last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0005236698785085882,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.236698785085882e-05,
      "grad_norm": 0.4392659366130829,
      "learning_rate": 2e-05,
      "loss": 0.5208,
      "step": 1
    },
    {
      "epoch": 5.236698785085882e-05,
      "eval_loss": 0.6633045673370361,
      "eval_runtime": 722.4901,
      "eval_samples_per_second": 11.13,
      "eval_steps_per_second": 5.565,
      "step": 1
    },
    {
      "epoch": 0.00010473397570171764,
      "grad_norm": 0.5572344660758972,
      "learning_rate": 4e-05,
      "loss": 0.6158,
      "step": 2
    },
    {
      "epoch": 0.00015710096355257646,
      "grad_norm": 0.6260502338409424,
      "learning_rate": 6e-05,
      "loss": 0.6381,
      "step": 3
    },
    {
      "epoch": 0.00015710096355257646,
      "eval_loss": 0.6599137187004089,
      "eval_runtime": 722.6318,
      "eval_samples_per_second": 11.127,
      "eval_steps_per_second": 5.564,
      "step": 3
    },
    {
      "epoch": 0.00020946795140343527,
      "grad_norm": 0.47606730461120605,
      "learning_rate": 8e-05,
      "loss": 0.5238,
      "step": 4
    },
    {
      "epoch": 0.0002618349392542941,
      "grad_norm": 0.46592414379119873,
      "learning_rate": 0.0001,
      "loss": 0.5844,
      "step": 5
    },
    {
      "epoch": 0.0003142019271051529,
      "grad_norm": 0.5224045515060425,
      "learning_rate": 0.00012,
      "loss": 0.6448,
      "step": 6
    },
    {
      "epoch": 0.0003142019271051529,
      "eval_loss": 0.6183779835700989,
      "eval_runtime": 723.037,
      "eval_samples_per_second": 11.121,
      "eval_steps_per_second": 5.561,
      "step": 6
    },
    {
      "epoch": 0.00036656891495601173,
      "grad_norm": 0.38573431968688965,
      "learning_rate": 0.00014,
      "loss": 0.4903,
      "step": 7
    },
    {
      "epoch": 0.00041893590280687055,
      "grad_norm": 0.4179316759109497,
      "learning_rate": 0.00016,
      "loss": 0.4992,
      "step": 8
    },
    {
      "epoch": 0.00047130289065772937,
      "grad_norm": 0.6486102342605591,
      "learning_rate": 0.00018,
      "loss": 0.6087,
      "step": 9
    },
    {
      "epoch": 0.00047130289065772937,
      "eval_loss": 0.5971018075942993,
      "eval_runtime": 722.94,
      "eval_samples_per_second": 11.123,
      "eval_steps_per_second": 5.562,
      "step": 9
    },
    {
      "epoch": 0.0005236698785085882,
      "grad_norm": 0.5129334330558777,
      "learning_rate": 0.0002,
      "loss": 0.5634,
      "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": 4104787234652160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}