last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02372948388372553,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007909827961241843,
      "eval_loss": 4.051351070404053,
      "eval_runtime": 96.1654,
      "eval_samples_per_second": 5.543,
      "eval_steps_per_second": 2.776,
      "step": 1
    },
    {
      "epoch": 0.002372948388372553,
      "grad_norm": 1.3411372900009155,
      "learning_rate": 6e-05,
      "loss": 4.1091,
      "step": 3
    },
    {
      "epoch": 0.003954913980620921,
      "eval_loss": 3.9182193279266357,
      "eval_runtime": 96.0651,
      "eval_samples_per_second": 5.548,
      "eval_steps_per_second": 2.779,
      "step": 5
    },
    {
      "epoch": 0.004745896776745106,
      "grad_norm": 1.1928367614746094,
      "learning_rate": 0.00012,
      "loss": 3.6438,
      "step": 6
    },
    {
      "epoch": 0.007118845165117659,
      "grad_norm": 1.346867561340332,
      "learning_rate": 0.00018,
      "loss": 3.4395,
      "step": 9
    },
    {
      "epoch": 0.007909827961241843,
      "eval_loss": 3.1203250885009766,
      "eval_runtime": 96.4553,
      "eval_samples_per_second": 5.526,
      "eval_steps_per_second": 2.768,
      "step": 10
    },
    {
      "epoch": 0.009491793553490212,
      "grad_norm": 0.9624531865119934,
      "learning_rate": 0.00019510565162951537,
      "loss": 2.9168,
      "step": 12
    },
    {
      "epoch": 0.011864741941862765,
      "grad_norm": 1.5339550971984863,
      "learning_rate": 0.00017071067811865476,
      "loss": 2.8234,
      "step": 15
    },
    {
      "epoch": 0.011864741941862765,
      "eval_loss": 2.461575508117676,
      "eval_runtime": 96.3641,
      "eval_samples_per_second": 5.531,
      "eval_steps_per_second": 2.771,
      "step": 15
    },
    {
      "epoch": 0.014237690330235317,
      "grad_norm": 1.5237758159637451,
      "learning_rate": 0.00013090169943749476,
      "loss": 2.411,
      "step": 18
    },
    {
      "epoch": 0.015819655922483685,
      "eval_loss": 2.0676076412200928,
      "eval_runtime": 96.4467,
      "eval_samples_per_second": 5.526,
      "eval_steps_per_second": 2.768,
      "step": 20
    },
    {
      "epoch": 0.01661063871860787,
      "grad_norm": 1.1589887142181396,
      "learning_rate": 8.435655349597689e-05,
      "loss": 1.8789,
      "step": 21
    },
    {
      "epoch": 0.018983587106980424,
      "grad_norm": 1.2269470691680908,
      "learning_rate": 4.12214747707527e-05,
      "loss": 2.0367,
      "step": 24
    },
    {
      "epoch": 0.019774569903104607,
      "eval_loss": 1.9202778339385986,
      "eval_runtime": 96.3717,
      "eval_samples_per_second": 5.531,
      "eval_steps_per_second": 2.771,
      "step": 25
    },
    {
      "epoch": 0.021356535495352977,
      "grad_norm": 1.1548726558685303,
      "learning_rate": 1.0899347581163221e-05,
      "loss": 1.7571,
      "step": 27
    },
    {
      "epoch": 0.02372948388372553,
      "grad_norm": 1.1918234825134277,
      "learning_rate": 0.0,
      "loss": 1.8205,
      "step": 30
    },
    {
      "epoch": 0.02372948388372553,
      "eval_loss": 1.8914263248443604,
      "eval_runtime": 96.1971,
      "eval_samples_per_second": 5.541,
      "eval_steps_per_second": 2.776,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4939038012211200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}