sarahyo941's picture
Model save
3f08754 verified
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 48,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.06299212598425197,
"grad_norm": 1.6827600002288818,
"learning_rate": 2.0000000000000003e-06,
"loss": 0.9807,
"step": 1
},
{
"epoch": 0.12598425196850394,
"grad_norm": 1.6266508102416992,
"learning_rate": 4.000000000000001e-06,
"loss": 0.9458,
"step": 2
},
{
"epoch": 0.1889763779527559,
"grad_norm": 1.6120834350585938,
"learning_rate": 6e-06,
"loss": 0.9554,
"step": 3
},
{
"epoch": 0.25196850393700787,
"grad_norm": 1.5498627424240112,
"learning_rate": 8.000000000000001e-06,
"loss": 0.9262,
"step": 4
},
{
"epoch": 0.31496062992125984,
"grad_norm": 2.2747952938079834,
"learning_rate": 1e-05,
"loss": 0.9567,
"step": 5
},
{
"epoch": 0.3779527559055118,
"grad_norm": 2.724945068359375,
"learning_rate": 9.986661418317759e-06,
"loss": 0.94,
"step": 6
},
{
"epoch": 0.4409448818897638,
"grad_norm": 2.948086977005005,
"learning_rate": 9.946716840375552e-06,
"loss": 1.0088,
"step": 7
},
{
"epoch": 0.5039370078740157,
"grad_norm": 2.763153553009033,
"learning_rate": 9.880379387779637e-06,
"loss": 1.0134,
"step": 8
},
{
"epoch": 0.5669291338582677,
"grad_norm": 2.4897592067718506,
"learning_rate": 9.78800299954203e-06,
"loss": 1.0267,
"step": 9
},
{
"epoch": 0.6299212598425197,
"grad_norm": 2.5692765712738037,
"learning_rate": 9.670080543662742e-06,
"loss": 0.9758,
"step": 10
},
{
"epoch": 0.6929133858267716,
"grad_norm": 2.352846622467041,
"learning_rate": 9.527241187465735e-06,
"loss": 0.9602,
"step": 11
},
{
"epoch": 0.7559055118110236,
"grad_norm": 2.4274556636810303,
"learning_rate": 9.36024704071904e-06,
"loss": 1.0215,
"step": 12
},
{
"epoch": 0.8188976377952756,
"grad_norm": 2.3770511150360107,
"learning_rate": 9.16998908944939e-06,
"loss": 0.9639,
"step": 13
},
{
"epoch": 0.8818897637795275,
"grad_norm": 2.4261667728424072,
"learning_rate": 8.957482442146271e-06,
"loss": 0.9426,
"step": 14
},
{
"epoch": 0.9448818897637795,
"grad_norm": 2.2574000358581543,
"learning_rate": 8.72386091371891e-06,
"loss": 0.98,
"step": 15
},
{
"epoch": 1.0,
"grad_norm": 2.383176565170288,
"learning_rate": 8.470370976103171e-06,
"loss": 0.9405,
"step": 16
},
{
"epoch": 1.0629921259842519,
"grad_norm": 2.3710620403289795,
"learning_rate": 8.198365107794457e-06,
"loss": 0.8291,
"step": 17
},
{
"epoch": 1.125984251968504,
"grad_norm": 2.4993762969970703,
"learning_rate": 7.909294577789765e-06,
"loss": 0.83,
"step": 18
},
{
"epoch": 1.188976377952756,
"grad_norm": 2.5276553630828857,
"learning_rate": 7.604701702439652e-06,
"loss": 0.8006,
"step": 19
},
{
"epoch": 1.2519685039370079,
"grad_norm": 2.4743146896362305,
"learning_rate": 7.286211616523193e-06,
"loss": 0.7527,
"step": 20
},
{
"epoch": 1.3149606299212597,
"grad_norm": 2.530345916748047,
"learning_rate": 6.95552360245078e-06,
"loss": 0.785,
"step": 21
},
{
"epoch": 1.3779527559055118,
"grad_norm": 2.6544010639190674,
"learning_rate": 6.614402023857231e-06,
"loss": 0.8194,
"step": 22
},
{
"epoch": 1.4409448818897639,
"grad_norm": 2.5227644443511963,
"learning_rate": 6.264666911958404e-06,
"loss": 0.7987,
"step": 23
},
{
"epoch": 1.5039370078740157,
"grad_norm": 2.5911266803741455,
"learning_rate": 5.908184254897183e-06,
"loss": 0.7847,
"step": 24
},
{
"epoch": 1.5669291338582676,
"grad_norm": 2.3522489070892334,
"learning_rate": 5.546856041889374e-06,
"loss": 0.7723,
