last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.024198427102238355,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0024198427102238356,
      "grad_norm": 3.1973347663879395,
      "learning_rate": 2e-05,
      "loss": 44.3366,
      "step": 1
    },
    {
      "epoch": 0.0024198427102238356,
      "eval_loss": 11.064077377319336,
      "eval_runtime": 2.8623,
      "eval_samples_per_second": 60.79,
      "eval_steps_per_second": 30.395,
      "step": 1
    },
    {
      "epoch": 0.004839685420447671,
      "grad_norm": 4.02877140045166,
      "learning_rate": 4e-05,
      "loss": 44.3419,
      "step": 2
    },
    {
      "epoch": 0.007259528130671506,
      "grad_norm": 4.1353325843811035,
      "learning_rate": 6e-05,
      "loss": 44.3511,
      "step": 3
    },
    {
      "epoch": 0.007259528130671506,
      "eval_loss": 11.062400817871094,
      "eval_runtime": 2.498,
      "eval_samples_per_second": 69.656,
      "eval_steps_per_second": 34.828,
      "step": 3
    },
    {
      "epoch": 0.009679370840895343,
      "grad_norm": 3.9328174591064453,
      "learning_rate": 8e-05,
      "loss": 44.2043,
      "step": 4
    },
    {
      "epoch": 0.012099213551119177,
      "grad_norm": 3.8412272930145264,
      "learning_rate": 0.0001,
      "loss": 44.2909,
      "step": 5
    },
    {
      "epoch": 0.014519056261343012,
      "grad_norm": 3.7023684978485107,
      "learning_rate": 0.00012,
      "loss": 44.2077,
      "step": 6
    },
    {
      "epoch": 0.014519056261343012,
      "eval_loss": 11.053502082824707,
      "eval_runtime": 2.5656,
      "eval_samples_per_second": 67.821,
      "eval_steps_per_second": 33.91,
      "step": 6
    },
    {
      "epoch": 0.01693889897156685,
      "grad_norm": 4.074832439422607,
      "learning_rate": 0.00014,
      "loss": 44.1527,
      "step": 7
    },
    {
      "epoch": 0.019358741681790685,
      "grad_norm": 3.816525936126709,
      "learning_rate": 0.00016,
      "loss": 44.3855,
      "step": 8
    },
    {
      "epoch": 0.021778584392014518,
      "grad_norm": 3.581460475921631,
      "learning_rate": 0.00018,
      "loss": 44.1724,
      "step": 9
    },
    {
      "epoch": 0.021778584392014518,
      "eval_loss": 11.039490699768066,
      "eval_runtime": 2.5528,
      "eval_samples_per_second": 68.16,
      "eval_steps_per_second": 34.08,
      "step": 9
    },
    {
      "epoch": 0.024198427102238355,
      "grad_norm": 2.9207425117492676,
      "learning_rate": 0.0002,
      "loss": 44.1771,
      "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": 74868326400.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}