trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.99660441426146,
  "eval_steps": 50,
  "global_step": 294,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 5e-05,
      "loss": 1.1147,
      "step": 10
    },
    {
      "epoch": 0.14,
      "learning_rate": 5e-05,
      "loss": 1.0413,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 5e-05,
      "loss": 1.0627,
      "step": 30
    },
    {
      "epoch": 0.27,
      "learning_rate": 5e-05,
      "loss": 1.0158,
      "step": 40
    },
    {
      "epoch": 0.34,
      "learning_rate": 5e-05,
      "loss": 0.9925,
      "step": 50
    },
    {
      "epoch": 0.34,
      "eval_loss": 1.054595708847046,
      "eval_runtime": 84.2428,
      "eval_samples_per_second": 3.11,
      "eval_steps_per_second": 0.783,
      "step": 50
    },
    {
      "epoch": 0.41,
      "learning_rate": 5e-05,
      "loss": 1.0051,
      "step": 60
    },
    {
      "epoch": 0.48,
      "learning_rate": 5e-05,
      "loss": 1.0116,
      "step": 70
    },
    {
      "epoch": 0.54,
      "learning_rate": 5e-05,
      "loss": 1.0378,
      "step": 80
    },
    {
      "epoch": 0.61,
      "learning_rate": 5e-05,
      "loss": 1.0309,
      "step": 90
    },
    {
      "epoch": 0.68,
      "learning_rate": 5e-05,
      "loss": 0.9873,
      "step": 100
    },
    {
      "epoch": 0.68,
      "eval_loss": 1.0322012901306152,
      "eval_runtime": 84.2753,
      "eval_samples_per_second": 3.109,
      "eval_steps_per_second": 0.783,
      "step": 100
    },
    {
      "epoch": 0.75,
      "learning_rate": 5e-05,
      "loss": 0.9827,
      "step": 110
    },
    {
      "epoch": 0.81,
      "learning_rate": 5e-05,
      "loss": 0.9831,
      "step": 120
    },
    {
      "epoch": 0.88,
      "learning_rate": 5e-05,
      "loss": 1.009,
      "step": 130
    },
    {
      "epoch": 0.95,
      "learning_rate": 5e-05,
      "loss": 1.0236,
      "step": 140
    },
    {
      "epoch": 1.02,
      "learning_rate": 5e-05,
      "loss": 0.9577,
      "step": 150
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.0260435342788696,
      "eval_runtime": 84.2852,
      "eval_samples_per_second": 3.108,
      "eval_steps_per_second": 0.783,
      "step": 150
    },
    {
      "epoch": 1.09,
      "learning_rate": 5e-05,
      "loss": 0.8567,
      "step": 160
    },
    {
      "epoch": 1.15,
      "learning_rate": 5e-05,
      "loss": 0.9293,
      "step": 170
    },
    {
      "epoch": 1.22,
      "learning_rate": 5e-05,
      "loss": 0.8656,
      "step": 180
    },
    {
      "epoch": 1.29,
      "learning_rate": 5e-05,
      "loss": 0.7937,
      "step": 190
    },
    {
      "epoch": 1.36,
      "learning_rate": 5e-05,
      "loss": 0.8504,
      "step": 200
    },
    {
      "epoch": 1.36,
      "eval_loss": 1.0647752285003662,
      "eval_runtime": 84.2818,
      "eval_samples_per_second": 3.109,
      "eval_steps_per_second": 0.783,
      "step": 200
    },
    {
      "epoch": 1.43,
      "learning_rate": 5e-05,
      "loss": 0.887,
      "step": 210
    },
    {
      "epoch": 1.49,
      "learning_rate": 5e-05,
      "loss": 0.7825,
      "step": 220
    },
    {
      "epoch": 1.56,
      "learning_rate": 5e-05,
      "loss": 0.8271,
      "step": 230
    },
    {
      "epoch": 1.63,
      "learning_rate": 5e-05,
      "loss": 0.8465,
      "step": 240
    },
    {
      "epoch": 1.7,
      "learning_rate": 5e-05,
      "loss": 0.9154,
      "step": 250
    },
    {
      "epoch": 1.7,
      "eval_loss": 1.0422441959381104,
      "eval_runtime": 84.2906,
      "eval_samples_per_second": 3.108,
      "eval_steps_per_second": 0.783,
      "step": 250
    },
    {
      "epoch": 1.77,
      "learning_rate": 5e-05,
      "loss": 0.7803,
      "step": 260
    },
    {
      "epoch": 1.83,
      "learning_rate": 5e-05,
      "loss": 0.8642,
      "step": 270
    },
    {
      "epoch": 1.9,
      "learning_rate": 5e-05,
      "loss": 0.8112,
      "step": 280
    },
    {
      "epoch": 1.97,
      "learning_rate": 5e-05,
      "loss": 0.8171,
      "step": 290
    },
    {
      "epoch": 2.0,
      "step": 294,
      "total_flos": 4.649929554906317e+17,
      "train_loss": 0.9311216080269846,
      "train_runtime": 4522.2924,
      "train_samples_per_second": 1.042,
      "train_steps_per_second": 0.065
    }
  ],
  "logging_steps": 10,
  "max_steps": 294,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 250,
  "total_flos": 4.649929554906317e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}