mariagrandury's picture
End of training
1c7d2cb verified
{
"best_metric": 0.2140692864641761,
"best_model_checkpoint": "/content/drive/MyDrive/model_outputs/trial_0/checkpoint-292",
"epoch": 1.0,
"eval_steps": 500,
"global_step": 292,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03424657534246575,
"grad_norm": 8.432939529418945,
"learning_rate": 8.18270817933484e-07,
"loss": 2.2234,
"step": 10
},
{
"epoch": 0.0684931506849315,
"grad_norm": 8.3140230178833,
"learning_rate": 1.636541635866968e-06,
"loss": 2.2403,
"step": 20
},
{
"epoch": 0.10273972602739725,
"grad_norm": 7.506945610046387,
"learning_rate": 2.454812453800452e-06,
"loss": 2.2297,
"step": 30
},
{
"epoch": 0.136986301369863,
"grad_norm": 8.531285285949707,
"learning_rate": 3.273083271733936e-06,
"loss": 2.1823,
"step": 40
},
{
"epoch": 0.17123287671232876,
"grad_norm": 7.951968669891357,
"learning_rate": 4.091354089667421e-06,
"loss": 2.1696,
"step": 50
},
{
"epoch": 0.2054794520547945,
"grad_norm": 8.328391075134277,
"learning_rate": 4.909624907600904e-06,
"loss": 2.1415,
"step": 60
},
{
"epoch": 0.23972602739726026,
"grad_norm": 7.337668418884277,
"learning_rate": 5.727895725534388e-06,
"loss": 2.1206,
"step": 70
},
{
"epoch": 0.273972602739726,
"grad_norm": 7.09800386428833,
"learning_rate": 6.546166543467872e-06,
"loss": 2.0949,
"step": 80
},
{
"epoch": 0.3082191780821918,
"grad_norm": 6.549777507781982,
"learning_rate": 7.364437361401356e-06,
"loss": 2.076,
"step": 90
},
{
"epoch": 0.3424657534246575,
"grad_norm": 7.464539051055908,
"learning_rate": 8.182708179334841e-06,
"loss": 2.096,
"step": 100
},
{
"epoch": 0.3767123287671233,
"grad_norm": 6.982883930206299,
"learning_rate": 9.000978997268324e-06,
"loss": 2.0615,
"step": 110
},
{
"epoch": 0.410958904109589,
"grad_norm": 7.149113655090332,
"learning_rate": 9.819249815201808e-06,
"loss": 2.0785,
"step": 120
},
{
"epoch": 0.4452054794520548,
"grad_norm": 6.507534503936768,
"learning_rate": 1.0637520633135292e-05,
"loss": 2.0865,
"step": 130
},
{
"epoch": 0.4794520547945205,
"grad_norm": 6.1932196617126465,
"learning_rate": 1.1455791451068777e-05,
"loss": 2.0533,
"step": 140
},
{
"epoch": 0.5136986301369864,
"grad_norm": 7.072916507720947,
"learning_rate": 1.1783771409556479e-05,
"loss": 2.0613,
"step": 150
},
{
"epoch": 0.547945205479452,
"grad_norm": 8.155049324035645,
"learning_rate": 1.1621460508598402e-05,
"loss": 2.0284,
"step": 160
},
{
"epoch": 0.5821917808219178,
"grad_norm": 9.3668851852417,
"learning_rate": 1.1459149607640322e-05,
"loss": 2.0137,
"step": 170
},
{
"epoch": 0.6164383561643836,
"grad_norm": 8.165146827697754,
"learning_rate": 1.1296838706682245e-05,
"loss": 1.9835,
"step": 180
},
{
"epoch": 0.6506849315068494,
"grad_norm": 9.0431547164917,
"learning_rate": 1.1134527805724166e-05,
"loss": 1.9934,
"step": 190
},
{
"epoch": 0.684931506849315,
"grad_norm": 8.92784309387207,
"learning_rate": 1.0972216904766088e-05,
"loss": 2.0081,
"step": 200
},
{
"epoch": 0.7191780821917808,
