Downloads · 30 days
30
8% of all-time downloads
yuvraj17/Llama3-8B-SuperNova-Spectrum-Hermes-DPO
Llama3-8B-SuperNova-Spectrum-Hermes-DPO is a text generation model from yuvraj17. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This model is a DPO fine-tuned version of my DARETIES merged Model yuvraj17/Llama3-8B-SuperNova-Spectrum-dareties on the yuvraj17/chatml-OpenHermes2.5-dpo-binarized-alpha-2k dataset.
Downloads · 30 days
30
8% of all-time downloads
All-time downloads
395
Public
Parameters
8B
16.1 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors16.1 GB · 100%
From the Hugging Face model README
This model is a DPO fine-tuned version of my DARE_TIES merged Model yuvraj17/Llama3-8B-SuperNova-Spectrum-dare_ties on the yuvraj17/chatml-OpenHermes2.5-dpo-binarized-alpha-2k dataset.
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>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']
)
The following hyperparameters were used during training:
!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"])
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 18.00 |
| IFEval (0-Shot) | 46.91 |
| BBH (3-Shot) | 21.24 |
| MATH Lvl 5 (4-Shot) | 5.14 |
| GPQA (0-shot) | 6.94 |
| MuSR (0-shot) | 9.62 |
| MMLU-PRO (5-shot) | 18.16 |