trainer_state.json

{
  "best_metric": 2.203190326690674,
  "best_model_checkpoint": "out/checkpoint-9",
  "epoch": 0.036036036036036036,
  "eval_steps": 1,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 3.431358814239502,
      "learning_rate": 0.0002,
      "loss": 3.5363,
      "step": 1
    },
    {
      "epoch": 0.0,
      "eval_loss": 2.7476656436920166,
      "eval_runtime": 746.9283,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 1
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.838670253753662,
      "learning_rate": 0.0002,
      "loss": 3.0594,
      "step": 2
    },
    {
      "epoch": 0.01,
      "eval_loss": 2.539435625076294,
      "eval_runtime": 747.0674,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 2
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3010876178741455,
      "learning_rate": 0.0002,
      "loss": 2.4007,
      "step": 3
    },
    {
      "epoch": 0.01,
      "eval_loss": 2.4365687370300293,
      "eval_runtime": 747.0903,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 3
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.1130387783050537,
      "learning_rate": 0.0002,
      "loss": 2.0025,
      "step": 4
    },
    {
      "epoch": 0.02,
      "eval_loss": 2.3684797286987305,
      "eval_runtime": 747.0694,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 4
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.9912111759185791,
      "learning_rate": 0.0002,
      "loss": 1.9871,
      "step": 5
    },
    {
      "epoch": 0.02,
      "eval_loss": 2.324511766433716,
      "eval_runtime": 747.1038,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 5
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.8513877987861633,
      "learning_rate": 0.0002,
      "loss": 2.9169,
      "step": 6
    },
    {
      "epoch": 0.02,
      "eval_loss": 2.282069683074951,
      "eval_runtime": 747.081,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 6
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8679900169372559,
      "learning_rate": 0.0002,
      "loss": 2.6585,
      "step": 7
    },
    {
      "epoch": 0.03,
      "eval_loss": 2.250554084777832,
      "eval_runtime": 747.1089,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 7
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.9949858784675598,
      "learning_rate": 0.0002,
      "loss": 1.9087,
      "step": 8
    },
    {
      "epoch": 0.03,
      "eval_loss": 2.2261030673980713,
      "eval_runtime": 747.1156,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 8
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8005623817443848,
      "learning_rate": 0.0002,
      "loss": 1.8623,
      "step": 9
    },
    {
      "epoch": 0.04,
      "eval_loss": 2.203190326690674,
      "eval_runtime": 747.1469,
      "eval_samples_per_second": 0.365,
      "eval_steps_per_second": 0.365,
      "step": 9
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1,
  "total_flos": 1577402807353344.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}