last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.001222643354933366,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00012226433549333658,
      "grad_norm": 1.249587893486023,
      "learning_rate": 2e-05,
      "loss": 3.0705,
      "step": 1
    },
    {
      "epoch": 0.00012226433549333658,
      "eval_loss": 2.9430017471313477,
      "eval_runtime": 607.8703,
      "eval_samples_per_second": 5.666,
      "eval_steps_per_second": 2.833,
      "step": 1
    },
    {
      "epoch": 0.00024452867098667317,
      "grad_norm": 1.208764672279358,
      "learning_rate": 4e-05,
      "loss": 3.1993,
      "step": 2
    },
    {
      "epoch": 0.0003667930064800098,
      "grad_norm": 1.2434773445129395,
      "learning_rate": 6e-05,
      "loss": 2.9123,
      "step": 3
    },
    {
      "epoch": 0.0003667930064800098,
      "eval_loss": 2.9410884380340576,
      "eval_runtime": 582.6295,
      "eval_samples_per_second": 5.911,
      "eval_steps_per_second": 2.956,
      "step": 3
    },
    {
      "epoch": 0.0004890573419733463,
      "grad_norm": 1.2777798175811768,
      "learning_rate": 8e-05,
      "loss": 2.7318,
      "step": 4
    },
    {
      "epoch": 0.000611321677466683,
      "grad_norm": 0.9337401390075684,
      "learning_rate": 0.0001,
      "loss": 2.1764,
      "step": 5
    },
    {
      "epoch": 0.0007335860129600196,
      "grad_norm": 1.0842753648757935,
      "learning_rate": 0.00012,
      "loss": 2.9188,
      "step": 6
    },
    {
      "epoch": 0.0007335860129600196,
      "eval_loss": 2.9117157459259033,
      "eval_runtime": 583.5461,
      "eval_samples_per_second": 5.902,
      "eval_steps_per_second": 2.951,
      "step": 6
    },
    {
      "epoch": 0.0008558503484533561,
      "grad_norm": 1.3478213548660278,
      "learning_rate": 0.00014,
      "loss": 2.9173,
      "step": 7
    },
    {
      "epoch": 0.0009781146839466927,
      "grad_norm": 1.4442219734191895,
      "learning_rate": 0.00016,
      "loss": 3.3116,
      "step": 8
    },
    {
      "epoch": 0.0011003790194400293,
      "grad_norm": 1.2907761335372925,
      "learning_rate": 0.00018,
      "loss": 2.5688,
      "step": 9
    },
    {
      "epoch": 0.0011003790194400293,
      "eval_loss": 2.823617935180664,
      "eval_runtime": 606.1784,
      "eval_samples_per_second": 5.681,
      "eval_steps_per_second": 2.841,
      "step": 9
    },
    {
      "epoch": 0.001222643354933366,
      "grad_norm": 1.5255097150802612,
      "learning_rate": 0.0002,
      "loss": 2.6961,
      "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": 1935445139128320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}