"grad_norm": 9.184441566467285,
"learning_rate": 1.0809906003808009e-05,
"loss": 1.9567,
"step": 210
},
{
"epoch": 0.7534246575342466,
"grad_norm": 10.262762069702148,
"learning_rate": 1.0647595102849931e-05,
"loss": 1.966,
"step": 220
},
{
"epoch": 0.7876712328767124,
"grad_norm": 7.367489337921143,
"learning_rate": 1.0485284201891854e-05,
"loss": 2.051,
"step": 230
},
{
"epoch": 0.821917808219178,
"grad_norm": 8.76405143737793,
"learning_rate": 1.0322973300933776e-05,
"loss": 2.0785,
"step": 240
},
{
"epoch": 0.8561643835616438,
"grad_norm": 7.100874900817871,
"learning_rate": 1.0160662399975697e-05,
"loss": 1.9977,
"step": 250
},
{
"epoch": 0.8904109589041096,
"grad_norm": 8.239387512207031,
"learning_rate": 9.998351499017618e-06,
"loss": 2.0099,
"step": 260
},
{
"epoch": 0.9246575342465754,
"grad_norm": 7.704412460327148,
"learning_rate": 9.83604059805954e-06,
"loss": 1.9851,
"step": 270
},
{
"epoch": 0.958904109589041,
"grad_norm": 7.8163652420043945,
"learning_rate": 9.673729697101463e-06,
"loss": 2.0085,
"step": 280
},
{
"epoch": 0.9931506849315068,
"grad_norm": 8.153071403503418,
"learning_rate": 9.511418796143383e-06,
"loss": 1.9962,
"step": 290
},
{
"epoch": 1.0,
"eval_classification_report": {
"accuracy": 0.2375,
"ar": {
"f1-score": 0.0622568093385214,
"precision": 0.1568627450980392,
"recall": 0.038834951456310676,
"support": 206.0
},
"cl": {
"f1-score": 0.2011173184357542,
"precision": 0.16901408450704225,
"recall": 0.2482758620689655,
"support": 290.0
},
"co": {
"f1-score": 0.31543624161073824,
"precision": 0.23383084577114427,
"recall": 0.4845360824742268,
"support": 291.0
},
"es": {
"f1-score": 0.27364185110663986,
"precision": 0.3119266055045872,
"recall": 0.24372759856630824,
"support": 279.0
},
"macro avg": {
"f1-score": 0.2140692864641761,
"precision": 0.24922489036795273,
"recall": 0.22042150190811827,
"support": 2000.0
},
"mx": {
"f1-score": 0.22950819672131148,
"precision": 0.28426395939086296,
"recall": 0.19243986254295534,
"support": 291.0
},
"pe": {
"f1-score": 0.10610079575596817,
"precision": 0.23255813953488372,
"recall": 0.06872852233676977,
"support": 291.0
},
"pr": {
"f1-score": 0.5030674846625767,
"precision": 0.6612903225806451,
"recall": 0.40594059405940597,
"support": 101.0
},
"uy": {
"f1-score": 0.2354948805460751,
"precision": 0.19327731092436976,
"recall": 0.30131004366812225,
"support": 229.0
},
"ve": {
"f1-score": 0.0,
"precision": 0.0,
"recall": 0.0,
"support": 22.0
},
"weighted avg": {
"f1-score": 0.22084365412222065,
"precision": 0.24892308331106963,
"recall": 0.2375,
"support": 2000.0
}
},
"eval_f1": 0.2140692864641761,
"eval_loss": 1.964627742767334,
"eval_runtime": 3.948,
"eval_samples_per_second": 506.591,
"eval_steps_per_second": 31.662,
"step": 292
}
],
"logging_steps": 10,
"max_steps": 876,
"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": false
},
"attributes": {}
}
},
"total_flos": 306938335993344.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}