last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.013888888888888888,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001388888888888889,
      "grad_norm": 2.6438210010528564,
      "learning_rate": 2e-05,
      "loss": 2.7967,
      "step": 1
    },
    {
      "epoch": 0.002777777777777778,
      "grad_norm": 2.5853610038757324,
      "learning_rate": 4e-05,
      "loss": 2.5778,
      "step": 2
    },
    {
      "epoch": 0.004166666666666667,
      "grad_norm": 2.5614705085754395,
      "learning_rate": 6e-05,
      "loss": 2.7649,
      "step": 3
    },
    {
      "epoch": 0.004166666666666667,
      "eval_loss": 3.1153173446655273,
      "eval_runtime": 12.5427,
      "eval_samples_per_second": 24.237,
      "eval_steps_per_second": 12.119,
      "step": 3
    },
    {
      "epoch": 0.005555555555555556,
      "grad_norm": 3.3419907093048096,
      "learning_rate": 8e-05,
      "loss": 3.1696,
      "step": 4
    },
    {
      "epoch": 0.006944444444444444,
      "grad_norm": 2.6265079975128174,
      "learning_rate": 0.0001,
      "loss": 2.9326,
      "step": 5
    },
    {
      "epoch": 0.008333333333333333,
      "grad_norm": 2.834522247314453,
      "learning_rate": 0.00012,
      "loss": 3.1474,
      "step": 6
    },
    {
      "epoch": 0.008333333333333333,
      "eval_loss": 2.9960439205169678,
      "eval_runtime": 13.2297,
      "eval_samples_per_second": 22.979,
      "eval_steps_per_second": 11.489,
      "step": 6
    },
    {
      "epoch": 0.009722222222222222,
      "grad_norm": 2.6992099285125732,
      "learning_rate": 0.00014,
      "loss": 2.9552,
      "step": 7
    },
    {
      "epoch": 0.011111111111111112,
      "grad_norm": 2.701378107070923,
      "learning_rate": 0.00016,
      "loss": 2.8407,
      "step": 8
    },
    {
      "epoch": 0.0125,
      "grad_norm": 3.067340135574341,
      "learning_rate": 0.00018,
      "loss": 2.6516,
      "step": 9
    },
    {
      "epoch": 0.0125,
      "eval_loss": 2.6926474571228027,
      "eval_runtime": 13.8071,
      "eval_samples_per_second": 22.018,
      "eval_steps_per_second": 11.009,
      "step": 9
    },
    {
      "epoch": 0.013888888888888888,
      "grad_norm": 3.9425275325775146,
      "learning_rate": 0.0002,
      "loss": 2.7927,
      "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": 89038163804160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}