"step": 25
},
{
"epoch": 1.6299212598425197,
"grad_norm": 2.3171679973602295,
"learning_rate": 5.182610115288296e-06,
"loss": 0.7435,
"step": 26
},
{
"epoch": 1.6929133858267718,
"grad_norm": 2.3236382007598877,
"learning_rate": 4.817389884711706e-06,
"loss": 0.7955,
"step": 27
},
{
"epoch": 1.7559055118110236,
"grad_norm": 2.3249104022979736,
"learning_rate": 4.4531439581106295e-06,
"loss": 0.756,
"step": 28
},
{
"epoch": 1.8188976377952755,
"grad_norm": 2.308413028717041,
"learning_rate": 4.091815745102818e-06,
"loss": 0.7696,
"step": 29
},
{
"epoch": 1.8818897637795275,
"grad_norm": 2.3917453289031982,
"learning_rate": 3.7353330880415963e-06,
"loss": 0.7891,
"step": 30
},
{
"epoch": 1.9448818897637796,
"grad_norm": 2.2722179889678955,
"learning_rate": 3.3855979761427705e-06,
"loss": 0.7326,
"step": 31
},
{
"epoch": 2.0,
"grad_norm": 2.3211171627044678,
"learning_rate": 3.044476397549221e-06,
"loss": 0.7675,
"step": 32
},
{
"epoch": 2.062992125984252,
"grad_norm": 2.114405632019043,
"learning_rate": 2.7137883834768076e-06,
"loss": 0.6823,
"step": 33
},
{
"epoch": 2.1259842519685037,
"grad_norm": 2.106955051422119,
"learning_rate": 2.3952982975603494e-06,
"loss": 0.6718,
"step": 34
},
{
"epoch": 2.188976377952756,
"grad_norm": 2.1824538707733154,
"learning_rate": 2.0907054222102367e-06,
"loss": 0.6405,
"step": 35
},
{
"epoch": 2.251968503937008,
"grad_norm": 2.105780839920044,
"learning_rate": 1.8016348922055448e-06,
"loss": 0.6769,
"step": 36
},
{
"epoch": 2.3149606299212597,
"grad_norm": 2.0345335006713867,
"learning_rate": 1.5296290238968303e-06,
"loss": 0.6514,
"step": 37
},
{
"epoch": 2.377952755905512,
"grad_norm": 1.9925103187561035,
"learning_rate": 1.2761390862810907e-06,
"loss": 0.6376,
"step": 38
},
{
"epoch": 2.440944881889764,
"grad_norm": 2.051919937133789,
"learning_rate": 1.04251755785373e-06,
"loss": 0.666,
"step": 39
},
{
"epoch": 2.5039370078740157,
"grad_norm": 1.9680358171463013,
"learning_rate": 8.30010910550611e-07,
"loss": 0.667,
"step": 40
},
{
"epoch": 2.5669291338582676,
"grad_norm": 2.007450819015503,
"learning_rate": 6.397529592809615e-07,
"loss": 0.6699,
"step": 41
},
{
"epoch": 2.6299212598425195,
"grad_norm": 1.990918517112732,
"learning_rate": 4.727588125342669e-07,
"loss": 0.6151,
"step": 42
},
{
"epoch": 2.6929133858267718,
"grad_norm": 1.9402188062667847,
"learning_rate": 3.299194563372604e-07,
"loss": 0.6667,
"step": 43
},
{
"epoch": 2.7559055118110236,
"grad_norm": 1.8792628049850464,
"learning_rate": 2.1199700045797077e-07,
"loss": 0.6395,
"step": 44
},
{
"epoch": 2.8188976377952755,
"grad_norm": 1.8017092943191528,
"learning_rate": 1.196206122203647e-07,
"loss": 0.6417,
"step": 45
},
{
"epoch": 2.8818897637795278,
"grad_norm": 1.8651630878448486,
"learning_rate": 5.3283159624448745e-08,
"loss": 0.6361,
"step": 46
},
{
"epoch": 2.9448818897637796,
"grad_norm": 1.874280571937561,
"learning_rate": 1.333858168224178e-08,
"loss": 0.6426,
"step": 47
},
{
"epoch": 3.0,
"grad_norm": 1.9368858337402344,
"learning_rate": 0.0,
"loss": 0.6177,
"step": 48
},
{
"epoch": 3.0,
"step": 48,
"total_flos": 1.2587951231441306e+17,
"train_loss": 0.8018192363282045,
"train_runtime": 3899.5488,
"train_samples_per_second": 0.786,
"train_steps_per_second": 0.012
}
],
"logging_steps": 1.0,
"max_steps": 48,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.2587951231441306e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}