pavitemple's picture
Model save
d26b324 verified
{
"best_metric": 0.390625,
"best_model_checkpoint": "finetuned-Accident-MultipleLabels-Video-subset-v2-checkpointing/checkpoint-4",
"epoch": 17.02857142857143,
"eval_steps": 500,
"global_step": 35,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.06,
"eval_accuracy": 0.359375,
"eval_loss": 1.7265217304229736,
"eval_runtime": 228.6142,
"eval_samples_per_second": 0.28,
"eval_steps_per_second": 0.035,
"step": 2
},
{
"epoch": 1.06,
"eval_accuracy": 0.390625,
"eval_loss": 1.6976414918899536,
"eval_runtime": 225.5421,
"eval_samples_per_second": 0.284,
"eval_steps_per_second": 0.035,
"step": 4
},
{
"epoch": 2.06,
"eval_accuracy": 0.359375,
"eval_loss": 1.7503018379211426,
"eval_runtime": 176.7414,
"eval_samples_per_second": 0.362,
"eval_steps_per_second": 0.045,
"step": 6
},
{
"epoch": 3.06,
"eval_accuracy": 0.3125,
"eval_loss": 1.8830795288085938,
"eval_runtime": 175.1171,
"eval_samples_per_second": 0.365,
"eval_steps_per_second": 0.046,
"step": 8
},
{
"epoch": 4.06,
"learning_rate": 4.032258064516129e-05,
"loss": 1.7254,
"step": 10
},
{
"epoch": 4.06,
"eval_accuracy": 0.171875,
"eval_loss": 2.0285067558288574,
"eval_runtime": 184.7182,
"eval_samples_per_second": 0.346,
"eval_steps_per_second": 0.043,
"step": 10
},
{
"epoch": 5.06,
"eval_accuracy": 0.28125,
"eval_loss": 2.039059638977051,
"eval_runtime": 178.8587,
"eval_samples_per_second": 0.358,
"eval_steps_per_second": 0.045,
"step": 12
},
{
"epoch": 6.06,
"eval_accuracy": 0.328125,
"eval_loss": 1.9736883640289307,
"eval_runtime": 179.8546,
"eval_samples_per_second": 0.356,
"eval_steps_per_second": 0.044,
"step": 14
},
{
"epoch": 7.06,
"eval_accuracy": 0.375,
"eval_loss": 1.8998150825500488,
"eval_runtime": 172.5948,
"eval_samples_per_second": 0.371,
"eval_steps_per_second": 0.046,
"step": 16
},
{
"epoch": 8.06,
"eval_accuracy": 0.375,
"eval_loss": 1.8785591125488281,
"eval_runtime": 169.9356,
"eval_samples_per_second": 0.377,
"eval_steps_per_second": 0.047,
"step": 18
},
{
"epoch": 9.06,
"learning_rate": 2.4193548387096777e-05,
"loss": 1.394,
"step": 20
},
{
"epoch": 9.06,
"eval_accuracy": 0.34375,
"eval_loss": 1.9054185152053833,
"eval_runtime": 179.2066,
"eval_samples_per_second": 0.357,
"eval_steps_per_second": 0.045,
"step": 20
},
{
"epoch": 10.06,
"eval_accuracy": 0.328125,
"eval_loss": 1.9473646879196167,
"eval_runtime": 173.6519,
"eval_samples_per_second": 0.369,
"eval_steps_per_second": 0.046,
"step": 22
},
{
"epoch": 11.06,
"eval_accuracy": 0.328125,
"eval_loss": 2.003166675567627,
"eval_runtime": 174.1677,
"eval_samples_per_second": 0.367,
"eval_steps_per_second": 0.046,
"step": 24
},
{
"epoch": 12.06,
"eval_accuracy": 0.328125,
"eval_loss": 2.072862386703491,
"eval_runtime": 196.1804,
"eval_samples_per_second": 0.326,
"eval_steps_per_second": 0.041,
"step": 26
},
{
"epoch": 13.06,
"eval_accuracy": 0.34375,
"eval_loss": 2.108105421066284,
"eval_runtime": 180.4592,
"eval_samples_per_second": 0.355,
"eval_steps_per_second": 0.044,
"step": 28
},
{
"epoch": 14.06,
"learning_rate": 8.064516129032258e-06,
"loss": 1.285,
"step": 30
},
{
"epoch": 14.06,
"eval_accuracy": 0.328125,
"eval_loss": 2.1189751625061035,
"eval_runtime": 180.6898,
"eval_samples_per_second": 0.354,
"eval_steps_per_second": 0.044,
"step": 30
},
{
"epoch": 15.06,
"eval_accuracy": 0.34375,
"eval_loss": 2.1188015937805176,
"eval_runtime": 172.3883,
"eval_samples_per_second": 0.371,
"eval_steps_per_second": 0.046,
"step": 32
},
{
"epoch": 16.06,
"eval_accuracy": 0.359375,
"eval_loss": 2.1155242919921875,
"eval_runtime": 169.8241,
"eval_samples_per_second": 0.377,
"eval_steps_per_second": 0.047,
"step": 34
},
{
"epoch": 17.03,
"eval_accuracy": 0.359375,
"eval_loss": 2.116347312927246,
"eval_runtime": 177.6799,
"eval_samples_per_second": 0.36,
"eval_steps_per_second": 0.045,
"step": 35
},
{
"epoch": 17.03,
"step": 35,
"total_flos": 1.3956545706511565e+18,
"train_loss": 1.4362963676452636,
"train_runtime": 8809.684,
"train_samples_per_second": 0.127,
"train_steps_per_second": 0.004
}
],
"logging_steps": 10,
"max_steps": 35,
"num_train_epochs": 9223372036854775807,
"save_steps": 500,
"total_flos": 1.3956545706511565e+18,
"trial_name": null,
"trial_params": null
}