---
base_model: Qwen/Qwen2.5-7B
library_name: peft
license: apache-2.0
tags:
- llama-factory
- lora
- generated_from_trainer
model-index:
- name: Inverse-Qwen-2.5-BlackBox-7B-LoRA-Adapter
  results: []
pipeline_tag: text-generation
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# Inverse-Qwen-2.5-BlackBox-7B-LoRA-Adapter 

This model is a fine-tuned version of [Qwen/Qwen2.5-7B](https://huggingface.co/Qwen/Qwen2.5-7B) on the inv_qwen_inf-ins_660k dataset for paper [Beyond One-Size-Fits-All: Inversion Learning for Highly Effective NLG Evaluation Prompts](arxiv.org/abs/2504.21117).

Code: https://github.com/blackboxllm/llm_evaluation_inversion/tree/main.

## Model description

LoRA Adapter for Inverse-Qwen-2.5-7B. Please use with the originl Qwen2.5-7B base model.


The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 8
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 32
- total_train_batch_size: 1024
- total_eval_batch_size: 4
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3.0

### Framework versions

- PEFT 0.12.0
- Transformers 4.46.1
- Pytorch 2.4.0+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3
---
license: apache-2.0
---