last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.014378145219266714,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014378145219266715,
      "grad_norm": 27.511749267578125,
      "learning_rate": 2e-05,
      "loss": 17.122,
      "step": 1
    },
    {
      "epoch": 0.0014378145219266715,
      "eval_loss": 17.40503692626953,
      "eval_runtime": 74.0046,
      "eval_samples_per_second": 3.959,
      "eval_steps_per_second": 1.986,
      "step": 1
    },
    {
      "epoch": 0.002875629043853343,
      "grad_norm": 32.051795959472656,
      "learning_rate": 4e-05,
      "loss": 17.5674,
      "step": 2
    },
    {
      "epoch": 0.004313443565780014,
      "grad_norm": 32.293006896972656,
      "learning_rate": 6e-05,
      "loss": 17.8159,
      "step": 3
    },
    {
      "epoch": 0.004313443565780014,
      "eval_loss": 16.918561935424805,
      "eval_runtime": 74.3408,
      "eval_samples_per_second": 3.941,
      "eval_steps_per_second": 1.977,
      "step": 3
    },
    {
      "epoch": 0.005751258087706686,
      "grad_norm": 24.69898223876953,
      "learning_rate": 8e-05,
      "loss": 16.4401,
      "step": 4
    },
    {
      "epoch": 0.007189072609633357,
      "grad_norm": 25.496450424194336,
      "learning_rate": 0.0001,
      "loss": 16.1233,
      "step": 5
    },
    {
      "epoch": 0.008626887131560028,
      "grad_norm": 28.253496170043945,
      "learning_rate": 0.00012,
      "loss": 14.2261,
      "step": 6
    },
    {
      "epoch": 0.008626887131560028,
      "eval_loss": 12.743063926696777,
      "eval_runtime": 74.5592,
      "eval_samples_per_second": 3.93,
      "eval_steps_per_second": 1.972,
      "step": 6
    },
    {
      "epoch": 0.010064701653486701,
      "grad_norm": 21.585725784301758,
      "learning_rate": 0.00014,
      "loss": 13.1111,
      "step": 7
    },
    {
      "epoch": 0.011502516175413372,
      "grad_norm": 23.57619285583496,
      "learning_rate": 0.00016,
      "loss": 11.0064,
      "step": 8
    },
    {
      "epoch": 0.012940330697340043,
      "grad_norm": 30.479141235351562,
      "learning_rate": 0.00018,
      "loss": 10.0326,
      "step": 9
    },
    {
      "epoch": 0.012940330697340043,
      "eval_loss": 8.028653144836426,
      "eval_runtime": 74.255,
      "eval_samples_per_second": 3.946,
      "eval_steps_per_second": 1.98,
      "step": 9
    },
    {
      "epoch": 0.014378145219266714,
      "grad_norm": 26.972896575927734,
      "learning_rate": 0.0002,
      "loss": 7.7974,
      "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": 2612537673646080.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}