last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.013214403700033036,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013214403700033035,
      "grad_norm": 35878.53515625,
      "learning_rate": 2e-05,
      "loss": 42.0391,
      "step": 1
    },
    {
      "epoch": 0.0013214403700033035,
      "eval_loss": 10.801332473754883,
      "eval_runtime": 37.0602,
      "eval_samples_per_second": 8.608,
      "eval_steps_per_second": 4.317,
      "step": 1
    },
    {
      "epoch": 0.002642880740006607,
      "grad_norm": 40164.90625,
      "learning_rate": 4e-05,
      "loss": 41.0703,
      "step": 2
    },
    {
      "epoch": 0.003964321110009911,
      "grad_norm": 22879.40234375,
      "learning_rate": 6e-05,
      "loss": 42.3828,
      "step": 3
    },
    {
      "epoch": 0.003964321110009911,
      "eval_loss": 10.28499698638916,
      "eval_runtime": 36.8196,
      "eval_samples_per_second": 8.664,
      "eval_steps_per_second": 4.346,
      "step": 3
    },
    {
      "epoch": 0.005285761480013214,
      "grad_norm": 26312.224609375,
      "learning_rate": 8e-05,
      "loss": 39.375,
      "step": 4
    },
    {
      "epoch": 0.006607201850016518,
      "grad_norm": 30341.833984375,
      "learning_rate": 0.0001,
      "loss": 42.0781,
      "step": 5
    },
    {
      "epoch": 0.007928642220019821,
      "grad_norm": 72681.9140625,
      "learning_rate": 0.00012,
      "loss": 42.0234,
      "step": 6
    },
    {
      "epoch": 0.007928642220019821,
      "eval_loss": 9.394837379455566,
      "eval_runtime": 36.7072,
      "eval_samples_per_second": 8.69,
      "eval_steps_per_second": 4.359,
      "step": 6
    },
    {
      "epoch": 0.009250082590023126,
      "grad_norm": 22986.171875,
      "learning_rate": 0.00014,
      "loss": 30.3867,
      "step": 7
    },
    {
      "epoch": 0.010571522960026428,
      "grad_norm": 9130.1826171875,
      "learning_rate": 0.00016,
      "loss": 33.0547,
      "step": 8
    },
    {
      "epoch": 0.011892963330029732,
      "grad_norm": 27018.677734375,
      "learning_rate": 0.00018,
      "loss": 28.6055,
      "step": 9
    },
    {
      "epoch": 0.011892963330029732,
      "eval_loss": 7.373456954956055,
      "eval_runtime": 32.2691,
      "eval_samples_per_second": 9.886,
      "eval_steps_per_second": 4.958,
      "step": 9
    },
    {
      "epoch": 0.013214403700033036,
      "grad_norm": 43709.67578125,
      "learning_rate": 0.0002,
      "loss": 25.9141,
      "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": 15385378160640.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}