last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0006326333362621913,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.029259291802126e-05,
      "grad_norm": 0.6534746289253235,
      "learning_rate": 2e-05,
      "loss": 2.9102,
      "step": 1
    },
    {
      "epoch": 7.029259291802126e-05,
      "eval_loss": 3.0221028327941895,
      "eval_runtime": 229.449,
      "eval_samples_per_second": 26.106,
      "eval_steps_per_second": 13.053,
      "step": 1
    },
    {
      "epoch": 0.00014058518583604253,
      "grad_norm": 1.0394901037216187,
      "learning_rate": 4e-05,
      "loss": 3.4188,
      "step": 2
    },
    {
      "epoch": 0.00021087777875406378,
      "grad_norm": 0.607863187789917,
      "learning_rate": 6e-05,
      "loss": 3.0918,
      "step": 3
    },
    {
      "epoch": 0.00021087777875406378,
      "eval_loss": 3.014835834503174,
      "eval_runtime": 228.4096,
      "eval_samples_per_second": 26.225,
      "eval_steps_per_second": 13.112,
      "step": 3
    },
    {
      "epoch": 0.00028117037167208506,
      "grad_norm": 0.725269615650177,
      "learning_rate": 8e-05,
      "loss": 3.004,
      "step": 4
    },
    {
      "epoch": 0.00035146296459010633,
      "grad_norm": 0.7683107256889343,
      "learning_rate": 0.0001,
      "loss": 3.303,
      "step": 5
    },
    {
      "epoch": 0.00042175555750812756,
      "grad_norm": 0.743407130241394,
      "learning_rate": 0.00012,
      "loss": 2.7819,
      "step": 6
    },
    {
      "epoch": 0.00042175555750812756,
      "eval_loss": 2.949836015701294,
      "eval_runtime": 228.2654,
      "eval_samples_per_second": 26.241,
      "eval_steps_per_second": 13.121,
      "step": 6
    },
    {
      "epoch": 0.0004920481504261489,
      "grad_norm": 0.9011140465736389,
      "learning_rate": 0.00014,
      "loss": 2.8048,
      "step": 7
    },
    {
      "epoch": 0.0005623407433441701,
      "grad_norm": 0.8253324031829834,
      "learning_rate": 0.00016,
      "loss": 2.7209,
      "step": 8
    },
    {
      "epoch": 0.0006326333362621913,
      "grad_norm": 1.0869470834732056,
      "learning_rate": 0.00018,
      "loss": 3.0211,
      "step": 9
    },
    {
      "epoch": 0.0006326333362621913,
      "eval_loss": 2.7636959552764893,
      "eval_runtime": 228.2803,
      "eval_samples_per_second": 26.24,
      "eval_steps_per_second": 13.12,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2035916419891200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}