last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005198180636777128,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005198180636777127,
      "grad_norm": 21.3779239654541,
      "learning_rate": 2e-05,
      "loss": 6.7677,
      "step": 1
    },
    {
      "epoch": 0.0005198180636777127,
      "eval_loss": 1.7232974767684937,
      "eval_runtime": 10.9514,
      "eval_samples_per_second": 74.055,
      "eval_steps_per_second": 37.073,
      "step": 1
    },
    {
      "epoch": 0.0010396361273554255,
      "grad_norm": 19.957937240600586,
      "learning_rate": 4e-05,
      "loss": 6.6812,
      "step": 2
    },
    {
      "epoch": 0.0015594541910331384,
      "grad_norm": 19.227352142333984,
      "learning_rate": 6e-05,
      "loss": 6.2624,
      "step": 3
    },
    {
      "epoch": 0.0015594541910331384,
      "eval_loss": 1.7243677377700806,
      "eval_runtime": 9.923,
      "eval_samples_per_second": 81.73,
      "eval_steps_per_second": 40.915,
      "step": 3
    },
    {
      "epoch": 0.002079272254710851,
      "grad_norm": 24.774314880371094,
      "learning_rate": 8e-05,
      "loss": 6.3232,
      "step": 4
    },
    {
      "epoch": 0.002599090318388564,
      "grad_norm": 26.77846336364746,
      "learning_rate": 0.0001,
      "loss": 6.5702,
      "step": 5
    },
    {
      "epoch": 0.0031189083820662767,
      "grad_norm": 15.674995422363281,
      "learning_rate": 0.00012,
      "loss": 7.1034,
      "step": 6
    },
    {
      "epoch": 0.0031189083820662767,
      "eval_loss": 1.7164443731307983,
      "eval_runtime": 9.9852,
      "eval_samples_per_second": 81.22,
      "eval_steps_per_second": 40.66,
      "step": 6
    },
    {
      "epoch": 0.0036387264457439896,
      "grad_norm": 19.863548278808594,
      "learning_rate": 0.00014,
      "loss": 7.4573,
      "step": 7
    },
    {
      "epoch": 0.004158544509421702,
      "grad_norm": 19.914533615112305,
      "learning_rate": 0.00016,
      "loss": 7.0145,
      "step": 8
    },
    {
      "epoch": 0.004678362573099415,
      "grad_norm": 27.733673095703125,
      "learning_rate": 0.00018,
      "loss": 6.6241,
      "step": 9
    },
    {
      "epoch": 0.004678362573099415,
      "eval_loss": 1.6902105808258057,
      "eval_runtime": 9.9899,
      "eval_samples_per_second": 81.182,
      "eval_steps_per_second": 40.641,
      "step": 9
    },
    {
      "epoch": 0.005198180636777128,
      "grad_norm": 22.087255477905273,
      "learning_rate": 0.0002,
      "loss": 6.5024,
      "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": 175457131560960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}