last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004081632653061225,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00040816326530612246,
      "grad_norm": 1.3611687421798706,
      "learning_rate": 2e-05,
      "loss": 3.5911,
      "step": 1
    },
    {
      "epoch": 0.00040816326530612246,
      "eval_loss": 1.0385665893554688,
      "eval_runtime": 42.6964,
      "eval_samples_per_second": 24.171,
      "eval_steps_per_second": 12.085,
      "step": 1
    },
    {
      "epoch": 0.0008163265306122449,
      "grad_norm": 1.398386836051941,
      "learning_rate": 4e-05,
      "loss": 3.761,
      "step": 2
    },
    {
      "epoch": 0.0012244897959183673,
      "grad_norm": 1.3859976530075073,
      "learning_rate": 6e-05,
      "loss": 3.6131,
      "step": 3
    },
    {
      "epoch": 0.0012244897959183673,
      "eval_loss": 1.0361061096191406,
      "eval_runtime": 41.4119,
      "eval_samples_per_second": 24.92,
      "eval_steps_per_second": 12.46,
      "step": 3
    },
    {
      "epoch": 0.0016326530612244899,
      "grad_norm": 1.6692824363708496,
      "learning_rate": 8e-05,
      "loss": 4.3543,
      "step": 4
    },
    {
      "epoch": 0.0020408163265306124,
      "grad_norm": 1.6006004810333252,
      "learning_rate": 0.0001,
      "loss": 4.6853,
      "step": 5
    },
    {
      "epoch": 0.0024489795918367346,
      "grad_norm": 1.3115887641906738,
      "learning_rate": 0.00012,
      "loss": 3.5072,
      "step": 6
    },
    {
      "epoch": 0.0024489795918367346,
      "eval_loss": 1.0073992013931274,
      "eval_runtime": 41.4927,
      "eval_samples_per_second": 24.872,
      "eval_steps_per_second": 12.436,
      "step": 6
    },
    {
      "epoch": 0.002857142857142857,
      "grad_norm": 1.4988536834716797,
      "learning_rate": 0.00014,
      "loss": 3.4837,
      "step": 7
    },
    {
      "epoch": 0.0032653061224489797,
      "grad_norm": 2.0944082736968994,
      "learning_rate": 0.00016,
      "loss": 4.0673,
      "step": 8
    },
    {
      "epoch": 0.003673469387755102,
      "grad_norm": 1.5092041492462158,
      "learning_rate": 0.00018,
      "loss": 3.5947,
      "step": 9
    },
    {
      "epoch": 0.003673469387755102,
      "eval_loss": 0.9217643737792969,
      "eval_runtime": 41.5439,
      "eval_samples_per_second": 24.841,
      "eval_steps_per_second": 12.421,
      "step": 9
    },
    {
      "epoch": 0.004081632653061225,
      "grad_norm": 1.365379810333252,
      "learning_rate": 0.0002,
      "loss": 3.1071,
      "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": 3566006697984000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}