last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5405405405405406,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05405405405405406,
      "grad_norm": 0.4434233009815216,
      "learning_rate": 2e-05,
      "loss": 11.8518,
      "step": 1
    },
    {
      "epoch": 0.05405405405405406,
      "eval_loss": 11.87682819366455,
      "eval_runtime": 0.6037,
      "eval_samples_per_second": 13.251,
      "eval_steps_per_second": 6.626,
      "step": 1
    },
    {
      "epoch": 0.10810810810810811,
      "grad_norm": 0.4468322694301605,
      "learning_rate": 4e-05,
      "loss": 11.8776,
      "step": 2
    },
    {
      "epoch": 0.16216216216216217,
      "grad_norm": 0.5244847536087036,
      "learning_rate": 6e-05,
      "loss": 11.9195,
      "step": 3
    },
    {
      "epoch": 0.16216216216216217,
      "eval_loss": 11.876558303833008,
      "eval_runtime": 0.1169,
      "eval_samples_per_second": 68.45,
      "eval_steps_per_second": 34.225,
      "step": 3
    },
    {
      "epoch": 0.21621621621621623,
      "grad_norm": 0.43882226943969727,
      "learning_rate": 8e-05,
      "loss": 11.9086,
      "step": 4
    },
    {
      "epoch": 0.2702702702702703,
      "grad_norm": 0.40448427200317383,
      "learning_rate": 0.0001,
      "loss": 11.9007,
      "step": 5
    },
    {
      "epoch": 0.32432432432432434,
      "grad_norm": 0.4453897774219513,
      "learning_rate": 0.00012,
      "loss": 11.8834,
      "step": 6
    },
    {
      "epoch": 0.32432432432432434,
      "eval_loss": 11.873939514160156,
      "eval_runtime": 0.1194,
      "eval_samples_per_second": 67.028,
      "eval_steps_per_second": 33.514,
      "step": 6
    },
    {
      "epoch": 0.3783783783783784,
      "grad_norm": 0.3570092022418976,
      "learning_rate": 0.00014,
      "loss": 11.8717,
      "step": 7
    },
    {
      "epoch": 0.43243243243243246,
      "grad_norm": 0.5594825744628906,
      "learning_rate": 0.00016,
      "loss": 11.9122,
      "step": 8
    },
    {
      "epoch": 0.4864864864864865,
      "grad_norm": 0.45945826172828674,
      "learning_rate": 0.00018,
      "loss": 11.8951,
      "step": 9
    },
    {
      "epoch": 0.4864864864864865,
      "eval_loss": 11.869675636291504,
      "eval_runtime": 0.1191,
      "eval_samples_per_second": 67.148,
      "eval_steps_per_second": 33.574,
      "step": 9
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 0.452753484249115,
      "learning_rate": 0.0002,
      "loss": 11.9054,
      "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": 1345426292736.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}