last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.008821367311933349,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00098015192354815,
      "grad_norm": 4.2206292152404785,
      "learning_rate": 2e-05,
      "loss": 1.8528,
      "step": 1
    },
    {
      "epoch": 0.00098015192354815,
      "eval_loss": 1.81155264377594,
      "eval_runtime": 15.4398,
      "eval_samples_per_second": 27.85,
      "eval_steps_per_second": 13.925,
      "step": 1
    },
    {
      "epoch": 0.0019603038470963,
      "grad_norm": 4.969163417816162,
      "learning_rate": 4e-05,
      "loss": 2.028,
      "step": 2
    },
    {
      "epoch": 0.00294045577064445,
      "grad_norm": 4.201023578643799,
      "learning_rate": 6e-05,
      "loss": 1.8649,
      "step": 3
    },
    {
      "epoch": 0.00294045577064445,
      "eval_loss": 1.6717958450317383,
      "eval_runtime": 15.4361,
      "eval_samples_per_second": 27.857,
      "eval_steps_per_second": 13.928,
      "step": 3
    },
    {
      "epoch": 0.0039206076941926,
      "grad_norm": 3.7689459323883057,
      "learning_rate": 8e-05,
      "loss": 1.4737,
      "step": 4
    },
    {
      "epoch": 0.00490075961774075,
      "grad_norm": 3.5083603858947754,
      "learning_rate": 0.0001,
      "loss": 1.3285,
      "step": 5
    },
    {
      "epoch": 0.0058809115412889,
      "grad_norm": 3.1085927486419678,
      "learning_rate": 0.00012,
      "loss": 1.2179,
      "step": 6
    },
    {
      "epoch": 0.0058809115412889,
      "eval_loss": 0.8347481489181519,
      "eval_runtime": 15.465,
      "eval_samples_per_second": 27.805,
      "eval_steps_per_second": 13.902,
      "step": 6
    },
    {
      "epoch": 0.00686106346483705,
      "grad_norm": 2.3984503746032715,
      "learning_rate": 0.00014,
      "loss": 0.743,
      "step": 7
    },
    {
      "epoch": 0.0078412153883852,
      "grad_norm": 2.014052152633667,
      "learning_rate": 0.00016,
      "loss": 0.6128,
      "step": 8
    },
    {
      "epoch": 0.008821367311933349,
      "grad_norm": 1.7296559810638428,
      "learning_rate": 0.00018,
      "loss": 0.4875,
      "step": 9
    },
    {
      "epoch": 0.008821367311933349,
      "eval_loss": 0.4390629231929779,
      "eval_runtime": 15.4545,
      "eval_samples_per_second": 27.824,
      "eval_steps_per_second": 13.912,
      "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": 1815536845455360.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}