last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.011463250168577209,
  "eval_steps": 34,
  "global_step": 34,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00033715441672285906,
      "eval_loss": 2.3532843589782715,
      "eval_runtime": 338.8675,
      "eval_samples_per_second": 14.74,
      "eval_steps_per_second": 1.844,
      "step": 1
    },
    {
      "epoch": 0.0010114632501685772,
      "grad_norm": 0.8756303787231445,
      "learning_rate": 1.5e-05,
      "loss": 2.361,
      "step": 3
    },
    {
      "epoch": 0.0020229265003371545,
      "grad_norm": 0.9312941431999207,
      "learning_rate": 3e-05,
      "loss": 2.4681,
      "step": 6
    },
    {
      "epoch": 0.0030343897505057315,
      "grad_norm": 0.9662004709243774,
      "learning_rate": 4.5e-05,
      "loss": 2.1782,
      "step": 9
    },
    {
      "epoch": 0.004045853000674309,
      "grad_norm": 1.3127541542053223,
      "learning_rate": 4.999675562428437e-05,
      "loss": 2.2872,
      "step": 12
    },
    {
      "epoch": 0.0050573162508428865,
      "grad_norm": 1.5349361896514893,
      "learning_rate": 4.9979724954289244e-05,
      "loss": 1.8232,
      "step": 15
    },
    {
      "epoch": 0.006068779501011463,
      "grad_norm": 1.5614502429962158,
      "learning_rate": 4.994810682835951e-05,
      "loss": 1.5184,
      "step": 18
    },
    {
      "epoch": 0.0070802427511800405,
      "grad_norm": 1.413482904434204,
      "learning_rate": 4.990191971059033e-05,
      "loss": 1.2261,
      "step": 21
    },
    {
      "epoch": 0.008091706001348618,
      "grad_norm": 1.2501227855682373,
      "learning_rate": 4.984119057295783e-05,
      "loss": 1.1656,
      "step": 24
    },
    {
      "epoch": 0.009103169251517195,
      "grad_norm": 1.0704395771026611,
      "learning_rate": 4.976595487956823e-05,
      "loss": 1.0296,
      "step": 27
    },
    {
      "epoch": 0.010114632501685773,
      "grad_norm": 0.8690694570541382,
      "learning_rate": 4.967625656594782e-05,
      "loss": 0.834,
      "step": 30
    },
    {
      "epoch": 0.011126095751854349,
      "grad_norm": 0.9992819428443909,
      "learning_rate": 4.957214801338581e-05,
      "loss": 0.9255,
      "step": 33
    },
    {
      "epoch": 0.011463250168577209,
      "eval_loss": 0.8507078289985657,
      "eval_runtime": 341.3748,
      "eval_samples_per_second": 14.632,
      "eval_steps_per_second": 1.831,
      "step": 34
    }
  ],
  "logging_steps": 3,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 34,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.753463114622566e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}