last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007654688496704231,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008505209440782479,
      "grad_norm": 7.013522148132324,
      "learning_rate": 2e-05,
      "loss": 11.2339,
      "step": 1
    },
    {
      "epoch": 0.0008505209440782479,
      "eval_loss": 2.182969570159912,
      "eval_runtime": 18.5884,
      "eval_samples_per_second": 26.63,
      "eval_steps_per_second": 13.342,
      "step": 1
    },
    {
      "epoch": 0.0017010418881564958,
      "grad_norm": 2.597937822341919,
      "learning_rate": 4e-05,
      "loss": 8.1504,
      "step": 2
    },
    {
      "epoch": 0.0025515628322347436,
      "grad_norm": 4.277220726013184,
      "learning_rate": 6e-05,
      "loss": 9.4615,
      "step": 3
    },
    {
      "epoch": 0.0025515628322347436,
      "eval_loss": 2.1793859004974365,
      "eval_runtime": 18.5435,
      "eval_samples_per_second": 26.694,
      "eval_steps_per_second": 13.374,
      "step": 3
    },
    {
      "epoch": 0.0034020837763129915,
      "grad_norm": 4.113152980804443,
      "learning_rate": 8e-05,
      "loss": 8.655,
      "step": 4
    },
    {
      "epoch": 0.00425260472039124,
      "grad_norm": 4.113445281982422,
      "learning_rate": 0.0001,
      "loss": 8.6883,
      "step": 5
    },
    {
      "epoch": 0.005103125664469487,
      "grad_norm": 3.8714869022369385,
      "learning_rate": 0.00012,
      "loss": 8.7348,
      "step": 6
    },
    {
      "epoch": 0.005103125664469487,
      "eval_loss": 2.145993709564209,
      "eval_runtime": 18.4695,
      "eval_samples_per_second": 26.801,
      "eval_steps_per_second": 13.428,
      "step": 6
    },
    {
      "epoch": 0.005953646608547736,
      "grad_norm": 4.330378532409668,
      "learning_rate": 0.00014,
      "loss": 8.4749,
      "step": 7
    },
    {
      "epoch": 0.006804167552625983,
      "grad_norm": 4.18934440612793,
      "learning_rate": 0.00016,
      "loss": 8.491,
      "step": 8
    },
    {
      "epoch": 0.007654688496704231,
      "grad_norm": 4.927352428436279,
      "learning_rate": 0.00018,
      "loss": 9.3576,
      "step": 9
    },
    {
      "epoch": 0.007654688496704231,
      "eval_loss": 2.026862621307373,
      "eval_runtime": 18.3928,
      "eval_samples_per_second": 26.913,
      "eval_steps_per_second": 13.484,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1506578150719488.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}