last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007860090391039498,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007860090391039497,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 18.5638,
      "step": 1
    },
    {
      "epoch": 0.0007860090391039497,
      "eval_loss": 17.82631492614746,
      "eval_runtime": 25.1316,
      "eval_samples_per_second": 10.664,
      "eval_steps_per_second": 10.664,
      "step": 1
    },
    {
      "epoch": 0.0015720180782078995,
      "grad_norm": 13.52338695526123,
      "learning_rate": 2e-05,
      "loss": 17.9821,
      "step": 2
    },
    {
      "epoch": 0.002358027117311849,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 18.3539,
      "step": 3
    },
    {
      "epoch": 0.002358027117311849,
      "eval_loss": 17.82631492614746,
      "eval_runtime": 24.3766,
      "eval_samples_per_second": 10.994,
      "eval_steps_per_second": 10.994,
      "step": 3
    },
    {
      "epoch": 0.003144036156415799,
      "grad_norm": 11.923038482666016,
      "learning_rate": 4e-05,
      "loss": 18.0281,
      "step": 4
    },
    {
      "epoch": 0.003930045195519749,
      "grad_norm": 12.763206481933594,
      "learning_rate": 6e-05,
      "loss": 17.3767,
      "step": 5
    },
    {
      "epoch": 0.004716054234623698,
      "grad_norm": 11.823870658874512,
      "learning_rate": 8e-05,
      "loss": 17.507,
      "step": 6
    },
    {
      "epoch": 0.004716054234623698,
      "eval_loss": 17.262771606445312,
      "eval_runtime": 24.3977,
      "eval_samples_per_second": 10.985,
      "eval_steps_per_second": 10.985,
      "step": 6
    },
    {
      "epoch": 0.005502063273727648,
      "grad_norm": 14.074959754943848,
      "learning_rate": 0.0001,
      "loss": 17.4521,
      "step": 7
    },
    {
      "epoch": 0.006288072312831598,
      "grad_norm": 16.047016143798828,
      "learning_rate": 0.00012,
      "loss": 17.0565,
      "step": 8
    },
    {
      "epoch": 0.007074081351935547,
      "grad_norm": 20.59668731689453,
      "learning_rate": 0.00014,
      "loss": 15.4486,
      "step": 9
    },
    {
      "epoch": 0.007074081351935547,
      "eval_loss": 15.391473770141602,
      "eval_runtime": 24.3734,
      "eval_samples_per_second": 10.996,
      "eval_steps_per_second": 10.996,
      "step": 9
    },
    {
      "epoch": 0.007860090391039498,
      "grad_norm": 39.99613952636719,
      "learning_rate": 0.00016,
      "loss": 15.2566,
      "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": 1306268836823040.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}