last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0018196706396142299,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.6393412792284596e-05,
      "eval_loss": NaN,
      "eval_runtime": 411.358,
      "eval_samples_per_second": 28.126,
      "eval_steps_per_second": 14.063,
      "step": 1
    },
    {
      "epoch": 0.000181967063961423,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.000363934127922846,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.000363934127922846,
      "eval_loss": NaN,
      "eval_runtime": 425.1705,
      "eval_samples_per_second": 27.213,
      "eval_steps_per_second": 13.606,
      "step": 10
    },
    {
      "epoch": 0.0005459011918842689,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.000727868255845692,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.000727868255845692,
      "eval_loss": NaN,
      "eval_runtime": 453.8405,
      "eval_samples_per_second": 25.494,
      "eval_steps_per_second": 12.747,
      "step": 20
    },
    {
      "epoch": 0.0009098353198071149,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.0010918023837685378,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0010918023837685378,
      "eval_loss": NaN,
      "eval_runtime": 412.8892,
      "eval_samples_per_second": 28.022,
      "eval_steps_per_second": 14.011,
      "step": 30
    },
    {
      "epoch": 0.0012737694477299609,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.001455736511691384,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.001455736511691384,
      "eval_loss": NaN,
      "eval_runtime": 411.0902,
      "eval_samples_per_second": 28.145,
      "eval_steps_per_second": 14.072,
      "step": 40
    },
    {
      "epoch": 0.0016377035756528068,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.0018196706396142299,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.0018196706396142299,
      "eval_loss": NaN,
      "eval_runtime": 410.2429,
      "eval_samples_per_second": 28.203,
      "eval_steps_per_second": 14.101,
      "step": 50
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 716757218623488.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}