last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0008520246235116195,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.520246235116194e-05,
      "grad_norm": 11.423274040222168,
      "learning_rate": 2e-05,
      "loss": 1.5967,
      "step": 1
    },
    {
      "epoch": 8.520246235116194e-05,
      "eval_loss": 1.6326349973678589,
      "eval_runtime": 94.7453,
      "eval_samples_per_second": 52.161,
      "eval_steps_per_second": 26.08,
      "step": 1
    },
    {
      "epoch": 0.00017040492470232388,
      "grad_norm": 16.681358337402344,
      "learning_rate": 4e-05,
      "loss": 1.9276,
      "step": 2
    },
    {
      "epoch": 0.00025560738705348587,
      "grad_norm": 13.098152160644531,
      "learning_rate": 6e-05,
      "loss": 1.2446,
      "step": 3
    },
    {
      "epoch": 0.00025560738705348587,
      "eval_loss": 1.537522315979004,
      "eval_runtime": 93.6785,
      "eval_samples_per_second": 52.755,
      "eval_steps_per_second": 26.377,
      "step": 3
    },
    {
      "epoch": 0.00034080984940464777,
      "grad_norm": 8.880377769470215,
      "learning_rate": 8e-05,
      "loss": 1.0321,
      "step": 4
    },
    {
      "epoch": 0.0004260123117558097,
      "grad_norm": 7.2637505531311035,
      "learning_rate": 0.0001,
      "loss": 0.9301,
      "step": 5
    },
    {
      "epoch": 0.0005112147741069717,
      "grad_norm": 10.159381866455078,
      "learning_rate": 0.00012,
      "loss": 1.2338,
      "step": 6
    },
    {
      "epoch": 0.0005112147741069717,
      "eval_loss": 0.8766170144081116,
      "eval_runtime": 93.7618,
      "eval_samples_per_second": 52.708,
      "eval_steps_per_second": 26.354,
      "step": 6
    },
    {
      "epoch": 0.0005964172364581337,
      "grad_norm": 5.066013336181641,
      "learning_rate": 0.00014,
      "loss": 0.5199,
      "step": 7
    },
    {
      "epoch": 0.0006816196988092955,
      "grad_norm": 8.70285701751709,
      "learning_rate": 0.00016,
      "loss": 0.631,
      "step": 8
    },
    {
      "epoch": 0.0007668221611604575,
      "grad_norm": 6.512222766876221,
      "learning_rate": 0.00018,
      "loss": 0.3651,
      "step": 9
    },
    {
      "epoch": 0.0007668221611604575,
      "eval_loss": 0.31702756881713867,
      "eval_runtime": 93.8477,
      "eval_samples_per_second": 52.66,
      "eval_steps_per_second": 26.33,
      "step": 9
    },
    {
      "epoch": 0.0008520246235116195,
      "grad_norm": 3.5256497859954834,
      "learning_rate": 0.0002,
      "loss": 0.3867,
      "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
}