trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 136,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14705882352941177,
      "grad_norm": 18.60810157022616,
      "learning_rate": 6.4285714285714295e-06,
      "loss": 3.0049,
      "step": 10
    },
    {
      "epoch": 0.29411764705882354,
      "grad_norm": 7.546445604794843,
      "learning_rate": 9.958613370680507e-06,
      "loss": 1.5679,
      "step": 20
    },
    {
      "epoch": 0.4411764705882353,
      "grad_norm": 6.79358887429206,
      "learning_rate": 9.631619841257477e-06,
      "loss": 1.17,
      "step": 30
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 6.982504247579255,
      "learning_rate": 8.999196223698599e-06,
      "loss": 0.9323,
      "step": 40
    },
    {
      "epoch": 0.7352941176470589,
      "grad_norm": 7.442261648915706,
      "learning_rate": 8.103047409137114e-06,
      "loss": 0.9835,
      "step": 50
    },
    {
      "epoch": 0.8823529411764706,
      "grad_norm": 7.2156252241121575,
      "learning_rate": 7.002269528256334e-06,
      "loss": 0.9241,
      "step": 60
    },
    {
      "epoch": 1.0294117647058822,
      "grad_norm": 4.57233394591064,
      "learning_rate": 5.76945288352031e-06,
      "loss": 0.818,
      "step": 70
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 5.70679901027974,
      "learning_rate": 4.485895014313198e-06,
      "loss": 0.5443,
      "step": 80
    },
    {
      "epoch": 1.3235294117647058,
      "grad_norm": 5.377159576531448,
      "learning_rate": 3.236239567254526e-06,
      "loss": 0.4745,
      "step": 90
    },
    {
      "epoch": 1.4705882352941178,
      "grad_norm": 5.859867601143253,
      "learning_rate": 2.102894508977182e-06,
      "loss": 0.5068,
      "step": 100
    },
    {
      "epoch": 1.6176470588235294,
      "grad_norm": 6.491101584304634,
      "learning_rate": 1.1605977698170001e-06,
      "loss": 0.4754,
      "step": 110
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 5.468108913057432,
      "learning_rate": 4.71488684597643e-07,
      "loss": 0.4828,
      "step": 120
    },
    {
      "epoch": 1.9117647058823528,
      "grad_norm": 6.171970457432134,
      "learning_rate": 8.101024213241826e-08,
      "loss": 0.4865,
      "step": 130
    },
    {
      "epoch": 2.0,
      "step": 136,
      "total_flos": 3208165081088.0,
      "train_loss": 0.9287934250691358,
      "train_runtime": 291.6194,
      "train_samples_per_second": 3.717,
      "train_steps_per_second": 0.466
    }
  ],
  "logging_steps": 10,
  "max_steps": 136,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3208165081088.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}