last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03710575139146568,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0037105751391465678,
      "grad_norm": 16.39884376525879,
      "learning_rate": 2e-05,
      "loss": 9.1969,
      "step": 1
    },
    {
      "epoch": 0.0037105751391465678,
      "eval_loss": 9.007013320922852,
      "eval_runtime": 26.0552,
      "eval_samples_per_second": 4.375,
      "eval_steps_per_second": 2.188,
      "step": 1
    },
    {
      "epoch": 0.0074211502782931356,
      "grad_norm": 18.28474235534668,
      "learning_rate": 4e-05,
      "loss": 8.8405,
      "step": 2
    },
    {
      "epoch": 0.011131725417439703,
      "grad_norm": 16.06081199645996,
      "learning_rate": 6e-05,
      "loss": 8.7897,
      "step": 3
    },
    {
      "epoch": 0.011131725417439703,
      "eval_loss": 8.362504005432129,
      "eval_runtime": 26.6541,
      "eval_samples_per_second": 4.277,
      "eval_steps_per_second": 2.139,
      "step": 3
    },
    {
      "epoch": 0.014842300556586271,
      "grad_norm": 19.3427791595459,
      "learning_rate": 8e-05,
      "loss": 8.3517,
      "step": 4
    },
    {
      "epoch": 0.01855287569573284,
      "grad_norm": 20.023731231689453,
      "learning_rate": 0.0001,
      "loss": 6.9941,
      "step": 5
    },
    {
      "epoch": 0.022263450834879406,
      "grad_norm": 19.472457885742188,
      "learning_rate": 0.00012,
      "loss": 5.532,
      "step": 6
    },
    {
      "epoch": 0.022263450834879406,
      "eval_loss": 3.4154958724975586,
      "eval_runtime": 26.993,
      "eval_samples_per_second": 4.223,
      "eval_steps_per_second": 2.112,
      "step": 6
    },
    {
      "epoch": 0.025974025974025976,
      "grad_norm": 15.347859382629395,
      "learning_rate": 0.00014,
      "loss": 3.3472,
      "step": 7
    },
    {
      "epoch": 0.029684601113172542,
      "grad_norm": 12.538643836975098,
      "learning_rate": 0.00016,
      "loss": 2.6497,
      "step": 8
    },
    {
      "epoch": 0.03339517625231911,
      "grad_norm": 10.789239883422852,
      "learning_rate": 0.00018,
      "loss": 1.4241,
      "step": 9
    },
    {
      "epoch": 0.03339517625231911,
      "eval_loss": 0.3812977075576782,
      "eval_runtime": 27.2928,
      "eval_samples_per_second": 4.177,
      "eval_steps_per_second": 2.088,
      "step": 9
    },
    {
      "epoch": 0.03710575139146568,
      "grad_norm": 6.846190929412842,
      "learning_rate": 0.0002,
      "loss": 0.1979,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.483774567120896e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}