last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.13850415512465375,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002770083102493075,
      "eval_loss": 0.8790126442909241,
      "eval_runtime": 11.2362,
      "eval_samples_per_second": 13.617,
      "eval_steps_per_second": 6.853,
      "step": 1
    },
    {
      "epoch": 0.013850415512465374,
      "grad_norm": 0.5528757572174072,
      "learning_rate": 5e-05,
      "loss": 0.8437,
      "step": 5
    },
    {
      "epoch": 0.027700831024930747,
      "grad_norm": 0.7231165766716003,
      "learning_rate": 0.0001,
      "loss": 0.8243,
      "step": 10
    },
    {
      "epoch": 0.027700831024930747,
      "eval_loss": 0.8458106517791748,
      "eval_runtime": 11.376,
      "eval_samples_per_second": 13.449,
      "eval_steps_per_second": 6.769,
      "step": 10
    },
    {
      "epoch": 0.04155124653739612,
      "grad_norm": 0.6403104662895203,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.6981,
      "step": 15
    },
    {
      "epoch": 0.055401662049861494,
      "grad_norm": 0.5902954339981079,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.6614,
      "step": 20
    },
    {
      "epoch": 0.055401662049861494,
      "eval_loss": 0.763834536075592,
      "eval_runtime": 11.7356,
      "eval_samples_per_second": 13.037,
      "eval_steps_per_second": 6.561,
      "step": 20
    },
    {
      "epoch": 0.06925207756232687,
      "grad_norm": 0.5696752071380615,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.6615,
      "step": 25
    },
    {
      "epoch": 0.08310249307479224,
      "grad_norm": 0.5728344321250916,
      "learning_rate": 5e-05,
      "loss": 0.7297,
      "step": 30
    },
    {
      "epoch": 0.08310249307479224,
      "eval_loss": 0.7355851531028748,
      "eval_runtime": 11.8476,
      "eval_samples_per_second": 12.914,
      "eval_steps_per_second": 6.499,
      "step": 30
    },
    {
      "epoch": 0.09695290858725762,
      "grad_norm": 0.5396150946617126,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.658,
      "step": 35
    },
    {
      "epoch": 0.11080332409972299,
      "grad_norm": 0.5392118096351624,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.6982,
      "step": 40
    },
    {
      "epoch": 0.11080332409972299,
      "eval_loss": 0.7300357222557068,
      "eval_runtime": 11.8185,
      "eval_samples_per_second": 12.946,
      "eval_steps_per_second": 6.515,
      "step": 40
    },
    {
      "epoch": 0.12465373961218837,
      "grad_norm": 0.6359060406684875,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.7617,
      "step": 45
    },
    {
      "epoch": 0.13850415512465375,
      "grad_norm": 0.504328727722168,
      "learning_rate": 0.0,
      "loss": 0.6836,
      "step": 50
    },
    {
      "epoch": 0.13850415512465375,
      "eval_loss": 0.7285895347595215,
      "eval_runtime": 11.7706,
      "eval_samples_per_second": 12.998,
      "eval_steps_per_second": 6.542,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.109738548887552e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}