File size: 2,651 Bytes
302527c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 |
---
language:
- en
library_name: transformers
license: apache-2.0
pipeline_tag: text-generation
tags:
- dpo
- rlhf
- trl
- autoquant
- gguf
---
# Llama3-8B-SuperNova-Spectrum-Hermes-DPO
This model is a **DPO fine-tuned** version of my `DARE_TIES` merged Model [`yuvraj17/Llama3-8B-SuperNova-Spectrum-dare_ties`](https://huggingface.co/yuvraj17/Llama3-8B-SuperNova-Spectrum-dare_ties) on the [yuvraj17/chatml-OpenHermes2.5-dpo-binarized-alpha-2k](https://huggingface.co/datasets/yuvraj17/chatml-OpenHermes2.5-dpo-binarized-alpha-2k) dataset.
## DPO (Direct Preference Optimization):
Direct Preference Optimization (DPO) is a fine-tuning technique that focuses on aligning a model's responses with human preferences or ranking data without requiring reinforcement learning steps, like in RLHF.
<figure>
<img src="https://cdn-uploads.huggingface.co/production/uploads/66137d95e8d2cda230ddcea6/kHcU5dkcSVqxEIWt_GRUB.png" width="1000" height="768">
<figcaption> DPO vs RLHF <a href="//arxiv.org/abs/2305.18290">Reference</a> </figcaption>
</figure>
## Training:
- Trained on **1x A40s (48GB VRAM)** using the [HuggingFace TRL](https://huggingface.co/docs/trl/index).
- **QLoRA**(`4-bit precision`) for 1 epoch
```
# LoRA configuration
peft_config = LoraConfig(
r=32,
lora_alpha=16,
lora_dropout=0.05,
bias="none",
task_type="CAUSAL_LM",
target_modules=['k_proj', 'gate_proj', 'v_proj', 'up_proj', 'q_proj', 'o_proj', 'down_proj']
)
```
### Training Params
The following hyperparameters were used during training:
- learning_rate: 5e-05
- beta=0.1
- num_devices: 1
- gradient_accumulation_steps: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- num_epochs: 1
### Training Time = **1:57:00** hours
### Weight & Biases Report
[Report-Link](https://api.wandb.ai/links/my-sft-team/d211juao)
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "yuvraj17/Llama3-8B-SuperNova-Spectrum-Hermes-DPO"
messages = [{"role": "user", "content": "What is a large language model?"}]
tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
```
## 🏆 Evaluation Scores
Coming Soon |