metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- winglian/no_robots_rlhf
- HuggingFaceH4/no_robots
base_model: teknium/OpenHermes-2.5-Mistral-7B
model-index:
- name: qlora-out
results: []
openhermes-2_5-dpo-no-robots
This model is a RL fine-tuned version of teknium/OpenHermes-2.5-Mistral-7B on a preference dataset derived from HuggingFace's no robots dataset using DPO.
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-07
- train_batch_size: 2
- eval_batch_size: 8
- seed: 42
- distributed_type: multi-GPU
- num_devices: 8
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- total_eval_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 20
- training_steps: 408
Training results
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.1+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 66.40 |
| AI2 Reasoning Challenge (25-Shot) | 64.93 |
| HellaSwag (10-Shot) | 84.30 |
| MMLU (5-Shot) | 63.86 |
| TruthfulQA (0-shot) | 52.12 |
| Winogrande (5-shot) | 77.90 |
| GSM8k (5-shot) | 55.27 |