last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0008564087924636027,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.515653249595585e-05,
      "grad_norm": 0.8959489464759827,
      "learning_rate": 2e-05,
      "loss": 0.8972,
      "step": 1
    },
    {
      "epoch": 9.515653249595585e-05,
      "eval_loss": 0.9797402024269104,
      "eval_runtime": 702.7202,
      "eval_samples_per_second": 3.149,
      "eval_steps_per_second": 3.149,
      "step": 1
    },
    {
      "epoch": 0.0001903130649919117,
      "grad_norm": 0.5574295520782471,
      "learning_rate": 4e-05,
      "loss": 0.8309,
      "step": 2
    },
    {
      "epoch": 0.00028546959748786756,
      "grad_norm": 0.5667484402656555,
      "learning_rate": 6e-05,
      "loss": 0.7712,
      "step": 3
    },
    {
      "epoch": 0.00028546959748786756,
      "eval_loss": 0.9666468501091003,
      "eval_runtime": 703.082,
      "eval_samples_per_second": 3.148,
      "eval_steps_per_second": 3.148,
      "step": 3
    },
    {
      "epoch": 0.0003806261299838234,
      "grad_norm": 0.8053474426269531,
      "learning_rate": 8e-05,
      "loss": 0.7217,
      "step": 4
    },
    {
      "epoch": 0.0004757826624797792,
      "grad_norm": 0.580493152141571,
      "learning_rate": 0.0001,
      "loss": 1.161,
      "step": 5
    },
    {
      "epoch": 0.0005709391949757351,
      "grad_norm": 0.7792345285415649,
      "learning_rate": 0.00012,
      "loss": 0.9679,
      "step": 6
    },
    {
      "epoch": 0.0005709391949757351,
      "eval_loss": 0.8787497878074646,
      "eval_runtime": 702.975,
      "eval_samples_per_second": 3.148,
      "eval_steps_per_second": 3.148,
      "step": 6
    },
    {
      "epoch": 0.000666095727471691,
      "grad_norm": 0.7983985543251038,
      "learning_rate": 0.00014,
      "loss": 0.9646,
      "step": 7
    },
    {
      "epoch": 0.0007612522599676468,
      "grad_norm": 0.7791344523429871,
      "learning_rate": 0.00016,
      "loss": 1.0793,
      "step": 8
    },
    {
      "epoch": 0.0008564087924636027,
      "grad_norm": 0.9845541715621948,
      "learning_rate": 0.00018,
      "loss": 0.875,
      "step": 9
    },
    {
      "epoch": 0.0008564087924636027,
      "eval_loss": 0.7046293020248413,
      "eval_runtime": 703.0342,
      "eval_samples_per_second": 3.148,
      "eval_steps_per_second": 3.148,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.145153356038144e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}