last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.001554001554001554,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001554001554001554,
      "grad_norm": 2.6164662837982178,
      "learning_rate": 2e-05,
      "loss": 5.5903,
      "step": 1
    },
    {
      "epoch": 0.0001554001554001554,
      "eval_loss": 5.979092121124268,
      "eval_runtime": 119.2544,
      "eval_samples_per_second": 22.725,
      "eval_steps_per_second": 11.362,
      "step": 1
    },
    {
      "epoch": 0.0003108003108003108,
      "grad_norm": 1.883692741394043,
      "learning_rate": 4e-05,
      "loss": 4.2089,
      "step": 2
    },
    {
      "epoch": 0.0004662004662004662,
      "grad_norm": 2.9674272537231445,
      "learning_rate": 6e-05,
      "loss": 5.9288,
      "step": 3
    },
    {
      "epoch": 0.0004662004662004662,
      "eval_loss": 5.970859527587891,
      "eval_runtime": 117.9013,
      "eval_samples_per_second": 22.985,
      "eval_steps_per_second": 11.493,
      "step": 3
    },
    {
      "epoch": 0.0006216006216006216,
      "grad_norm": 2.538806676864624,
      "learning_rate": 8e-05,
      "loss": 6.3257,
      "step": 4
    },
    {
      "epoch": 0.000777000777000777,
      "grad_norm": 1.9700177907943726,
      "learning_rate": 0.0001,
      "loss": 4.3584,
      "step": 5
    },
    {
      "epoch": 0.0009324009324009324,
      "grad_norm": 3.550747871398926,
      "learning_rate": 0.00012,
      "loss": 5.8421,
      "step": 6
    },
    {
      "epoch": 0.0009324009324009324,
      "eval_loss": 5.807605743408203,
      "eval_runtime": 117.9373,
      "eval_samples_per_second": 22.978,
      "eval_steps_per_second": 11.489,
      "step": 6
    },
    {
      "epoch": 0.0010878010878010878,
      "grad_norm": 2.7380995750427246,
      "learning_rate": 0.00014,
      "loss": 4.7198,
      "step": 7
    },
    {
      "epoch": 0.0012432012432012432,
      "grad_norm": 3.2857611179351807,
      "learning_rate": 0.00016,
      "loss": 5.6143,
      "step": 8
    },
    {
      "epoch": 0.0013986013986013986,
      "grad_norm": 3.5267786979675293,
      "learning_rate": 0.00018,
      "loss": 4.7434,
      "step": 9
    },
    {
      "epoch": 0.0013986013986013986,
      "eval_loss": 5.1117706298828125,
      "eval_runtime": 117.98,
      "eval_samples_per_second": 22.97,
      "eval_steps_per_second": 11.485,
      "step": 9
    },
    {
      "epoch": 0.001554001554001554,
      "grad_norm": 3.7728404998779297,
      "learning_rate": 0.0002,
      "loss": 4.7718,
      "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": 2728107266015232.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}