last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.042105263157894736,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004210526315789474,
      "grad_norm": 7.093989372253418,
      "learning_rate": 2e-05,
      "loss": 5.4952,
      "step": 1
    },
    {
      "epoch": 0.004210526315789474,
      "eval_loss": 5.3969902992248535,
      "eval_runtime": 33.7106,
      "eval_samples_per_second": 2.966,
      "eval_steps_per_second": 1.483,
      "step": 1
    },
    {
      "epoch": 0.008421052631578947,
      "grad_norm": 7.5623555183410645,
      "learning_rate": 4e-05,
      "loss": 5.5938,
      "step": 2
    },
    {
      "epoch": 0.01263157894736842,
      "grad_norm": 8.191666603088379,
      "learning_rate": 6e-05,
      "loss": 5.7261,
      "step": 3
    },
    {
      "epoch": 0.01263157894736842,
      "eval_loss": 5.265225410461426,
      "eval_runtime": 33.8391,
      "eval_samples_per_second": 2.955,
      "eval_steps_per_second": 1.478,
      "step": 3
    },
    {
      "epoch": 0.016842105263157894,
      "grad_norm": 7.726444244384766,
      "learning_rate": 8e-05,
      "loss": 5.2674,
      "step": 4
    },
    {
      "epoch": 0.021052631578947368,
      "grad_norm": 8.86080551147461,
      "learning_rate": 0.0001,
      "loss": 5.021,
      "step": 5
    },
    {
      "epoch": 0.02526315789473684,
      "grad_norm": 9.20016098022461,
      "learning_rate": 0.00012,
      "loss": 4.0351,
      "step": 6
    },
    {
      "epoch": 0.02526315789473684,
      "eval_loss": 2.8107352256774902,
      "eval_runtime": 33.8183,
      "eval_samples_per_second": 2.957,
      "eval_steps_per_second": 1.478,
      "step": 6
    },
    {
      "epoch": 0.029473684210526315,
      "grad_norm": 9.21462631225586,
      "learning_rate": 0.00014,
      "loss": 2.8038,
      "step": 7
    },
    {
      "epoch": 0.03368421052631579,
      "grad_norm": 10.491095542907715,
      "learning_rate": 0.00016,
      "loss": 1.4769,
      "step": 8
    },
    {
      "epoch": 0.037894736842105266,
      "grad_norm": 10.90908432006836,
      "learning_rate": 0.00018,
      "loss": 0.9007,
      "step": 9
    },
    {
      "epoch": 0.037894736842105266,
      "eval_loss": 0.18612316250801086,
      "eval_runtime": 33.8894,
      "eval_samples_per_second": 2.951,
      "eval_steps_per_second": 1.475,
      "step": 9
    },
    {
      "epoch": 0.042105263157894736,
      "grad_norm": 3.377394676208496,
      "learning_rate": 0.0002,
      "loss": 0.2287,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5637488844472320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}