last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.002677774471883368,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002975304968759298,
      "grad_norm": 4.187718868255615,
      "learning_rate": 2e-05,
      "loss": 6.1983,
      "step": 1
    },
    {
      "epoch": 0.0002975304968759298,
      "eval_loss": 3.89170503616333,
      "eval_runtime": 280.5373,
      "eval_samples_per_second": 5.047,
      "eval_steps_per_second": 2.524,
      "step": 1
    },
    {
      "epoch": 0.0005950609937518596,
      "grad_norm": 2.15948748588562,
      "learning_rate": 4e-05,
      "loss": 3.5477,
      "step": 2
    },
    {
      "epoch": 0.0008925914906277894,
      "grad_norm": 3.185936689376831,
      "learning_rate": 6e-05,
      "loss": 4.6576,
      "step": 3
    },
    {
      "epoch": 0.0008925914906277894,
      "eval_loss": 3.8822848796844482,
      "eval_runtime": 280.3445,
      "eval_samples_per_second": 5.051,
      "eval_steps_per_second": 2.525,
      "step": 3
    },
    {
      "epoch": 0.0011901219875037191,
      "grad_norm": 2.5385518074035645,
      "learning_rate": 8e-05,
      "loss": 4.4757,
      "step": 4
    },
    {
      "epoch": 0.001487652484379649,
      "grad_norm": 1.6990153789520264,
      "learning_rate": 0.0001,
      "loss": 3.1968,
      "step": 5
    },
    {
      "epoch": 0.0017851829812555787,
      "grad_norm": 2.8810431957244873,
      "learning_rate": 0.00012,
      "loss": 4.7495,
      "step": 6
    },
    {
      "epoch": 0.0017851829812555787,
      "eval_loss": 3.694185256958008,
      "eval_runtime": 280.1724,
      "eval_samples_per_second": 5.054,
      "eval_steps_per_second": 2.527,
      "step": 6
    },
    {
      "epoch": 0.0020827134781315083,
      "grad_norm": 1.4023734331130981,
      "learning_rate": 0.00014,
      "loss": 2.1711,
      "step": 7
    },
    {
      "epoch": 0.0023802439750074383,
      "grad_norm": 2.0291244983673096,
      "learning_rate": 0.00016,
      "loss": 3.006,
      "step": 8
    },
    {
      "epoch": 0.002677774471883368,
      "grad_norm": 2.446932315826416,
      "learning_rate": 0.00018,
      "loss": 3.5487,
      "step": 9
    },
    {
      "epoch": 0.002677774471883368,
      "eval_loss": 3.453287124633789,
      "eval_runtime": 281.5434,
      "eval_samples_per_second": 5.029,
      "eval_steps_per_second": 2.515,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2624890368688128.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}