avsolatorio's picture
Training in progress, step 8000, checkpoint
a30d463 verified
raw
history blame
6.81 kB
{
"best_metric": 0.040071167051792145,
"best_model_checkpoint": "doc-topic-model_eval-01_train-03/checkpoint-8000",
"epoch": 3.9447731755424065,
"eval_steps": 1000,
"global_step": 8000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.2465483234714004,
"grad_norm": 0.3881610631942749,
"learning_rate": 1.9950690335305722e-05,
"loss": 0.1669,
"step": 500
},
{
"epoch": 0.4930966469428008,
"grad_norm": 0.3564796447753906,
"learning_rate": 1.9901380670611442e-05,
"loss": 0.0935,
"step": 1000
},
{
"epoch": 0.4930966469428008,
"eval_accuracy": 0.981383845685049,
"eval_f1": 0.0,
"eval_loss": 0.08976549655199051,
"eval_precision": 0.0,
"eval_recall": 0.0,
"eval_runtime": 12.0327,
"eval_samples_per_second": 673.996,
"eval_steps_per_second": 2.659,
"step": 1000
},
{
"epoch": 0.7396449704142012,
"grad_norm": 0.4097365140914917,
"learning_rate": 1.9852071005917162e-05,
"loss": 0.0869,
"step": 1500
},
{
"epoch": 0.9861932938856016,
"grad_norm": 0.3858756422996521,
"learning_rate": 1.980276134122288e-05,
"loss": 0.0764,
"step": 2000
},
{
"epoch": 0.9861932938856016,
"eval_accuracy": 0.981383845685049,
"eval_f1": 0.0,
"eval_loss": 0.07017117738723755,
"eval_precision": 0.0,
"eval_recall": 0.0,
"eval_runtime": 12.2763,
"eval_samples_per_second": 660.624,
"eval_steps_per_second": 2.607,
"step": 2000
},
{
"epoch": 1.232741617357002,
"grad_norm": 0.3765203356742859,
"learning_rate": 1.9753451676528602e-05,
"loss": 0.068,
"step": 2500
},
{
"epoch": 1.4792899408284024,
"grad_norm": 0.3380969762802124,
"learning_rate": 1.9704142011834322e-05,
"loss": 0.0621,
"step": 3000
},
{
"epoch": 1.4792899408284024,
"eval_accuracy": 0.9819748572076478,
"eval_f1": 0.06950541410594088,
"eval_loss": 0.05700037255883217,
"eval_precision": 0.8911819887429644,
"eval_recall": 0.03616292348686715,
"eval_runtime": 14.2939,
"eval_samples_per_second": 567.375,
"eval_steps_per_second": 2.239,
"step": 3000
},
{
"epoch": 1.725838264299803,
"grad_norm": 0.3490855395793915,
"learning_rate": 1.965483234714004e-05,
"loss": 0.056,
"step": 3500
},
{
"epoch": 1.972386587771203,
"grad_norm": 0.4068734347820282,
"learning_rate": 1.9605522682445763e-05,
"loss": 0.0542,
"step": 4000
},
{
"epoch": 1.972386587771203,
"eval_accuracy": 0.9839533993792253,
"eval_f1": 0.2863985881759738,
"eval_loss": 0.049796320497989655,
"eval_precision": 0.8319296960820213,
"eval_recall": 0.17297297297297298,
"eval_runtime": 14.4691,
"eval_samples_per_second": 560.505,
"eval_steps_per_second": 2.212,
"step": 4000
},
{
"epoch": 2.2189349112426036,
"grad_norm": 0.39391544461250305,
"learning_rate": 1.9556213017751483e-05,
"loss": 0.0488,
"step": 4500
},
{
"epoch": 2.465483234714004,
"grad_norm": 0.2698710858821869,
"learning_rate": 1.95069033530572e-05,
"loss": 0.0468,
"step": 5000
},
{
"epoch": 2.465483234714004,
"eval_accuracy": 0.9851807758266367,
"eval_f1": 0.4191111111111111,
"eval_loss": 0.04676121473312378,
"eval_precision": 0.7753340184994861,
"eval_recall": 0.2871716787209745,
"eval_runtime": 14.3877,
"eval_samples_per_second": 563.675,
"eval_steps_per_second": 2.224,
"step": 5000
},
{
"epoch": 2.712031558185404,
"grad_norm": 0.39123860001564026,
"learning_rate": 1.9457593688362923e-05,
"loss": 0.0456,
"step": 5500
},
{
"epoch": 2.9585798816568047,
"grad_norm": 0.4525609016418457,
"learning_rate": 1.940828402366864e-05,
"loss": 0.0441,
"step": 6000
},
{
"epoch": 2.9585798816568047,
"eval_accuracy": 0.9861062686905623,
"eval_f1": 0.4897725498360485,
"eval_loss": 0.04347195476293564,
"eval_precision": 0.774103323461665,
"eval_recall": 0.35820327369623145,
"eval_runtime": 14.2071,
"eval_samples_per_second": 570.842,
"eval_steps_per_second": 2.252,
"step": 6000
},
{
"epoch": 3.2051282051282053,
"grad_norm": 0.47004055976867676,
"learning_rate": 1.935897435897436e-05,
"loss": 0.0409,
"step": 6500
},
{
"epoch": 3.4516765285996054,
"grad_norm": 0.3331545293331146,
"learning_rate": 1.930966469428008e-05,
"loss": 0.0395,
"step": 7000
},
{
"epoch": 3.4516765285996054,
"eval_accuracy": 0.9860283175305072,
"eval_f1": 0.5278735632183909,
"eval_loss": 0.041799116879701614,
"eval_precision": 0.7115558424790187,
"eval_recall": 0.4195660449181576,
"eval_runtime": 14.3982,
"eval_samples_per_second": 563.265,
"eval_steps_per_second": 2.223,
"step": 7000
},
{
"epoch": 3.698224852071006,
"grad_norm": 0.46907347440719604,
"learning_rate": 1.92603550295858e-05,
"loss": 0.0382,
"step": 7500
},
{
"epoch": 3.9447731755424065,
"grad_norm": 0.3860171437263489,
"learning_rate": 1.921104536489152e-05,
"loss": 0.0384,
"step": 8000
},
{
"epoch": 3.9447731755424065,
"eval_accuracy": 0.9865853140014457,
"eval_f1": 0.5588440922861804,
"eval_loss": 0.040071167051792145,
"eval_precision": 0.7205528846153846,
"eval_recall": 0.45641416063951273,
"eval_runtime": 14.3408,
"eval_samples_per_second": 565.521,
"eval_steps_per_second": 2.231,
"step": 8000
}
],
"logging_steps": 500,
"max_steps": 202800,
"num_input_tokens_seen": 0,
"num_train_epochs": 100,
"save_steps": 1000,
"stateful_callbacks": {
"EarlyStoppingCallback": {
"args": {
"early_stopping_patience": 5,
"early_stopping_threshold": 0.0
},
"attributes": {
"early_stopping_patience_counter": 0
}
},
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 180338255777124.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}