last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03196164602477027,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0031961646024770275,
      "grad_norm": 1.6692992448806763,
      "learning_rate": 1e-05,
      "loss": 5.2274,
      "step": 1
    },
    {
      "epoch": 0.0031961646024770275,
      "eval_loss": 6.05078125,
      "eval_runtime": 100.6272,
      "eval_samples_per_second": 5.237,
      "eval_steps_per_second": 2.624,
      "step": 1
    },
    {
      "epoch": 0.006392329204954055,
      "grad_norm": 1.1485488414764404,
      "learning_rate": 2e-05,
      "loss": 5.9317,
      "step": 2
    },
    {
      "epoch": 0.009588493807431082,
      "grad_norm": 0.9786986708641052,
      "learning_rate": 3e-05,
      "loss": 5.7836,
      "step": 3
    },
    {
      "epoch": 0.009588493807431082,
      "eval_loss": 5.9795355796813965,
      "eval_runtime": 11.5612,
      "eval_samples_per_second": 45.583,
      "eval_steps_per_second": 22.835,
      "step": 3
    },
    {
      "epoch": 0.01278465840990811,
      "grad_norm": 1.0295538902282715,
      "learning_rate": 4e-05,
      "loss": 5.6168,
      "step": 4
    },
    {
      "epoch": 0.015980823012385136,
      "grad_norm": 1.2927321195602417,
      "learning_rate": 5e-05,
      "loss": 5.281,
      "step": 5
    },
    {
      "epoch": 0.019176987614862164,
      "grad_norm": 1.1961203813552856,
      "learning_rate": 6e-05,
      "loss": 5.202,
      "step": 6
    },
    {
      "epoch": 0.019176987614862164,
      "eval_loss": 5.88431978225708,
      "eval_runtime": 11.6355,
      "eval_samples_per_second": 45.292,
      "eval_steps_per_second": 22.689,
      "step": 6
    },
    {
      "epoch": 0.022373152217339192,
      "grad_norm": 2.880462646484375,
      "learning_rate": 7e-05,
      "loss": 5.6402,
      "step": 7
    },
    {
      "epoch": 0.02556931681981622,
      "grad_norm": 4.132447719573975,
      "learning_rate": 8e-05,
      "loss": 5.5549,
      "step": 8
    },
    {
      "epoch": 0.028765481422293248,
      "grad_norm": 2.7458152770996094,
      "learning_rate": 9e-05,
      "loss": 5.3865,
      "step": 9
    },
    {
      "epoch": 0.028765481422293248,
      "eval_loss": 5.598178863525391,
      "eval_runtime": 11.6299,
      "eval_samples_per_second": 45.314,
      "eval_steps_per_second": 22.7,
      "step": 9
    },
    {
      "epoch": 0.03196164602477027,
      "grad_norm": 3.523580312728882,
      "learning_rate": 0.0001,
      "loss": 5.2641,
      "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": 6970544184360960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}