trainer_state.json

{
  "best_metric": 0.5984224677085876,
  "best_model_checkpoint": "/mnt/beegfs/farid/mlora/outputs/xnli/aya-101/ar,en,es,fr,hi,ru,sw,zh,th,tr/rank4_lr5e-5/checkpoint-1000",
  "epoch": 0.004074332115108031,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0020371660575540153,
      "grad_norm": 1.2531261444091797,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.1205,
      "step": 500
    },
    {
      "epoch": 0.0020371660575540153,
      "eval_accuracy": 0.41,
      "eval_f1": 0.3911797336690286,
      "eval_loss": 1.0892360210418701,
      "eval_runtime": 38.2869,
      "eval_samples_per_second": 5.224,
      "eval_steps_per_second": 5.224,
      "step": 500
    },
    {
      "epoch": 0.004074332115108031,
      "grad_norm": 8.405991554260254,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.8882,
      "step": 1000
    },
    {
      "epoch": 0.004074332115108031,
      "eval_accuracy": 0.775,
      "eval_f1": 0.7729114275922142,
      "eval_loss": 0.5984224677085876,
      "eval_runtime": 37.8206,
      "eval_samples_per_second": 5.288,
      "eval_steps_per_second": 5.288,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 6000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 1.33861145088e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}