last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0005088864292711474,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.088864292711474e-05,
      "grad_norm": 6.617193222045898,
      "learning_rate": 2e-05,
      "loss": 4.3838,
      "step": 1
    },
    {
      "epoch": 5.088864292711474e-05,
      "eval_loss": 4.6908650398254395,
      "eval_runtime": 1200.5721,
      "eval_samples_per_second": 6.893,
      "eval_steps_per_second": 3.447,
      "step": 1
    },
    {
      "epoch": 0.00010177728585422949,
      "grad_norm": 5.9834723472595215,
      "learning_rate": 4e-05,
      "loss": 4.1114,
      "step": 2
    },
    {
      "epoch": 0.00015266592878134422,
      "grad_norm": 6.40416145324707,
      "learning_rate": 6e-05,
      "loss": 4.3085,
      "step": 3
    },
    {
      "epoch": 0.00015266592878134422,
      "eval_loss": 4.5709028244018555,
      "eval_runtime": 1208.4645,
      "eval_samples_per_second": 6.848,
      "eval_steps_per_second": 3.424,
      "step": 3
    },
    {
      "epoch": 0.00020355457170845897,
      "grad_norm": 6.498128414154053,
      "learning_rate": 8e-05,
      "loss": 4.2921,
      "step": 4
    },
    {
      "epoch": 0.0002544432146355737,
      "grad_norm": 7.741085052490234,
      "learning_rate": 0.0001,
      "loss": 4.3936,
      "step": 5
    },
    {
      "epoch": 0.00030533185756268844,
      "grad_norm": 5.888733863830566,
      "learning_rate": 0.00012,
      "loss": 3.6026,
      "step": 6
    },
    {
      "epoch": 0.00030533185756268844,
      "eval_loss": 3.5490193367004395,
      "eval_runtime": 1208.0744,
      "eval_samples_per_second": 6.85,
      "eval_steps_per_second": 3.425,
      "step": 6
    },
    {
      "epoch": 0.00035622050048980317,
      "grad_norm": 6.340505123138428,
      "learning_rate": 0.00014,
      "loss": 3.6984,
      "step": 7
    },
    {
      "epoch": 0.00040710914341691794,
      "grad_norm": 5.083515644073486,
      "learning_rate": 0.00016,
      "loss": 3.2718,
      "step": 8
    },
    {
      "epoch": 0.00045799778634403267,
      "grad_norm": 3.3060462474823,
      "learning_rate": 0.00018,
      "loss": 2.8094,
      "step": 9
    },
    {
      "epoch": 0.00045799778634403267,
      "eval_loss": 2.9464833736419678,
      "eval_runtime": 1207.6921,
      "eval_samples_per_second": 6.852,
      "eval_steps_per_second": 3.426,
      "step": 9
    },
    {
      "epoch": 0.0005088864292711474,
      "grad_norm": 4.704765319824219,
      "learning_rate": 0.0002,
      "loss": 2.5608,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3822182832537600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}