last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0538922155688624,
  "eval_steps": 3,
  "global_step": 11,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09580838323353294,
      "grad_norm": 18.69622802734375,
      "learning_rate": 1e-05,
      "loss": 8.3105,
      "step": 1
    },
    {
      "epoch": 0.09580838323353294,
      "eval_loss": 0.5051048994064331,
      "eval_runtime": 12.4846,
      "eval_samples_per_second": 1.442,
      "eval_steps_per_second": 0.721,
      "step": 1
    },
    {
      "epoch": 0.19161676646706588,
      "grad_norm": 19.64305877685547,
      "learning_rate": 2e-05,
      "loss": 8.379,
      "step": 2
    },
    {
      "epoch": 0.2874251497005988,
      "grad_norm": 18.88804817199707,
      "learning_rate": 3e-05,
      "loss": 8.352,
      "step": 3
    },
    {
      "epoch": 0.2874251497005988,
      "eval_loss": 0.4333694279193878,
      "eval_runtime": 0.8932,
      "eval_samples_per_second": 20.153,
      "eval_steps_per_second": 10.076,
      "step": 3
    },
    {
      "epoch": 0.38323353293413176,
      "grad_norm": 18.877365112304688,
      "learning_rate": 4e-05,
      "loss": 8.279,
      "step": 4
    },
    {
      "epoch": 0.47904191616766467,
      "grad_norm": 18.81561851501465,
      "learning_rate": 5e-05,
      "loss": 7.9473,
      "step": 5
    },
    {
      "epoch": 0.5748502994011976,
      "grad_norm": 19.72781753540039,
      "learning_rate": 6e-05,
      "loss": 7.7265,
      "step": 6
    },
    {
      "epoch": 0.5748502994011976,
      "eval_loss": 0.3338497579097748,
      "eval_runtime": 0.8954,
      "eval_samples_per_second": 20.102,
      "eval_steps_per_second": 10.051,
      "step": 6
    },
    {
      "epoch": 0.6706586826347305,
      "grad_norm": 19.34385871887207,
      "learning_rate": 7e-05,
      "loss": 7.1555,
      "step": 7
    },
    {
      "epoch": 0.7664670658682635,
      "grad_norm": 19.580270767211914,
      "learning_rate": 8e-05,
      "loss": 6.3322,
      "step": 8
    },
    {
      "epoch": 0.8622754491017964,
      "grad_norm": 19.722665786743164,
      "learning_rate": 9e-05,
      "loss": 5.4843,
      "step": 9
    },
    {
      "epoch": 0.8622754491017964,
      "eval_loss": 0.16169598698616028,
      "eval_runtime": 0.9032,
      "eval_samples_per_second": 19.929,
      "eval_steps_per_second": 9.964,
      "step": 9
    },
    {
      "epoch": 0.9580838323353293,
      "grad_norm": 16.325122833251953,
      "learning_rate": 0.0001,
      "loss": 3.4328,
      "step": 10
    },
    {
      "epoch": 1.0538922155688624,
      "grad_norm": 14.050758361816406,
      "learning_rate": 0.0,
      "loss": 2.343,
      "step": 11
    }
  ],
  "logging_steps": 1,
  "max_steps": 11,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 4027520444792832.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}