last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0014875702066387587,
  "eval_steps": 10,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.814282581125023e-05,
      "eval_loss": 0.48583489656448364,
      "eval_runtime": 1288.389,
      "eval_samples_per_second": 8.568,
      "eval_steps_per_second": 4.284,
      "step": 1
    },
    {
      "epoch": 0.00019071412905625112,
      "grad_norm": 4.6333489418029785,
      "learning_rate": 5e-05,
      "loss": 1.8997,
      "step": 5
    },
    {
      "epoch": 0.00038142825811250225,
      "grad_norm": 3.166757822036743,
      "learning_rate": 0.0001,
      "loss": 1.7327,
      "step": 10
    },
    {
      "epoch": 0.00038142825811250225,
      "eval_loss": 0.39591124653816223,
      "eval_runtime": 1292.9351,
      "eval_samples_per_second": 8.538,
      "eval_steps_per_second": 4.269,
      "step": 10
    },
    {
      "epoch": 0.0005721423871687534,
      "grad_norm": 2.7471635341644287,
      "learning_rate": 9.619397662556435e-05,
      "loss": 1.522,
      "step": 15
    },
    {
      "epoch": 0.0007628565162250045,
      "grad_norm": 2.795982599258423,
      "learning_rate": 8.535533905932738e-05,
      "loss": 1.356,
      "step": 20
    },
    {
      "epoch": 0.0007628565162250045,
      "eval_loss": 0.35060152411460876,
      "eval_runtime": 1298.9399,
      "eval_samples_per_second": 8.498,
      "eval_steps_per_second": 4.25,
      "step": 20
    },
    {
      "epoch": 0.0009535706452812557,
      "grad_norm": 2.849712371826172,
      "learning_rate": 6.91341716182545e-05,
      "loss": 1.4202,
      "step": 25
    },
    {
      "epoch": 0.0011442847743375068,
      "grad_norm": 2.5111687183380127,
      "learning_rate": 5e-05,
      "loss": 1.2296,
      "step": 30
    },
    {
      "epoch": 0.0011442847743375068,
      "eval_loss": 0.3362519145011902,
      "eval_runtime": 1297.0349,
      "eval_samples_per_second": 8.511,
      "eval_steps_per_second": 4.256,
      "step": 30
    },
    {
      "epoch": 0.001334998903393758,
      "grad_norm": 2.6420953273773193,
      "learning_rate": 3.086582838174551e-05,
      "loss": 1.4156,
      "step": 35
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3670824330395648e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}