last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.024953212726138492,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002495321272613849,
      "grad_norm": 2.5409350395202637,
      "learning_rate": 2e-05,
      "loss": 1.5255,
      "step": 1
    },
    {
      "epoch": 0.002495321272613849,
      "eval_loss": 1.6741704940795898,
      "eval_runtime": 5.0211,
      "eval_samples_per_second": 33.658,
      "eval_steps_per_second": 16.929,
      "step": 1
    },
    {
      "epoch": 0.004990642545227698,
      "grad_norm": 2.8833096027374268,
      "learning_rate": 4e-05,
      "loss": 1.4442,
      "step": 2
    },
    {
      "epoch": 0.007485963817841547,
      "grad_norm": 2.7640318870544434,
      "learning_rate": 6e-05,
      "loss": 1.8815,
      "step": 3
    },
    {
      "epoch": 0.007485963817841547,
      "eval_loss": 1.6381901502609253,
      "eval_runtime": 5.0134,
      "eval_samples_per_second": 33.71,
      "eval_steps_per_second": 16.955,
      "step": 3
    },
    {
      "epoch": 0.009981285090455396,
      "grad_norm": 2.6299636363983154,
      "learning_rate": 8e-05,
      "loss": 1.4253,
      "step": 4
    },
    {
      "epoch": 0.012476606363069246,
      "grad_norm": 2.9032979011535645,
      "learning_rate": 0.0001,
      "loss": 1.9614,
      "step": 5
    },
    {
      "epoch": 0.014971927635683094,
      "grad_norm": 2.814908266067505,
      "learning_rate": 0.00012,
      "loss": 1.3926,
      "step": 6
    },
    {
      "epoch": 0.014971927635683094,
      "eval_loss": 1.3469496965408325,
      "eval_runtime": 5.1022,
      "eval_samples_per_second": 33.123,
      "eval_steps_per_second": 16.659,
      "step": 6
    },
    {
      "epoch": 0.017467248908296942,
      "grad_norm": 2.572765827178955,
      "learning_rate": 0.00014,
      "loss": 1.1907,
      "step": 7
    },
    {
      "epoch": 0.019962570180910792,
      "grad_norm": 3.265836477279663,
      "learning_rate": 0.00016,
      "loss": 0.9563,
      "step": 8
    },
    {
      "epoch": 0.022457891453524642,
      "grad_norm": 3.0395562648773193,
      "learning_rate": 0.00018,
      "loss": 0.7423,
      "step": 9
    },
    {
      "epoch": 0.022457891453524642,
      "eval_loss": 1.1905124187469482,
      "eval_runtime": 5.1614,
      "eval_samples_per_second": 32.743,
      "eval_steps_per_second": 16.468,
      "step": 9
    },
    {
      "epoch": 0.024953212726138492,
      "grad_norm": 5.918942451477051,
      "learning_rate": 0.0002,
      "loss": 1.4546,
      "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": 500097330708480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}