last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004024684733029246,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00013415615776764153,
      "eval_loss": NaN,
      "eval_runtime": 562.1292,
      "eval_samples_per_second": 5.584,
      "eval_steps_per_second": 2.793,
      "step": 1
    },
    {
      "epoch": 0.0004024684733029246,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0006707807888382077,
      "eval_loss": NaN,
      "eval_runtime": 563.7868,
      "eval_samples_per_second": 5.568,
      "eval_steps_per_second": 2.785,
      "step": 5
    },
    {
      "epoch": 0.0008049369466058492,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0012074054199087737,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0013415615776764154,
      "eval_loss": NaN,
      "eval_runtime": 563.5642,
      "eval_samples_per_second": 5.57,
      "eval_steps_per_second": 2.786,
      "step": 10
    },
    {
      "epoch": 0.0016098738932116983,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.002012342366514623,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.002012342366514623,
      "eval_loss": NaN,
      "eval_runtime": 562.0579,
      "eval_samples_per_second": 5.585,
      "eval_steps_per_second": 2.793,
      "step": 15
    },
    {
      "epoch": 0.0024148108398175474,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.002683123155352831,
      "eval_loss": NaN,
      "eval_runtime": 561.9752,
      "eval_samples_per_second": 5.586,
      "eval_steps_per_second": 2.794,
      "step": 20
    },
    {
      "epoch": 0.002817279313120472,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0032197477864233967,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.0033539039441910384,
      "eval_loss": NaN,
      "eval_runtime": 562.084,
      "eval_samples_per_second": 5.585,
      "eval_steps_per_second": 2.793,
      "step": 25
    },
    {
      "epoch": 0.0036222162597263213,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.004024684733029246,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.004024684733029246,
      "eval_loss": NaN,
      "eval_runtime": 562.2583,
      "eval_samples_per_second": 5.583,
      "eval_steps_per_second": 2.792,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.0639510886940672e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}