last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02289639381797367,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0022896393817973667,
      "grad_norm": 4.149505615234375,
      "learning_rate": 2e-05,
      "loss": 4.2034,
      "step": 1
    },
    {
      "epoch": 0.0022896393817973667,
      "eval_loss": 4.1785359382629395,
      "eval_runtime": 29.4614,
      "eval_samples_per_second": 6.245,
      "eval_steps_per_second": 3.123,
      "step": 1
    },
    {
      "epoch": 0.0045792787635947334,
      "grad_norm": 4.028379917144775,
      "learning_rate": 4e-05,
      "loss": 4.7035,
      "step": 2
    },
    {
      "epoch": 0.006868918145392101,
      "grad_norm": 3.707000732421875,
      "learning_rate": 6e-05,
      "loss": 4.149,
      "step": 3
    },
    {
      "epoch": 0.006868918145392101,
      "eval_loss": 4.107508182525635,
      "eval_runtime": 29.7334,
      "eval_samples_per_second": 6.188,
      "eval_steps_per_second": 3.094,
      "step": 3
    },
    {
      "epoch": 0.009158557527189467,
      "grad_norm": 3.4329545497894287,
      "learning_rate": 8e-05,
      "loss": 3.7555,
      "step": 4
    },
    {
      "epoch": 0.011448196908986834,
      "grad_norm": 3.9176552295684814,
      "learning_rate": 0.0001,
      "loss": 4.0517,
      "step": 5
    },
    {
      "epoch": 0.013737836290784202,
      "grad_norm": 3.739713430404663,
      "learning_rate": 0.00012,
      "loss": 3.7634,
      "step": 6
    },
    {
      "epoch": 0.013737836290784202,
      "eval_loss": 3.4562952518463135,
      "eval_runtime": 29.7619,
      "eval_samples_per_second": 6.182,
      "eval_steps_per_second": 3.091,
      "step": 6
    },
    {
      "epoch": 0.01602747567258157,
      "grad_norm": 3.4949848651885986,
      "learning_rate": 0.00014,
      "loss": 3.4091,
      "step": 7
    },
    {
      "epoch": 0.018317115054378934,
      "grad_norm": 5.311485290527344,
      "learning_rate": 0.00016,
      "loss": 3.3507,
      "step": 8
    },
    {
      "epoch": 0.0206067544361763,
      "grad_norm": 9.926668167114258,
      "learning_rate": 0.00018,
      "loss": 3.4155,
      "step": 9
    },
    {
      "epoch": 0.0206067544361763,
      "eval_loss": 3.2308547496795654,
      "eval_runtime": 29.8059,
      "eval_samples_per_second": 6.173,
      "eval_steps_per_second": 3.087,
      "step": 9
    },
    {
      "epoch": 0.02289639381797367,
      "grad_norm": 10.94023609161377,
      "learning_rate": 0.0002,
      "loss": 3.003,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4491064828231680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}