last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0023216669568750362,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00023216669568750363,
      "grad_norm": 2.633310079574585,
      "learning_rate": 2e-05,
      "loss": 2.4602,
      "step": 1
    },
    {
      "epoch": 0.00023216669568750363,
      "eval_loss": 2.4733028411865234,
      "eval_runtime": 198.1616,
      "eval_samples_per_second": 4.577,
      "eval_steps_per_second": 4.577,
      "step": 1
    },
    {
      "epoch": 0.00046433339137500726,
      "grad_norm": 2.2018356323242188,
      "learning_rate": 4e-05,
      "loss": 2.4447,
      "step": 2
    },
    {
      "epoch": 0.0006965000870625109,
      "grad_norm": 2.590804100036621,
      "learning_rate": 6e-05,
      "loss": 2.4806,
      "step": 3
    },
    {
      "epoch": 0.0006965000870625109,
      "eval_loss": 2.3103156089782715,
      "eval_runtime": 197.5106,
      "eval_samples_per_second": 4.592,
      "eval_steps_per_second": 4.592,
      "step": 3
    },
    {
      "epoch": 0.0009286667827500145,
      "grad_norm": 3.1405253410339355,
      "learning_rate": 8e-05,
      "loss": 2.319,
      "step": 4
    },
    {
      "epoch": 0.0011608334784375181,
      "grad_norm": 2.377209424972534,
      "learning_rate": 0.0001,
      "loss": 1.6571,
      "step": 5
    },
    {
      "epoch": 0.0013930001741250218,
      "grad_norm": 3.2652180194854736,
      "learning_rate": 0.00012,
      "loss": 1.4567,
      "step": 6
    },
    {
      "epoch": 0.0013930001741250218,
      "eval_loss": 1.3078285455703735,
      "eval_runtime": 197.6739,
      "eval_samples_per_second": 4.588,
      "eval_steps_per_second": 4.588,
      "step": 6
    },
    {
      "epoch": 0.0016251668698125253,
      "grad_norm": 2.901350975036621,
      "learning_rate": 0.00014,
      "loss": 1.6501,
      "step": 7
    },
    {
      "epoch": 0.001857333565500029,
      "grad_norm": 3.143646478652954,
      "learning_rate": 0.00016,
      "loss": 1.0855,
      "step": 8
    },
    {
      "epoch": 0.0020895002611875328,
      "grad_norm": 2.208082675933838,
      "learning_rate": 0.00018,
      "loss": 0.8768,
      "step": 9
    },
    {
      "epoch": 0.0020895002611875328,
      "eval_loss": 0.6551823616027832,
      "eval_runtime": 197.4739,
      "eval_samples_per_second": 4.593,
      "eval_steps_per_second": 4.593,
      "step": 9
    },
    {
      "epoch": 0.0023216669568750362,
      "grad_norm": 1.9000203609466553,
      "learning_rate": 0.0002,
      "loss": 0.5152,
      "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": 6990386933268480.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}