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": 0.42628446221351624,
      "learning_rate": 2e-05,
      "loss": 0.1517,
      "step": 1
    },
    {
      "epoch": 0.004210526315789474,
      "eval_loss": 0.24416916072368622,
      "eval_runtime": 29.4222,
      "eval_samples_per_second": 3.399,
      "eval_steps_per_second": 1.699,
      "step": 1
    },
    {
      "epoch": 0.008421052631578947,
      "grad_norm": 0.433437317609787,
      "learning_rate": 4e-05,
      "loss": 0.1501,
      "step": 2
    },
    {
      "epoch": 0.01263157894736842,
      "grad_norm": 0.3588545322418213,
      "learning_rate": 6e-05,
      "loss": 0.1181,
      "step": 3
    },
    {
      "epoch": 0.01263157894736842,
      "eval_loss": 0.2361646592617035,
      "eval_runtime": 29.4356,
      "eval_samples_per_second": 3.397,
      "eval_steps_per_second": 1.699,
      "step": 3
    },
    {
      "epoch": 0.016842105263157894,
      "grad_norm": 1.395027756690979,
      "learning_rate": 8e-05,
      "loss": 0.4504,
      "step": 4
    },
    {
      "epoch": 0.021052631578947368,
      "grad_norm": 1.5977227687835693,
      "learning_rate": 0.0001,
      "loss": 0.4201,
      "step": 5
    },
    {
      "epoch": 0.02526315789473684,
      "grad_norm": 1.294385552406311,
      "learning_rate": 0.00012,
      "loss": 0.3502,
      "step": 6
    },
    {
      "epoch": 0.02526315789473684,
      "eval_loss": 0.14960011839866638,
      "eval_runtime": 29.4235,
      "eval_samples_per_second": 3.399,
      "eval_steps_per_second": 1.699,
      "step": 6
    },
    {
      "epoch": 0.029473684210526315,
      "grad_norm": 0.3078320622444153,
      "learning_rate": 0.00014,
      "loss": 0.0808,
      "step": 7
    },
    {
      "epoch": 0.03368421052631579,
      "grad_norm": 0.6034978032112122,
      "learning_rate": 0.00016,
      "loss": 0.0923,
      "step": 8
    },
    {
      "epoch": 0.037894736842105266,
      "grad_norm": 0.5751827955245972,
      "learning_rate": 0.00018,
      "loss": 0.0495,
      "step": 9
    },
    {
      "epoch": 0.037894736842105266,
      "eval_loss": 0.04392697289586067,
      "eval_runtime": 29.5398,
      "eval_samples_per_second": 3.385,
      "eval_steps_per_second": 1.693,
      "step": 9
    },
    {
      "epoch": 0.042105263157894736,
      "grad_norm": 0.9490816593170166,
      "learning_rate": 0.0002,
      "loss": 0.0434,
      "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": 5589685858467840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}