last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.019782393669634024,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0019782393669634025,
      "grad_norm": 1.4230250120162964,
      "learning_rate": 2e-05,
      "loss": 3.2187,
      "step": 1
    },
    {
      "epoch": 0.003956478733926805,
      "grad_norm": 1.3153244256973267,
      "learning_rate": 4e-05,
      "loss": 3.2003,
      "step": 2
    },
    {
      "epoch": 0.005934718100890208,
      "grad_norm": 1.88228178024292,
      "learning_rate": 6e-05,
      "loss": 3.3319,
      "step": 3
    },
    {
      "epoch": 0.005934718100890208,
      "eval_loss": 3.3604626655578613,
      "eval_runtime": 12.0631,
      "eval_samples_per_second": 17.657,
      "eval_steps_per_second": 8.87,
      "step": 3
    },
    {
      "epoch": 0.00791295746785361,
      "grad_norm": 1.8003789186477661,
      "learning_rate": 8e-05,
      "loss": 3.35,
      "step": 4
    },
    {
      "epoch": 0.009891196834817012,
      "grad_norm": 1.578134298324585,
      "learning_rate": 0.0001,
      "loss": 3.1165,
      "step": 5
    },
    {
      "epoch": 0.011869436201780416,
      "grad_norm": 1.6146352291107178,
      "learning_rate": 0.00012,
      "loss": 3.1276,
      "step": 6
    },
    {
      "epoch": 0.011869436201780416,
      "eval_loss": 3.161781072616577,
      "eval_runtime": 14.4687,
      "eval_samples_per_second": 14.721,
      "eval_steps_per_second": 7.395,
      "step": 6
    },
    {
      "epoch": 0.013847675568743818,
      "grad_norm": 1.8498936891555786,
      "learning_rate": 0.00014,
      "loss": 3.0711,
      "step": 7
    },
    {
      "epoch": 0.01582591493570722,
      "grad_norm": 1.8731578588485718,
      "learning_rate": 0.00016,
      "loss": 2.6369,
      "step": 8
    },
    {
      "epoch": 0.017804154302670624,
      "grad_norm": 2.1558492183685303,
      "learning_rate": 0.00018,
      "loss": 2.8718,
      "step": 9
    },
    {
      "epoch": 0.017804154302670624,
      "eval_loss": 2.8179714679718018,
      "eval_runtime": 12.1815,
      "eval_samples_per_second": 17.486,
      "eval_steps_per_second": 8.784,
      "step": 9
    },
    {
      "epoch": 0.019782393669634024,
      "grad_norm": 2.164289712905884,
      "learning_rate": 0.0002,
      "loss": 2.5339,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}