last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03759398496240601,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0037593984962406013,
      "grad_norm": 2.4439609050750732,
      "learning_rate": 2e-05,
      "loss": 5.3841,
      "step": 1
    },
    {
      "epoch": 0.0037593984962406013,
      "eval_loss": 1.6148337125778198,
      "eval_runtime": 33.6548,
      "eval_samples_per_second": 3.328,
      "eval_steps_per_second": 1.664,
      "step": 1
    },
    {
      "epoch": 0.007518796992481203,
      "grad_norm": 2.6649999618530273,
      "learning_rate": 4e-05,
      "loss": 7.2761,
      "step": 2
    },
    {
      "epoch": 0.011278195488721804,
      "grad_norm": 2.454400062561035,
      "learning_rate": 6e-05,
      "loss": 7.3359,
      "step": 3
    },
    {
      "epoch": 0.011278195488721804,
      "eval_loss": 1.6110070943832397,
      "eval_runtime": 33.5914,
      "eval_samples_per_second": 3.334,
      "eval_steps_per_second": 1.667,
      "step": 3
    },
    {
      "epoch": 0.015037593984962405,
      "grad_norm": 2.5507595539093018,
      "learning_rate": 8e-05,
      "loss": 4.5487,
      "step": 4
    },
    {
      "epoch": 0.018796992481203006,
      "grad_norm": 3.046490430831909,
      "learning_rate": 0.0001,
      "loss": 6.1722,
      "step": 5
    },
    {
      "epoch": 0.022556390977443608,
      "grad_norm": 2.6542139053344727,
      "learning_rate": 0.00012,
      "loss": 7.2555,
      "step": 6
    },
    {
      "epoch": 0.022556390977443608,
      "eval_loss": 1.5455684661865234,
      "eval_runtime": 33.649,
      "eval_samples_per_second": 3.328,
      "eval_steps_per_second": 1.664,
      "step": 6
    },
    {
      "epoch": 0.02631578947368421,
      "grad_norm": 3.145287275314331,
      "learning_rate": 0.00014,
      "loss": 4.8095,
      "step": 7
    },
    {
      "epoch": 0.03007518796992481,
      "grad_norm": 4.9834747314453125,
      "learning_rate": 0.00016,
      "loss": 5.3171,
      "step": 8
    },
    {
      "epoch": 0.03383458646616541,
      "grad_norm": 2.429388999938965,
      "learning_rate": 0.00018,
      "loss": 2.8386,
      "step": 9
    },
    {
      "epoch": 0.03383458646616541,
      "eval_loss": 1.2319965362548828,
      "eval_runtime": 33.5889,
      "eval_samples_per_second": 3.334,
      "eval_steps_per_second": 1.667,
      "step": 9
    },
    {
      "epoch": 0.03759398496240601,
      "grad_norm": 2.8795113563537598,
      "learning_rate": 0.0002,
      "loss": 5.3087,
      "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": 5620048142008320.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}