Instructions to use htlou/mm-interp-AA_preference_random_0_40 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use htlou/mm-interp-AA_preference_random_0_40 with Transformers:

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

pipe = pipeline("image-text-to-text", model="htlou/mm-interp-AA_preference_random_0_40")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForImageTextToText

processor = AutoProcessor.from_pretrained("htlou/mm-interp-AA_preference_random_0_40")
model = AutoModelForImageTextToText.from_pretrained("htlou/mm-interp-AA_preference_random_0_40")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use htlou/mm-interp-AA_preference_random_0_40 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "htlou/mm-interp-AA_preference_random_0_40"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "htlou/mm-interp-AA_preference_random_0_40",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/htlou/mm-interp-AA_preference_random_0_40

SGLang

How to use htlou/mm-interp-AA_preference_random_0_40 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 "htlou/mm-interp-AA_preference_random_0_40" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "htlou/mm-interp-AA_preference_random_0_40",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "htlou/mm-interp-AA_preference_random_0_40" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "htlou/mm-interp-AA_preference_random_0_40",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use htlou/mm-interp-AA_preference_random_0_40 with Docker Model Runner:
```
docker model run hf.co/htlou/mm-interp-AA_preference_random_0_40
```

AA_preference_random_0_40

This model is a fine-tuned version of llava-hf/llava-v1.6-mistral-7b-hf on the AA_preference_random_0_40 dataset. It achieves the following results on the evaluation set:

Loss: 0.5902
Rewards/chosen: 1.1361
Rewards/rejected: -0.8413
Rewards/accuracies: 0.7552
Rewards/margins: 1.9774
Logps/rejected: -247.4370
Logps/chosen: -255.4082
Logits/rejected: -2.3629
Logits/chosen: -2.3927

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-06
train_batch_size: 8
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 8
gradient_accumulation_steps: 4
total_train_batch_size: 256
total_eval_batch_size: 64
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 10
num_epochs: 3.0

Training results

Training Loss	Epoch	Step	Validation Loss	Rewards/chosen	Rewards/rejected	Rewards/accuracies	Rewards/margins	Logps/rejected	Logps/chosen	Logits/rejected	Logits/chosen
0.5323	0.9346	50	0.5824	0.9205	-0.3508	0.7396	1.2713	-242.5313	-257.5635	-2.3512	-2.3802
0.2441	1.8692	100	0.5841	1.0720	-0.7661	0.7708	1.8380	-246.6841	-256.0490	-2.3634	-2.3957
0.1203	2.8037	150	0.5899	1.1373	-0.8378	0.7760	1.9751	-247.4010	-255.3957	-2.3639	-2.3938

Framework versions

Transformers 4.45.2
Pytorch 2.4.0+cu121
Datasets 2.21.0
Tokenizers 0.20.3

Downloads last month: 4

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for htlou/mm-interp-AA_preference_random_0_40

Base model

llava-hf/llava-v1.6-mistral-7b-hf

Finetuned

(188)

this model