last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0016354567012838335,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00016354567012838336,
      "grad_norm": 0.853318452835083,
      "learning_rate": 2e-05,
      "loss": 4.1796,
      "step": 1
    },
    {
      "epoch": 0.00016354567012838336,
      "eval_loss": 3.870612859725952,
      "eval_runtime": 117.9532,
      "eval_samples_per_second": 21.831,
      "eval_steps_per_second": 10.92,
      "step": 1
    },
    {
      "epoch": 0.0003270913402567667,
      "grad_norm": 0.7066583633422852,
      "learning_rate": 4e-05,
      "loss": 3.2798,
      "step": 2
    },
    {
      "epoch": 0.00049063701038515,
      "grad_norm": 0.5949627161026001,
      "learning_rate": 6e-05,
      "loss": 2.8722,
      "step": 3
    },
    {
      "epoch": 0.00049063701038515,
      "eval_loss": 3.8621253967285156,
      "eval_runtime": 116.8527,
      "eval_samples_per_second": 22.036,
      "eval_steps_per_second": 11.022,
      "step": 3
    },
    {
      "epoch": 0.0006541826805135334,
      "grad_norm": 1.0430587530136108,
      "learning_rate": 8e-05,
      "loss": 3.5746,
      "step": 4
    },
    {
      "epoch": 0.0008177283506419167,
      "grad_norm": 0.8327656984329224,
      "learning_rate": 0.0001,
      "loss": 3.6405,
      "step": 5
    },
    {
      "epoch": 0.0009812740207703,
      "grad_norm": 1.4123550653457642,
      "learning_rate": 0.00012,
      "loss": 3.7486,
      "step": 6
    },
    {
      "epoch": 0.0009812740207703,
      "eval_loss": 3.747291326522827,
      "eval_runtime": 116.9365,
      "eval_samples_per_second": 22.02,
      "eval_steps_per_second": 11.015,
      "step": 6
    },
    {
      "epoch": 0.0011448196908986834,
      "grad_norm": 1.0004723072052002,
      "learning_rate": 0.00014,
      "loss": 3.2888,
      "step": 7
    },
    {
      "epoch": 0.0013083653610270669,
      "grad_norm": 1.0540218353271484,
      "learning_rate": 0.00016,
      "loss": 3.4393,
      "step": 8
    },
    {
      "epoch": 0.0014719110311554502,
      "grad_norm": 0.7315677404403687,
      "learning_rate": 0.00018,
      "loss": 2.3771,
      "step": 9
    },
    {
      "epoch": 0.0014719110311554502,
      "eval_loss": 3.418764352798462,
      "eval_runtime": 116.9534,
      "eval_samples_per_second": 22.017,
      "eval_steps_per_second": 11.013,
      "step": 9
    },
    {
      "epoch": 0.0016354567012838335,
      "grad_norm": 2.133103847503662,
      "learning_rate": 0.0002,
      "loss": 3.9408,
      "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": 3166204172697600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}