trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.13793103448275862,
  "eval_steps": 1,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.013793103448275862,
      "grad_norm": 6.989287853240967,
      "learning_rate": 2.2727272727272728e-06,
      "loss": 1.8926,
      "step": 1
    },
    {
      "epoch": 0.013793103448275862,
      "eval_loss": 2.144650459289551,
      "eval_runtime": 17.2592,
      "eval_samples_per_second": 1.159,
      "eval_steps_per_second": 0.579,
      "step": 1
    },
    {
      "epoch": 0.027586206896551724,
      "grad_norm": 6.281332492828369,
      "learning_rate": 4.5454545454545455e-06,
      "loss": 1.986,
      "step": 2
    },
    {
      "epoch": 0.027586206896551724,
      "eval_loss": 2.0905685424804688,
      "eval_runtime": 17.6184,
      "eval_samples_per_second": 1.135,
      "eval_steps_per_second": 0.568,
      "step": 2
    },
    {
      "epoch": 0.041379310344827586,
      "grad_norm": 4.347537040710449,
      "learning_rate": 6.818181818181818e-06,
      "loss": 1.9355,
      "step": 3
    },
    {
      "epoch": 0.041379310344827586,
      "eval_loss": 1.9983774423599243,
      "eval_runtime": 17.5928,
      "eval_samples_per_second": 1.137,
      "eval_steps_per_second": 0.568,
      "step": 3
    },
    {
      "epoch": 0.05517241379310345,
      "grad_norm": 4.809764385223389,
      "learning_rate": 9.090909090909091e-06,
      "loss": 1.7509,
      "step": 4
    },
    {
      "epoch": 0.05517241379310345,
      "eval_loss": 1.8737837076187134,
      "eval_runtime": 17.6375,
      "eval_samples_per_second": 1.134,
      "eval_steps_per_second": 0.567,
      "step": 4
    },
    {
      "epoch": 0.06896551724137931,
      "grad_norm": 3.4548702239990234,
      "learning_rate": 1.1363636363636365e-05,
      "loss": 1.8838,
      "step": 5
    },
    {
      "epoch": 0.06896551724137931,
      "eval_loss": 1.7746165990829468,
      "eval_runtime": 18.0257,
      "eval_samples_per_second": 1.11,
      "eval_steps_per_second": 0.555,
      "step": 5
    },
    {
      "epoch": 0.08275862068965517,
      "grad_norm": 3.1943702697753906,
      "learning_rate": 1.3636363636363637e-05,
      "loss": 1.7707,
      "step": 6
    },
    {
      "epoch": 0.08275862068965517,
      "eval_loss": 1.6792665719985962,
      "eval_runtime": 17.7498,
      "eval_samples_per_second": 1.127,
      "eval_steps_per_second": 0.563,
      "step": 6
    },
    {
      "epoch": 0.09655172413793103,
      "grad_norm": 3.318288564682007,
      "learning_rate": 1.590909090909091e-05,
      "loss": 1.7171,
      "step": 7
    },
    {
      "epoch": 0.09655172413793103,
      "eval_loss": 1.5874873399734497,
      "eval_runtime": 17.6295,
      "eval_samples_per_second": 1.134,
      "eval_steps_per_second": 0.567,
      "step": 7
    },
    {
      "epoch": 0.1103448275862069,
      "grad_norm": 3.210330009460449,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 1.5734,
      "step": 8
    },
    {
      "epoch": 0.1103448275862069,
      "eval_loss": 1.535287618637085,
      "eval_runtime": 17.6232,
      "eval_samples_per_second": 1.135,
      "eval_steps_per_second": 0.567,
      "step": 8
    },
    {
      "epoch": 0.12413793103448276,
      "grad_norm": 3.2319107055664062,
      "learning_rate": 2.0454545454545457e-05,
      "loss": 1.7986,
      "step": 9
    },
    {
      "epoch": 0.12413793103448276,
      "eval_loss": 1.467301607131958,
      "eval_runtime": 17.5824,
      "eval_samples_per_second": 1.138,
      "eval_steps_per_second": 0.569,
      "step": 9
    }
  ],
  "logging_steps": 1,
  "max_steps": 216,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.169340664455168e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}