last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.40816326530612246,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04081632653061224,
      "grad_norm": 2.1020636558532715,
      "learning_rate": 2e-05,
      "loss": 1.7373,
      "step": 1
    },
    {
      "epoch": 0.04081632653061224,
      "eval_loss": 2.7169904708862305,
      "eval_runtime": 1.7685,
      "eval_samples_per_second": 6.22,
      "eval_steps_per_second": 3.393,
      "step": 1
    },
    {
      "epoch": 0.08163265306122448,
      "grad_norm": 1.407194972038269,
      "learning_rate": 4e-05,
      "loss": 1.2073,
      "step": 2
    },
    {
      "epoch": 0.12244897959183673,
      "grad_norm": 1.6883673667907715,
      "learning_rate": 6e-05,
      "loss": 1.4486,
      "step": 3
    },
    {
      "epoch": 0.12244897959183673,
      "eval_loss": 2.6832287311553955,
      "eval_runtime": 0.382,
      "eval_samples_per_second": 28.799,
      "eval_steps_per_second": 15.708,
      "step": 3
    },
    {
      "epoch": 0.16326530612244897,
      "grad_norm": 1.6880388259887695,
      "learning_rate": 8e-05,
      "loss": 1.3996,
      "step": 4
    },
    {
      "epoch": 0.20408163265306123,
      "grad_norm": 2.286900043487549,
      "learning_rate": 0.0001,
      "loss": 1.3002,
      "step": 5
    },
    {
      "epoch": 0.24489795918367346,
      "grad_norm": 3.153114080429077,
      "learning_rate": 0.00012,
      "loss": 1.2338,
      "step": 6
    },
    {
      "epoch": 0.24489795918367346,
      "eval_loss": 1.9889538288116455,
      "eval_runtime": 0.384,
      "eval_samples_per_second": 28.643,
      "eval_steps_per_second": 15.623,
      "step": 6
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 3.2458138465881348,
      "learning_rate": 0.00014,
      "loss": 1.1605,
      "step": 7
    },
    {
      "epoch": 0.32653061224489793,
      "grad_norm": 3.635159492492676,
      "learning_rate": 0.00016,
      "loss": 0.9088,
      "step": 8
    },
    {
      "epoch": 0.3673469387755102,
      "grad_norm": 2.9575371742248535,
      "learning_rate": 0.00018,
      "loss": 0.6034,
      "step": 9
    },
    {
      "epoch": 0.3673469387755102,
      "eval_loss": 0.5413809418678284,
      "eval_runtime": 0.3785,
      "eval_samples_per_second": 29.065,
      "eval_steps_per_second": 15.854,
      "step": 9
    },
    {
      "epoch": 0.40816326530612246,
      "grad_norm": 5.943121433258057,
      "learning_rate": 0.0002,
      "loss": 0.6172,
      "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": 1647822923366400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}