Instructions to use Kukedlc/NeuralKrishna-7B-V2-DPO with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Kukedlc/NeuralKrishna-7B-V2-DPO with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Kukedlc/NeuralKrishna-7B-V2-DPO")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Kukedlc/NeuralKrishna-7B-V2-DPO")
model = AutoModelForCausalLM.from_pretrained("Kukedlc/NeuralKrishna-7B-V2-DPO")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Kukedlc/NeuralKrishna-7B-V2-DPO with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Kukedlc/NeuralKrishna-7B-V2-DPO"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Kukedlc/NeuralKrishna-7B-V2-DPO",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Kukedlc/NeuralKrishna-7B-V2-DPO

SGLang

How to use Kukedlc/NeuralKrishna-7B-V2-DPO with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Kukedlc/NeuralKrishna-7B-V2-DPO" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Kukedlc/NeuralKrishna-7B-V2-DPO",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Kukedlc/NeuralKrishna-7B-V2-DPO" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Kukedlc/NeuralKrishna-7B-V2-DPO",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Kukedlc/NeuralKrishna-7B-V2-DPO with Docker Model Runner:
```
docker model run hf.co/Kukedlc/NeuralKrishna-7B-V2-DPO
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Neural Krishna DPO

Fine-tuning + lnegth(choose)

Training Args:

# LoRA configuration
peft_config = LoraConfig(
    r=16,
    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']
)

# Model to fine-tune
model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.float16,
    load_in_4bit=True
)
model.config.use_cache = False



# Training arguments
training_args = TrainingArguments(
    per_device_train_batch_size=4,
    gradient_accumulation_steps=4,
    gradient_checkpointing=True,
    learning_rate=5e-5,
    lr_scheduler_type="cosine",
    max_steps=120,
    save_strategy="no",
    logging_steps=1,
    output_dir=new_model,
    optim="paged_adamw_32bit",
    warmup_steps=50,
    bf16=True,
    report_to="wandb",
)

# Create DPO trainer
dpo_trainer = DPOTrainer(
    model,
    args=training_args,
    train_dataset=dataset,
    tokenizer=tokenizer,
    peft_config=peft_config,
    beta=0.1,
    max_prompt_length=1024,
    max_length=1536,
)

# Fine-tune model with DPO
dpo_trainer.train()

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	76.00
AI2 Reasoning Challenge (25-Shot)	74.06
HellaSwag (10-Shot)	88.97
MMLU (5-Shot)	64.41
TruthfulQA (0-shot)	76.19
Winogrande (5-shot)	84.29
GSM8k (5-shot)	68.08

Downloads last month: 92

Safetensors

Model size

7B params

Tensor type

F16

Model tree for Kukedlc/NeuralKrishna-7B-V2-DPO

Merges

2 models

Quantizations

3 models

Spaces using Kukedlc/NeuralKrishna-7B-V2-DPO 8

Evaluation results

normalized accuracy on AI2 Reasoning Challenge (25-Shot)
test set Open LLM Leaderboard

74.060
normalized accuracy on HellaSwag (10-Shot)
validation set Open LLM Leaderboard

88.970
accuracy on MMLU (5-Shot)
test set Open LLM Leaderboard

64.410
mc2 on TruthfulQA (0-shot)
validation set Open LLM Leaderboard

76.190
accuracy on Winogrande (5-shot)
validation set Open LLM Leaderboard

84.290
accuracy on GSM8k (5-shot)
test set Open LLM Leaderboard

68.080