last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.09345794392523364,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009345794392523364,
      "grad_norm": 2.8966872692108154,
      "learning_rate": 2e-05,
      "loss": 8.5992,
      "step": 1
    },
    {
      "epoch": 0.009345794392523364,
      "eval_loss": 9.027192115783691,
      "eval_runtime": 2.7807,
      "eval_samples_per_second": 16.183,
      "eval_steps_per_second": 8.271,
      "step": 1
    },
    {
      "epoch": 0.018691588785046728,
      "grad_norm": 3.358009099960327,
      "learning_rate": 4e-05,
      "loss": 8.693,
      "step": 2
    },
    {
      "epoch": 0.028037383177570093,
      "grad_norm": 4.049684047698975,
      "learning_rate": 6e-05,
      "loss": 8.6413,
      "step": 3
    },
    {
      "epoch": 0.028037383177570093,
      "eval_loss": 9.013434410095215,
      "eval_runtime": 1.4012,
      "eval_samples_per_second": 32.115,
      "eval_steps_per_second": 16.415,
      "step": 3
    },
    {
      "epoch": 0.037383177570093455,
      "grad_norm": 3.2869625091552734,
      "learning_rate": 8e-05,
      "loss": 9.643,
      "step": 4
    },
    {
      "epoch": 0.04672897196261682,
      "grad_norm": 3.9068827629089355,
      "learning_rate": 0.0001,
      "loss": 9.2311,
      "step": 5
    },
    {
      "epoch": 0.056074766355140186,
      "grad_norm": 4.157618999481201,
      "learning_rate": 0.00012,
      "loss": 8.7479,
      "step": 6
    },
    {
      "epoch": 0.056074766355140186,
      "eval_loss": 8.742019653320312,
      "eval_runtime": 1.4092,
      "eval_samples_per_second": 31.933,
      "eval_steps_per_second": 16.321,
      "step": 6
    },
    {
      "epoch": 0.06542056074766354,
      "grad_norm": 7.017149448394775,
      "learning_rate": 0.00014,
      "loss": 7.9013,
      "step": 7
    },
    {
      "epoch": 0.07476635514018691,
      "grad_norm": 9.943305015563965,
      "learning_rate": 0.00016,
      "loss": 8.9031,
      "step": 8
    },
    {
      "epoch": 0.08411214953271028,
      "grad_norm": 6.8315110206604,
      "learning_rate": 0.00018,
      "loss": 7.1672,
      "step": 9
    },
    {
      "epoch": 0.08411214953271028,
      "eval_loss": 7.236654281616211,
      "eval_runtime": 1.4012,
      "eval_samples_per_second": 32.114,
      "eval_steps_per_second": 16.414,
      "step": 9
    },
    {
      "epoch": 0.09345794392523364,
      "grad_norm": 5.995736598968506,
      "learning_rate": 0.0002,
      "loss": 7.5009,
      "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
}