last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0074850299401197605,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007485029940119761,
      "grad_norm": 25.40240478515625,
      "learning_rate": 2e-05,
      "loss": 2.15,
      "step": 1
    },
    {
      "epoch": 0.0007485029940119761,
      "eval_loss": 2.10892653465271,
      "eval_runtime": 217.0923,
      "eval_samples_per_second": 2.593,
      "eval_steps_per_second": 1.299,
      "step": 1
    },
    {
      "epoch": 0.0014970059880239522,
      "grad_norm": 18.43682861328125,
      "learning_rate": 4e-05,
      "loss": 1.7375,
      "step": 2
    },
    {
      "epoch": 0.002245508982035928,
      "grad_norm": 33.1661376953125,
      "learning_rate": 6e-05,
      "loss": 2.1489,
      "step": 3
    },
    {
      "epoch": 0.002245508982035928,
      "eval_loss": 1.6109232902526855,
      "eval_runtime": 218.0791,
      "eval_samples_per_second": 2.582,
      "eval_steps_per_second": 1.293,
      "step": 3
    },
    {
      "epoch": 0.0029940119760479044,
      "grad_norm": 20.124610900878906,
      "learning_rate": 8e-05,
      "loss": 1.2853,
      "step": 4
    },
    {
      "epoch": 0.0037425149700598802,
      "grad_norm": 10.917786598205566,
      "learning_rate": 0.0001,
      "loss": 0.8807,
      "step": 5
    },
    {
      "epoch": 0.004491017964071856,
      "grad_norm": 12.437239646911621,
      "learning_rate": 0.00012,
      "loss": 0.8447,
      "step": 6
    },
    {
      "epoch": 0.004491017964071856,
      "eval_loss": 0.5942753553390503,
      "eval_runtime": 218.8449,
      "eval_samples_per_second": 2.573,
      "eval_steps_per_second": 1.289,
      "step": 6
    },
    {
      "epoch": 0.005239520958083832,
      "grad_norm": 12.0173978805542,
      "learning_rate": 0.00014,
      "loss": 0.6145,
      "step": 7
    },
    {
      "epoch": 0.005988023952095809,
      "grad_norm": 6.952859401702881,
      "learning_rate": 0.00016,
      "loss": 0.4532,
      "step": 8
    },
    {
      "epoch": 0.006736526946107785,
      "grad_norm": 2.8368070125579834,
      "learning_rate": 0.00018,
      "loss": 0.2738,
      "step": 9
    },
    {
      "epoch": 0.006736526946107785,
      "eval_loss": 0.3004415035247803,
      "eval_runtime": 216.9304,
      "eval_samples_per_second": 2.595,
      "eval_steps_per_second": 1.3,
      "step": 9
    },
    {
      "epoch": 0.0074850299401197605,
      "grad_norm": 3.8192362785339355,
      "learning_rate": 0.0002,
      "loss": 0.3159,
      "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": 720954300825600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}