last-checkpoint/trainer_state.json

{
  "best_metric": 1.5300686359405518,
  "best_model_checkpoint": "./outputs/checkpoint-400",
  "epoch": 0.29143897996357016,
  "eval_steps": 100,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 1.765,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 1.631608486175537,
      "eval_runtime": 430.1354,
      "eval_samples_per_second": 14.586,
      "eval_steps_per_second": 1.825,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 1.6077,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.5870035886764526,
      "eval_runtime": 417.8578,
      "eval_samples_per_second": 15.015,
      "eval_steps_per_second": 1.879,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 1.5755,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.5563029050827026,
      "eval_runtime": 417.8787,
      "eval_samples_per_second": 15.014,
      "eval_steps_per_second": 1.879,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 1.5445,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.5300686359405518,
      "eval_runtime": 417.9846,
      "eval_samples_per_second": 15.01,
      "eval_steps_per_second": 1.878,
      "step": 400
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.298292292791122e+17,
  "trial_name": null,
  "trial_params": null
}