last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0007340976105122777,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.446992035040926e-05,
      "eval_loss": NaN,
      "eval_runtime": 2728.9412,
      "eval_samples_per_second": 6.305,
      "eval_steps_per_second": 3.153,
      "step": 1
    },
    {
      "epoch": 7.340976105122778e-05,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0001223496017520463,
      "eval_loss": NaN,
      "eval_runtime": 2728.657,
      "eval_samples_per_second": 6.306,
      "eval_steps_per_second": 3.153,
      "step": 5
    },
    {
      "epoch": 0.00014681952210245555,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.00022022928315368334,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0002446992035040926,
      "eval_loss": NaN,
      "eval_runtime": 2727.8611,
      "eval_samples_per_second": 6.308,
      "eval_steps_per_second": 3.154,
      "step": 10
    },
    {
      "epoch": 0.0002936390442049111,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.00036704880525613887,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.00036704880525613887,
      "eval_loss": NaN,
      "eval_runtime": 2727.2822,
      "eval_samples_per_second": 6.309,
      "eval_steps_per_second": 3.155,
      "step": 15
    },
    {
      "epoch": 0.0004404585663073667,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.0004893984070081852,
      "eval_loss": NaN,
      "eval_runtime": 2728.6679,
      "eval_samples_per_second": 6.306,
      "eval_steps_per_second": 3.153,
      "step": 20
    },
    {
      "epoch": 0.0005138683273585944,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0005872780884098222,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.0006117480087602315,
      "eval_loss": NaN,
      "eval_runtime": 2729.6987,
      "eval_samples_per_second": 6.304,
      "eval_steps_per_second": 3.152,
      "step": 25
    },
    {
      "epoch": 0.00066068784946105,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.0007340976105122777,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0007340976105122777,
      "eval_loss": NaN,
      "eval_runtime": 2730.3393,
      "eval_samples_per_second": 6.302,
      "eval_steps_per_second": 3.151,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.003811019259904e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}