last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.01638672675133142,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0016386726751331422,
      "grad_norm": 1.8315523862838745,
      "learning_rate": 2e-05,
      "loss": 1.8684,
      "step": 1
    },
    {
      "epoch": 0.0016386726751331422,
      "eval_loss": 1.9507054090499878,
      "eval_runtime": 72.8095,
      "eval_samples_per_second": 3.53,
      "eval_steps_per_second": 1.772,
      "step": 1
    },
    {
      "epoch": 0.0032773453502662844,
      "grad_norm": 1.7030572891235352,
      "learning_rate": 4e-05,
      "loss": 1.9973,
      "step": 2
    },
    {
      "epoch": 0.004916018025399427,
      "grad_norm": 1.5281317234039307,
      "learning_rate": 6e-05,
      "loss": 2.0103,
      "step": 3
    },
    {
      "epoch": 0.004916018025399427,
      "eval_loss": 1.9461408853530884,
      "eval_runtime": 76.6733,
      "eval_samples_per_second": 3.352,
      "eval_steps_per_second": 1.682,
      "step": 3
    },
    {
      "epoch": 0.006554690700532569,
      "grad_norm": 1.6242179870605469,
      "learning_rate": 8e-05,
      "loss": 2.0485,
      "step": 4
    },
    {
      "epoch": 0.00819336337566571,
      "grad_norm": 1.4690097570419312,
      "learning_rate": 0.0001,
      "loss": 1.8352,
      "step": 5
    },
    {
      "epoch": 0.009832036050798854,
      "grad_norm": 1.6833231449127197,
      "learning_rate": 0.00012,
      "loss": 1.8784,
      "step": 6
    },
    {
      "epoch": 0.009832036050798854,
      "eval_loss": 1.8704971075057983,
      "eval_runtime": 79.2873,
      "eval_samples_per_second": 3.241,
      "eval_steps_per_second": 1.627,
      "step": 6
    },
    {
      "epoch": 0.011470708725931995,
      "grad_norm": 1.779176950454712,
      "learning_rate": 0.00014,
      "loss": 2.1033,
      "step": 7
    },
    {
      "epoch": 0.013109381401065138,
      "grad_norm": 1.6860111951828003,
      "learning_rate": 0.00016,
      "loss": 1.7718,
      "step": 8
    },
    {
      "epoch": 0.014748054076198279,
      "grad_norm": 1.8793421983718872,
      "learning_rate": 0.00018,
      "loss": 1.7841,
      "step": 9
    },
    {
      "epoch": 0.014748054076198279,
      "eval_loss": 1.7667855024337769,
      "eval_runtime": 78.8217,
      "eval_samples_per_second": 3.261,
      "eval_steps_per_second": 1.637,
      "step": 9
    },
    {
      "epoch": 0.01638672675133142,
      "grad_norm": 1.7251274585723877,
      "learning_rate": 0.0002,
      "loss": 1.6624,
      "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": 1935445139128320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}