last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.018298261665141813,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0018298261665141812,
      "grad_norm": 0.12623006105422974,
      "learning_rate": 2e-05,
      "loss": 1.1082,
      "step": 1
    },
    {
      "epoch": 0.0036596523330283625,
      "grad_norm": 0.1358586847782135,
      "learning_rate": 4e-05,
      "loss": 1.2337,
      "step": 2
    },
    {
      "epoch": 0.0054894784995425435,
      "grad_norm": 0.1364835798740387,
      "learning_rate": 6e-05,
      "loss": 1.3508,
      "step": 3
    },
    {
      "epoch": 0.0054894784995425435,
      "eval_loss": 1.3069536685943604,
      "eval_runtime": 11.1524,
      "eval_samples_per_second": 20.713,
      "eval_steps_per_second": 10.401,
      "step": 3
    },
    {
      "epoch": 0.007319304666056725,
      "grad_norm": 0.1433553695678711,
      "learning_rate": 8e-05,
      "loss": 1.3129,
      "step": 4
    },
    {
      "epoch": 0.009149130832570906,
      "grad_norm": 0.13966652750968933,
      "learning_rate": 0.0001,
      "loss": 1.267,
      "step": 5
    },
    {
      "epoch": 0.010978956999085087,
      "grad_norm": 0.1373775750398636,
      "learning_rate": 0.00012,
      "loss": 1.1703,
      "step": 6
    },
    {
      "epoch": 0.010978956999085087,
      "eval_loss": 1.3050342798233032,
      "eval_runtime": 10.9745,
      "eval_samples_per_second": 21.049,
      "eval_steps_per_second": 10.57,
      "step": 6
    },
    {
      "epoch": 0.012808783165599268,
      "grad_norm": 0.16164709627628326,
      "learning_rate": 0.00014,
      "loss": 1.4877,
      "step": 7
    },
    {
      "epoch": 0.01463860933211345,
      "grad_norm": 0.1583903431892395,
      "learning_rate": 0.00016,
      "loss": 1.3104,
      "step": 8
    },
    {
      "epoch": 0.01646843549862763,
      "grad_norm": 0.14703314006328583,
      "learning_rate": 0.00018,
      "loss": 1.1769,
      "step": 9
    },
    {
      "epoch": 0.01646843549862763,
      "eval_loss": 1.2979130744934082,
      "eval_runtime": 15.1655,
      "eval_samples_per_second": 15.232,
      "eval_steps_per_second": 7.649,
      "step": 9
    },
    {
      "epoch": 0.018298261665141813,
      "grad_norm": 0.18663398921489716,
      "learning_rate": 0.0002,
      "loss": 1.2693,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 766287626108928.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}