trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.958904109589041,
  "eval_steps": 500,
  "global_step": 72,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0273972602739726,
      "grad_norm": 0.08680023374887928,
      "learning_rate": 1.25e-06,
      "loss": 0.5277,
      "step": 1
    },
    {
      "epoch": 0.136986301369863,
      "grad_norm": 0.11169400876358238,
      "learning_rate": 6.25e-06,
      "loss": 0.4765,
      "step": 5
    },
    {
      "epoch": 0.273972602739726,
      "grad_norm": 0.13066100210309342,
      "learning_rate": 9.975923633360985e-06,
      "loss": 0.4931,
      "step": 10
    },
    {
      "epoch": 0.410958904109589,
      "grad_norm": 0.090942347593835,
      "learning_rate": 9.707720325915105e-06,
      "loss": 0.4442,
      "step": 15
    },
    {
      "epoch": 0.547945205479452,
      "grad_norm": 0.0773742501512366,
      "learning_rate": 9.157348061512728e-06,
      "loss": 0.4414,
      "step": 20
    },
    {
      "epoch": 0.684931506849315,
      "grad_norm": 0.06469879367978734,
      "learning_rate": 8.357794774235094e-06,
      "loss": 0.4305,
      "step": 25
    },
    {
      "epoch": 0.821917808219178,
      "grad_norm": 0.07169259007148933,
      "learning_rate": 7.3569836841299905e-06,
      "loss": 0.4087,
      "step": 30
    },
    {
      "epoch": 0.958904109589041,
      "grad_norm": 0.05120584651824755,
      "learning_rate": 6.21490089951632e-06,
      "loss": 0.3805,
      "step": 35
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.38145706057548523,
      "eval_runtime": 19.6656,
      "eval_samples_per_second": 19.425,
      "eval_steps_per_second": 4.882,
      "step": 37
    },
    {
      "epoch": 1.0821917808219177,
      "grad_norm": 0.05410919141307303,
      "learning_rate": 5e-06,
      "loss": 0.3772,
      "step": 40
    },
    {
      "epoch": 1.2191780821917808,
      "grad_norm": 0.04666102361204336,
      "learning_rate": 3.7850991004836813e-06,
      "loss": 0.3614,
      "step": 45
    },
    {
      "epoch": 1.356164383561644,
      "grad_norm": 0.05063848991516576,
      "learning_rate": 2.6430163158700116e-06,
      "loss": 0.3645,
      "step": 50
    },
    {
      "epoch": 1.4931506849315068,
      "grad_norm": 0.05398338068942131,
      "learning_rate": 1.642205225764908e-06,
      "loss": 0.3708,
      "step": 55
    },
    {
      "epoch": 1.6301369863013697,
      "grad_norm": 0.0560834022659863,
      "learning_rate": 8.426519384872733e-07,
      "loss": 0.3529,
      "step": 60
    },
    {
      "epoch": 1.7671232876712328,
      "grad_norm": 0.057733023886196166,
      "learning_rate": 2.9227967408489653e-07,
      "loss": 0.3529,
      "step": 65
    },
    {
      "epoch": 1.904109589041096,
      "grad_norm": 0.04675257664571633,
      "learning_rate": 2.4076366639015914e-08,
      "loss": 0.3453,
      "step": 70
    },
    {
      "epoch": 1.958904109589041,
      "eval_loss": 0.37256091833114624,
      "eval_runtime": 19.6168,
      "eval_samples_per_second": 19.473,
      "eval_steps_per_second": 4.894,
      "step": 72
    },
    {
      "epoch": 1.958904109589041,
      "step": 72,
      "total_flos": 1.8376509591283302e+17,
      "train_loss": 0.39936936149994534,
      "train_runtime": 988.0556,
      "train_samples_per_second": 3.542,
      "train_steps_per_second": 0.073
    }
  ],
  "logging_steps": 5,
  "max_steps": 72,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8376509591283302e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}