last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007136485280999108,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007136485280999108,
      "grad_norm": 0.2924627363681793,
      "learning_rate": 2e-05,
      "loss": 1.4761,
      "step": 1
    },
    {
      "epoch": 0.0007136485280999108,
      "eval_loss": 0.3800647556781769,
      "eval_runtime": 67.4783,
      "eval_samples_per_second": 8.744,
      "eval_steps_per_second": 4.372,
      "step": 1
    },
    {
      "epoch": 0.0014272970561998216,
      "grad_norm": 0.2381795048713684,
      "learning_rate": 4e-05,
      "loss": 1.2496,
      "step": 2
    },
    {
      "epoch": 0.0021409455842997323,
      "grad_norm": 0.23130403459072113,
      "learning_rate": 6e-05,
      "loss": 1.2577,
      "step": 3
    },
    {
      "epoch": 0.0021409455842997323,
      "eval_loss": 0.3796656131744385,
      "eval_runtime": 67.5502,
      "eval_samples_per_second": 8.734,
      "eval_steps_per_second": 4.367,
      "step": 3
    },
    {
      "epoch": 0.002854594112399643,
      "grad_norm": 0.31006214022636414,
      "learning_rate": 8e-05,
      "loss": 1.2692,
      "step": 4
    },
    {
      "epoch": 0.003568242640499554,
      "grad_norm": 0.3327285051345825,
      "learning_rate": 0.0001,
      "loss": 1.5088,
      "step": 5
    },
    {
      "epoch": 0.004281891168599465,
      "grad_norm": 0.33807697892189026,
      "learning_rate": 0.00012,
      "loss": 1.5601,
      "step": 6
    },
    {
      "epoch": 0.004281891168599465,
      "eval_loss": 0.3732140362262726,
      "eval_runtime": 67.7591,
      "eval_samples_per_second": 8.707,
      "eval_steps_per_second": 4.354,
      "step": 6
    },
    {
      "epoch": 0.0049955396966993755,
      "grad_norm": 0.31830304861068726,
      "learning_rate": 0.00014,
      "loss": 1.2547,
      "step": 7
    },
    {
      "epoch": 0.005709188224799286,
      "grad_norm": 0.3602185547351837,
      "learning_rate": 0.00016,
      "loss": 1.1553,
      "step": 8
    },
    {
      "epoch": 0.006422836752899197,
      "grad_norm": 0.3723427951335907,
      "learning_rate": 0.00018,
      "loss": 1.3473,
      "step": 9
    },
    {
      "epoch": 0.006422836752899197,
      "eval_loss": 0.34222760796546936,
      "eval_runtime": 67.6386,
      "eval_samples_per_second": 8.723,
      "eval_steps_per_second": 4.361,
      "step": 9
    },
    {
      "epoch": 0.007136485280999108,
      "grad_norm": 0.3750506639480591,
      "learning_rate": 0.0002,
      "loss": 1.627,
      "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": 6794734731264000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}