last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00011021894994406389,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.1021894994406387e-05,
      "grad_norm": 0.9139910340309143,
      "learning_rate": 2e-05,
      "loss": 3.0642,
      "step": 1
    },
    {
      "epoch": 1.1021894994406387e-05,
      "eval_loss": 3.108307361602783,
      "eval_runtime": 1107.2078,
      "eval_samples_per_second": 34.503,
      "eval_steps_per_second": 17.252,
      "step": 1
    },
    {
      "epoch": 2.2043789988812775e-05,
      "grad_norm": 0.9620638489723206,
      "learning_rate": 4e-05,
      "loss": 3.1492,
      "step": 2
    },
    {
      "epoch": 3.306568498321916e-05,
      "grad_norm": 0.792957067489624,
      "learning_rate": 6e-05,
      "loss": 3.2571,
      "step": 3
    },
    {
      "epoch": 3.306568498321916e-05,
      "eval_loss": 3.1041641235351562,
      "eval_runtime": 1107.7452,
      "eval_samples_per_second": 34.486,
      "eval_steps_per_second": 17.243,
      "step": 3
    },
    {
      "epoch": 4.408757997762555e-05,
      "grad_norm": 0.7465797662734985,
      "learning_rate": 8e-05,
      "loss": 2.9506,
      "step": 4
    },
    {
      "epoch": 5.5109474972031944e-05,
      "grad_norm": 0.6941638588905334,
      "learning_rate": 0.0001,
      "loss": 3.0625,
      "step": 5
    },
    {
      "epoch": 6.613136996643832e-05,
      "grad_norm": 0.7453097701072693,
      "learning_rate": 0.00012,
      "loss": 3.0686,
      "step": 6
    },
    {
      "epoch": 6.613136996643832e-05,
      "eval_loss": 3.0480544567108154,
      "eval_runtime": 1108.0987,
      "eval_samples_per_second": 34.475,
      "eval_steps_per_second": 17.238,
      "step": 6
    },
    {
      "epoch": 7.715326496084471e-05,
      "grad_norm": 0.7151815891265869,
      "learning_rate": 0.00014,
      "loss": 3.1302,
      "step": 7
    },
    {
      "epoch": 8.81751599552511e-05,
      "grad_norm": 0.735742449760437,
      "learning_rate": 0.00016,
      "loss": 2.882,
      "step": 8
    },
    {
      "epoch": 9.91970549496575e-05,
      "grad_norm": 0.6919463872909546,
      "learning_rate": 0.00018,
      "loss": 2.7777,
      "step": 9
    },
    {
      "epoch": 9.91970549496575e-05,
      "eval_loss": 2.9666075706481934,
      "eval_runtime": 1106.0972,
      "eval_samples_per_second": 34.538,
      "eval_steps_per_second": 17.269,
      "step": 9
    },
    {
      "epoch": 0.00011021894994406389,
      "grad_norm": 0.7852161526679993,
      "learning_rate": 0.0002,
      "loss": 2.9233,
      "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": 277675172167680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}