|
{
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 6.9411764705882355,
|
|
"eval_steps": 25,
|
|
"global_step": 20,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.47058823529411764,
|
|
"grad_norm": 1.2416200637817383,
|
|
"learning_rate": 0.00019,
|
|
"loss": 9.549,
|
|
"step": 1
|
|
},
|
|
{
|
|
"epoch": 0.9411764705882353,
|
|
"grad_norm": 1.1442307233810425,
|
|
"learning_rate": 0.00018,
|
|
"loss": 9.4063,
|
|
"step": 2
|
|
},
|
|
{
|
|
"epoch": 1.0,
|
|
"grad_norm": 0.243339404463768,
|
|
"learning_rate": 0.00017,
|
|
"loss": 1.1454,
|
|
"step": 3
|
|
},
|
|
{
|
|
"epoch": 1.4705882352941178,
|
|
"grad_norm": 1.3809175491333008,
|
|
"learning_rate": 0.00016,
|
|
"loss": 8.9806,
|
|
"step": 4
|
|
},
|
|
{
|
|
"epoch": 1.9411764705882353,
|
|
"grad_norm": 1.642061710357666,
|
|
"learning_rate": 0.00015000000000000001,
|
|
"loss": 8.7867,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 2.0,
|
|
"grad_norm": 0.38183221220970154,
|
|
"learning_rate": 0.00014,
|
|
"loss": 1.0387,
|
|
"step": 6
|
|
},
|
|
{
|
|
"epoch": 2.4705882352941178,
|
|
"grad_norm": 2.3962018489837646,
|
|
"learning_rate": 0.00013000000000000002,
|
|
"loss": 8.2484,
|
|
"step": 7
|
|
},
|
|
{
|
|
"epoch": 2.9411764705882355,
|
|
"grad_norm": 2.1051485538482666,
|
|
"learning_rate": 0.00012,
|
|
"loss": 8.011,
|
|
"step": 8
|
|
},
|
|
{
|
|
"epoch": 3.0,
|
|
"grad_norm": 0.2919195592403412,
|
|
"learning_rate": 0.00011000000000000002,
|
|
"loss": 0.9644,
|
|
"step": 9
|
|
},
|
|
{
|
|
"epoch": 3.4705882352941178,
|
|
"grad_norm": 1.3799338340759277,
|
|
"learning_rate": 0.0001,
|
|
"loss": 7.6546,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 3.9411764705882355,
|
|
"grad_norm": 1.087898850440979,
|
|
"learning_rate": 9e-05,
|
|
"loss": 7.6277,
|
|
"step": 11
|
|
},
|
|
{
|
|
"epoch": 4.0,
|
|
"grad_norm": 0.2198086678981781,
|
|
"learning_rate": 8e-05,
|
|
"loss": 0.9398,
|
|
"step": 12
|
|
},
|
|
{
|
|
"epoch": 4.470588235294118,
|
|
"grad_norm": 0.8964143395423889,
|
|
"learning_rate": 7e-05,
|
|
"loss": 7.5336,
|
|
"step": 13
|
|
},
|
|
{
|
|
"epoch": 4.9411764705882355,
|
|
"grad_norm": 0.7436186671257019,
|
|
"learning_rate": 6e-05,
|
|
"loss": 7.4136,
|
|
"step": 14
|
|
},
|
|
{
|
|
"epoch": 5.0,
|
|
"grad_norm": 0.4537833034992218,
|
|
"learning_rate": 5e-05,
|
|
"loss": 0.9229,
|
|
"step": 15
|
|
},
|
|
{
|
|
"epoch": 5.470588235294118,
|
|
"grad_norm": 0.8003455996513367,
|
|
"learning_rate": 4e-05,
|
|
"loss": 7.3806,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 5.9411764705882355,
|
|
"grad_norm": 0.8167087435722351,
|
|
"learning_rate": 3e-05,
|
|
"loss": 7.4289,
|
|
"step": 17
|
|
},
|
|
{
|
|
"epoch": 6.0,
|
|
"grad_norm": 0.3596856892108917,
|
|
"learning_rate": 2e-05,
|
|
"loss": 0.9533,
|
|
"step": 18
|
|
},
|
|
{
|
|
"epoch": 6.470588235294118,
|
|
"grad_norm": 0.9435980319976807,
|
|
"learning_rate": 1e-05,
|
|
"loss": 7.4254,
|
|
"step": 19
|
|
},
|
|
{
|
|
"epoch": 6.9411764705882355,
|
|
"grad_norm": 0.6535025835037231,
|
|
"learning_rate": 0.0,
|
|
"loss": 7.3186,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 6.9411764705882355,
|
|
"step": 20,
|
|
"total_flos": 6374668143960.0,
|
|
"train_loss": 5.936464631557465,
|
|
"train_runtime": 569.9677,
|
|
"train_samples_per_second": 0.579,
|
|
"train_steps_per_second": 0.035
|
|
}
|
|
],
|
|
"logging_steps": 1.0,
|
|
"max_steps": 20,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 6374668143960.0,
|
|
"train_batch_size": 2,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|
|
|