trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9808429118773945,
  "eval_steps": 500,
  "global_step": 144,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.20434227330779056,
      "grad_norm": 0.719012975692749,
      "learning_rate": 1.2e-05,
      "loss": 0.5578,
      "step": 10
    },
    {
      "epoch": 0.4086845466155811,
      "grad_norm": 0.9851335287094116,
      "learning_rate": 1.995259033893236e-05,
      "loss": 0.4473,
      "step": 20
    },
    {
      "epoch": 0.6130268199233716,
      "grad_norm": 0.6076326370239258,
      "learning_rate": 1.9424380828337146e-05,
      "loss": 0.3542,
      "step": 30
    },
    {
      "epoch": 0.8173690932311622,
      "grad_norm": 0.5375417470932007,
      "learning_rate": 1.833997817889878e-05,
      "loss": 0.2408,
      "step": 40
    },
    {
      "epoch": 1.040868454661558,
      "grad_norm": 0.47686120867729187,
      "learning_rate": 1.676337994380903e-05,
      "loss": 0.1631,
      "step": 50
    },
    {
      "epoch": 1.2452107279693487,
      "grad_norm": 0.6329728960990906,
      "learning_rate": 1.4787631293572094e-05,
      "loss": 0.0694,
      "step": 60
    },
    {
      "epoch": 1.4495530012771392,
      "grad_norm": 0.3885073959827423,
      "learning_rate": 1.2529333823916807e-05,
      "loss": 0.0432,
      "step": 70
    },
    {
      "epoch": 1.6538952745849298,
      "grad_norm": 0.2986627519130707,
      "learning_rate": 1.0121764148019977e-05,
      "loss": 0.0326,
      "step": 80
    },
    {
      "epoch": 1.8582375478927204,
      "grad_norm": 0.207879900932312,
      "learning_rate": 7.707008389035102e-06,
      "loss": 0.0284,
      "step": 90
    },
    {
      "epoch": 2.081736909323116,
      "grad_norm": 0.14822977781295776,
      "learning_rate": 5.427576766953615e-06,
      "loss": 0.0157,
      "step": 100
    },
    {
      "epoch": 2.286079182630907,
      "grad_norm": 0.19800704717636108,
      "learning_rate": 3.4179931567925216e-06,
      "loss": 0.0053,
      "step": 110
    },
    {
      "epoch": 2.4904214559386975,
      "grad_norm": 0.293628454208374,
      "learning_rate": 1.7968559722048906e-06,
      "loss": 0.0083,
      "step": 120
    },
    {
      "epoch": 2.694763729246488,
      "grad_norm": 0.1427803635597229,
      "learning_rate": 6.598389126745209e-07,
      "loss": 0.0054,
      "step": 130
    },
    {
      "epoch": 2.8991060025542783,
      "grad_norm": 0.1450967788696289,
      "learning_rate": 7.404464507973608e-08,
      "loss": 0.005,
      "step": 140
    },
    {
      "epoch": 2.9808429118773945,
      "step": 144,
      "total_flos": 3.756952806007767e+17,
      "train_loss": 0.13742555860274783,
      "train_runtime": 2723.3625,
      "train_samples_per_second": 3.45,
      "train_steps_per_second": 0.053
    }
  ],
  "logging_steps": 10,
  "max_steps": 144,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.756952806007767e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}