checkpoint-250/trainer_state.json

{
  "best_metric": 0.7510997653007507,
  "best_model_checkpoint": "./mistral7b/08-02-24-Weni-Zeroshot-3.2.3-Mistral-7B-pipeline-config_zeroshot-2_max_steps-342_batch_256_2024-02-08/checkpoint-250",
  "epoch": 4.048582995951417,
  "eval_steps": 50,
  "global_step": 250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.81,
      "learning_rate": 0.0002,
      "loss": 1.0342,
      "step": 50
    },
    {
      "epoch": 0.81,
      "eval_loss": 0.8600249290466309,
      "eval_runtime": 139.2702,
      "eval_samples_per_second": 3.145,
      "eval_steps_per_second": 0.201,
      "step": 50
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0002,
      "loss": 0.8239,
      "step": 100
    },
    {
      "epoch": 1.62,
      "eval_loss": 0.8089107871055603,
      "eval_runtime": 139.1481,
      "eval_samples_per_second": 3.148,
      "eval_steps_per_second": 0.201,
      "step": 100
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.0002,
      "loss": 0.7904,
      "step": 150
    },
    {
      "epoch": 2.43,
      "eval_loss": 0.7841466069221497,
      "eval_runtime": 139.3201,
      "eval_samples_per_second": 3.144,
      "eval_steps_per_second": 0.201,
      "step": 150
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.0002,
      "loss": 0.7596,
      "step": 200
    },
    {
      "epoch": 3.24,
      "eval_loss": 0.7661702632904053,
      "eval_runtime": 138.9687,
      "eval_samples_per_second": 3.152,
      "eval_steps_per_second": 0.201,
      "step": 200
    },
    {
      "epoch": 4.05,
      "learning_rate": 0.0002,
      "loss": 0.7394,
      "step": 250
    },
    {
      "epoch": 4.05,
      "eval_loss": 0.7510997653007507,
      "eval_runtime": 139.2116,
      "eval_samples_per_second": 3.146,
      "eval_steps_per_second": 0.201,
      "step": 250
    }
  ],
  "logging_steps": 50,
  "max_steps": 342,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 50,
  "total_flos": 1.3989418538473882e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}