train_steps_per_second": 0.184
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 1,
|
||
|
|
"max_steps": 41,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"save_steps": 100,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 9533860949786624.0,
|
||
|
|
"train_batch_size": 1,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|