last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005647324579980234,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005647324579980235,
      "grad_norm": 6.657321453094482,
      "learning_rate": 2e-05,
      "loss": 3.0027,
      "step": 1
    },
    {
      "epoch": 0.0005647324579980235,
      "eval_loss": 2.7413086891174316,
      "eval_runtime": 209.2902,
      "eval_samples_per_second": 3.564,
      "eval_steps_per_second": 1.782,
      "step": 1
    },
    {
      "epoch": 0.001129464915996047,
      "grad_norm": 4.548604965209961,
      "learning_rate": 4e-05,
      "loss": 2.5806,
      "step": 2
    },
    {
      "epoch": 0.0016941973739940702,
      "grad_norm": 4.600951194763184,
      "learning_rate": 6e-05,
      "loss": 3.2098,
      "step": 3
    },
    {
      "epoch": 0.0016941973739940702,
      "eval_loss": 2.6281063556671143,
      "eval_runtime": 212.6516,
      "eval_samples_per_second": 3.508,
      "eval_steps_per_second": 1.754,
      "step": 3
    },
    {
      "epoch": 0.002258929831992094,
      "grad_norm": 5.961012840270996,
      "learning_rate": 8e-05,
      "loss": 3.546,
      "step": 4
    },
    {
      "epoch": 0.002823662289990117,
      "grad_norm": 3.73958158493042,
      "learning_rate": 0.0001,
      "loss": 3.039,
      "step": 5
    },
    {
      "epoch": 0.0033883947479881405,
      "grad_norm": 3.5475358963012695,
      "learning_rate": 0.00012,
      "loss": 2.3823,
      "step": 6
    },
    {
      "epoch": 0.0033883947479881405,
      "eval_loss": 2.191272020339966,
      "eval_runtime": 211.8851,
      "eval_samples_per_second": 3.521,
      "eval_steps_per_second": 1.76,
      "step": 6
    },
    {
      "epoch": 0.003953127205986164,
      "grad_norm": 2.7113640308380127,
      "learning_rate": 0.00014,
      "loss": 2.2146,
      "step": 7
    },
    {
      "epoch": 0.004517859663984188,
      "grad_norm": 2.97259521484375,
      "learning_rate": 0.00016,
      "loss": 1.9369,
      "step": 8
    },
    {
      "epoch": 0.005082592121982211,
      "grad_norm": 3.431048631668091,
      "learning_rate": 0.00018,
      "loss": 1.8664,
      "step": 9
    },
    {
      "epoch": 0.005082592121982211,
      "eval_loss": 2.243478536605835,
      "eval_runtime": 211.702,
      "eval_samples_per_second": 3.524,
      "eval_steps_per_second": 1.762,
      "step": 9
    },
    {
      "epoch": 0.005647324579980234,
      "grad_norm": 4.549256801605225,
      "learning_rate": 0.0002,
      "loss": 1.8522,
      "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": 1.533788723085312e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}