167 lines
4.8 KiB
JSON
167 lines
4.8 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 2.8941176470588235,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 63,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.23529411764705882,
|
||
|
|
"grad_norm": 1.6521763801574707,
|
||
|
|
"learning_rate": 9.999999999999999e-05,
|
||
|
|
"loss": 2.3013,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 5,
|
||
|
|
"tokens_per_second_per_gpu": 718.74
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.47058823529411764,
|
||
|
|
"grad_norm": 1.0924193859100342,
|
||
|
|
"learning_rate": 0.00014897709775520417,
|
||
|
|
"loss": 1.3236,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 10,
|
||
|
|
"tokens_per_second_per_gpu": 731.56
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7058823529411765,
|
||
|
|
"grad_norm": 0.6611356735229492,
|
||
|
|
"learning_rate": 0.00014282678705231832,
|
||
|
|
"loss": 0.5684,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 15,
|
||
|
|
"tokens_per_second_per_gpu": 722.65
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9411764705882353,
|
||
|
|
"grad_norm": 0.6052656173706055,
|
||
|
|
"learning_rate": 0.00013155799573326722,
|
||
|
|
"loss": 0.3071,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 20,
|
||
|
|
"tokens_per_second_per_gpu": 728.39
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.1411764705882352,
|
||
|
|
"grad_norm": 0.5231208801269531,
|
||
|
|
"learning_rate": 0.00011602111185918203,
|
||
|
|
"loss": 0.2304,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 25,
|
||
|
|
"tokens_per_second_per_gpu": 625.16
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.3764705882352941,
|
||
|
|
"grad_norm": 0.34258556365966797,
|
||
|
|
"learning_rate": 9.738861082512709e-05,
|
||
|
|
"loss": 0.2163,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 30,
|
||
|
|
"tokens_per_second_per_gpu": 724.72
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.611764705882353,
|
||
|
|
"grad_norm": 0.3575713038444519,
|
||
|
|
"learning_rate": 7.706657567761216e-05,
|
||
|
|
"loss": 0.2039,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 35,
|
||
|
|
"tokens_per_second_per_gpu": 723.32
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.8470588235294119,
|
||
|
|
"grad_norm": 0.4452803134918213,
|
||
|
|
"learning_rate": 5.658858846444006e-05,
|
||
|
|
"loss": 0.1938,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 40,
|
||
|
|
"tokens_per_second_per_gpu": 725.05
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.0470588235294116,
|
||
|
|
"grad_norm": 0.2797813415527344,
|
||
|
|
"learning_rate": 3.750000000000001e-05,
|
||
|
|
"loss": 0.1849,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 45,
|
||
|
|
"tokens_per_second_per_gpu": 607.58
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.2823529411764705,
|
||
|
|
"grad_norm": 0.36559921503067017,
|
||
|
|
"learning_rate": 2.1241311512366167e-05,
|
||
|
|
"loss": 0.1819,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 50,
|
||
|
|
"tokens_per_second_per_gpu": 730.88
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.5176470588235293,
|
||
|
|
"grad_norm": 0.2546774744987488,
|
||
|
|
"learning_rate": 9.039468659513327e-06,
|
||
|
|
"loss": 0.1818,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 55,
|
||
|
|
"tokens_per_second_per_gpu": 729.82
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 2.7529411764705882,
|
||
|
|
"grad_norm": 0.27984631061553955,
|
||
|
|
"learning_rate": 1.8152713029423283e-06,
|
||
|
|
"loss": 0.1846,
|
||
|
|
"memory/device_reserved (GiB)": 66.98,
|
||
|
|
"memory/max_active (GiB)": 57.73,
|
||
|
|
"memory/max_allocated (GiB)": 57.73,
|
||
|
|
"step": 60,
|
||
|
|
"tokens_per_second_per_gpu": 723.26
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 5,
|
||
|
|
"max_steps": 63,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 3,
|
||
|
|
"save_steps": 50,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 1.3854128835290726e+17,
|
||
|
|
"train_batch_size": 4,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|