last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0007845136995704787,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.845136995704788e-05,
      "grad_norm": 0.9127047061920166,
      "learning_rate": 2e-05,
      "loss": 2.1072,
      "step": 1
    },
    {
      "epoch": 7.845136995704788e-05,
      "eval_loss": 2.1771984100341797,
      "eval_runtime": 364.9344,
      "eval_samples_per_second": 14.709,
      "eval_steps_per_second": 7.355,
      "step": 1
    },
    {
      "epoch": 0.00015690273991409576,
      "grad_norm": 1.0036420822143555,
      "learning_rate": 4e-05,
      "loss": 2.4194,
      "step": 2
    },
    {
      "epoch": 0.00023535410987114363,
      "grad_norm": 1.1165186166763306,
      "learning_rate": 6e-05,
      "loss": 2.1131,
      "step": 3
    },
    {
      "epoch": 0.00023535410987114363,
      "eval_loss": 2.1711175441741943,
      "eval_runtime": 363.6197,
      "eval_samples_per_second": 14.763,
      "eval_steps_per_second": 7.381,
      "step": 3
    },
    {
      "epoch": 0.0003138054798281915,
      "grad_norm": 0.8401862382888794,
      "learning_rate": 8e-05,
      "loss": 1.9984,
      "step": 4
    },
    {
      "epoch": 0.00039225684978523936,
      "grad_norm": 0.746632993221283,
      "learning_rate": 0.0001,
      "loss": 2.2822,
      "step": 5
    },
    {
      "epoch": 0.00047070821974228727,
      "grad_norm": 0.830866277217865,
      "learning_rate": 0.00012,
      "loss": 2.1672,
      "step": 6
    },
    {
      "epoch": 0.00047070821974228727,
      "eval_loss": 2.1076629161834717,
      "eval_runtime": 363.7362,
      "eval_samples_per_second": 14.758,
      "eval_steps_per_second": 7.379,
      "step": 6
    },
    {
      "epoch": 0.0005491595896993352,
      "grad_norm": 0.7782320380210876,
      "learning_rate": 0.00014,
      "loss": 2.1435,
      "step": 7
    },
    {
      "epoch": 0.000627610959656383,
      "grad_norm": 0.712628960609436,
      "learning_rate": 0.00016,
      "loss": 1.8562,
      "step": 8
    },
    {
      "epoch": 0.0007060623296134309,
      "grad_norm": 0.6478232741355896,
      "learning_rate": 0.00018,
      "loss": 2.147,
      "step": 9
    },
    {
      "epoch": 0.0007060623296134309,
      "eval_loss": 1.9906666278839111,
      "eval_runtime": 363.6948,
      "eval_samples_per_second": 14.76,
      "eval_steps_per_second": 7.38,
      "step": 9
    },
    {
      "epoch": 0.0007845136995704787,
      "grad_norm": 0.7166277766227722,
      "learning_rate": 0.0002,
      "loss": 1.9556,
      "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": 3960342577152000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}