last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00040280351244662856,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.028035124466285e-05,
      "grad_norm": 8.708857536315918,
      "learning_rate": 2e-05,
      "loss": 5.2158,
      "step": 1
    },
    {
      "epoch": 4.028035124466285e-05,
      "eval_loss": 1.1413770914077759,
      "eval_runtime": 126.7757,
      "eval_samples_per_second": 82.461,
      "eval_steps_per_second": 41.23,
      "step": 1
    },
    {
      "epoch": 8.05607024893257e-05,
      "grad_norm": 9.573758125305176,
      "learning_rate": 4e-05,
      "loss": 4.4704,
      "step": 2
    },
    {
      "epoch": 0.00012084105373398856,
      "grad_norm": 9.162515640258789,
      "learning_rate": 6e-05,
      "loss": 4.0082,
      "step": 3
    },
    {
      "epoch": 0.00012084105373398856,
      "eval_loss": 1.1266303062438965,
      "eval_runtime": 127.5385,
      "eval_samples_per_second": 81.967,
      "eval_steps_per_second": 40.984,
      "step": 3
    },
    {
      "epoch": 0.0001611214049786514,
      "grad_norm": 7.617592811584473,
      "learning_rate": 8e-05,
      "loss": 4.689,
      "step": 4
    },
    {
      "epoch": 0.00020140175622331428,
      "grad_norm": 7.327009201049805,
      "learning_rate": 0.0001,
      "loss": 3.9992,
      "step": 5
    },
    {
      "epoch": 0.00024168210746797713,
      "grad_norm": 7.231747150421143,
      "learning_rate": 0.00012,
      "loss": 4.3443,
      "step": 6
    },
    {
      "epoch": 0.00024168210746797713,
      "eval_loss": 0.9814525842666626,
      "eval_runtime": 126.3317,
      "eval_samples_per_second": 82.75,
      "eval_steps_per_second": 41.375,
      "step": 6
    },
    {
      "epoch": 0.00028196245871264,
      "grad_norm": 6.902768611907959,
      "learning_rate": 0.00014,
      "loss": 3.3575,
      "step": 7
    },
    {
      "epoch": 0.0003222428099573028,
      "grad_norm": 7.082916736602783,
      "learning_rate": 0.00016,
      "loss": 3.3855,
      "step": 8
    },
    {
      "epoch": 0.0003625231612019657,
      "grad_norm": 7.2913899421691895,
      "learning_rate": 0.00018,
      "loss": 2.9172,
      "step": 9
    },
    {
      "epoch": 0.0003625231612019657,
      "eval_loss": 0.6874945759773254,
      "eval_runtime": 126.3353,
      "eval_samples_per_second": 82.748,
      "eval_steps_per_second": 41.374,
      "step": 9
    },
    {
      "epoch": 0.00040280351244662856,
      "grad_norm": 7.292717456817627,
      "learning_rate": 0.0002,
      "loss": 2.9434,
      "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": 75069652992000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}