last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0005248792777661138,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.248792777661138e-05,
      "grad_norm": 1.278374195098877,
      "learning_rate": 2e-05,
      "loss": 1.1436,
      "step": 1
    },
    {
      "epoch": 5.248792777661138e-05,
      "eval_loss": 1.349988579750061,
      "eval_runtime": 809.1013,
      "eval_samples_per_second": 4.957,
      "eval_steps_per_second": 4.957,
      "step": 1
    },
    {
      "epoch": 0.00010497585555322275,
      "grad_norm": 1.7088487148284912,
      "learning_rate": 4e-05,
      "loss": 1.3843,
      "step": 2
    },
    {
      "epoch": 0.00015746378332983415,
      "grad_norm": 1.2507978677749634,
      "learning_rate": 6e-05,
      "loss": 1.2253,
      "step": 3
    },
    {
      "epoch": 0.00015746378332983415,
      "eval_loss": 1.330771803855896,
      "eval_runtime": 808.891,
      "eval_samples_per_second": 4.959,
      "eval_steps_per_second": 4.959,
      "step": 3
    },
    {
      "epoch": 0.0002099517111064455,
      "grad_norm": 1.3818600177764893,
      "learning_rate": 8e-05,
      "loss": 1.0932,
      "step": 4
    },
    {
      "epoch": 0.0002624396388830569,
      "grad_norm": 1.3986132144927979,
      "learning_rate": 0.0001,
      "loss": 1.1926,
      "step": 5
    },
    {
      "epoch": 0.0003149275666596683,
      "grad_norm": 1.6934878826141357,
      "learning_rate": 0.00012,
      "loss": 1.2556,
      "step": 6
    },
    {
      "epoch": 0.0003149275666596683,
      "eval_loss": 1.2398279905319214,
      "eval_runtime": 808.9202,
      "eval_samples_per_second": 4.958,
      "eval_steps_per_second": 4.958,
      "step": 6
    },
    {
      "epoch": 0.00036741549443627963,
      "grad_norm": 1.7230865955352783,
      "learning_rate": 0.00014,
      "loss": 1.0982,
      "step": 7
    },
    {
      "epoch": 0.000419903422212891,
      "grad_norm": 1.092279076576233,
      "learning_rate": 0.00016,
      "loss": 0.8835,
      "step": 8
    },
    {
      "epoch": 0.0004723913499895024,
      "grad_norm": 1.9888521432876587,
      "learning_rate": 0.00018,
      "loss": 1.1342,
      "step": 9
    },
    {
      "epoch": 0.0004723913499895024,
      "eval_loss": 1.1055036783218384,
      "eval_runtime": 808.9744,
      "eval_samples_per_second": 4.958,
      "eval_steps_per_second": 4.958,
      "step": 9
    },
    {
      "epoch": 0.0005248792777661138,
      "grad_norm": 1.3874531984329224,
      "learning_rate": 0.0002,
      "loss": 1.0177,
      "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": 7418872835604480.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}