Edit model card

tulu2-13b-cost-UI-UF-7bjudge

This model is a fine-tuned version of allenai/tulu-2-13b on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6924
  • Rewards/chosen: 0.0238
  • Rewards/rejected: 0.0217
  • Rewards/accuracies: 0.5640
  • Rewards/margins: 0.0021
  • Rewards/margins Max: 0.0393
  • Rewards/margins Min: -0.0362
  • Rewards/margins Std: 0.0335
  • Logps/rejected: -313.6311
  • Logps/chosen: -321.7533
  • Logits/rejected: -1.0117
  • Logits/chosen: -1.1541

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-07
  • train_batch_size: 1
  • eval_batch_size: 4
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 8
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 16
  • total_eval_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 1

Training results

Training Loss Epoch Step Validation Loss Rewards/chosen Rewards/rejected Rewards/accuracies Rewards/margins Rewards/margins Max Rewards/margins Min Rewards/margins Std Logps/rejected Logps/chosen Logits/rejected Logits/chosen
0.6643 1.0 1350 0.6924 0.0238 0.0217 0.5640 0.0021 0.0393 -0.0362 0.0335 -313.6311 -321.7533 -1.0117 -1.1541

Framework versions

  • PEFT 0.7.1
  • Transformers 4.39.0.dev0
  • Pytorch 2.1.2+cu121
  • Datasets 2.14.6
  • Tokenizers 0.15.2
Downloads last month
0
Unable to determine this model’s pipeline type. Check the docs .

Adapter for