OpenAssistant-SFT-7-Llama-30B-HF

Instructions to use gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF with Transformers:

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

pipe = pipeline("text-generation", model="gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF")
model = AutoModelForCausalLM.from_pretrained("gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF

SGLang

How to use gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF 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 "gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF" \
    --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": "gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF",
		"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 "gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF" \
        --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": "gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF with Docker Model Runner:
```
docker model run hf.co/gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF
```

OpenAssistant LLaMA 30B SFT 7 HF

This in HF format repo of OpenAssistant's LLaMA 30B SFT 7.

It is the result of merging the XORs from the above repo with the original Llama 30B weights.

This is epoch 7 of OpenAssistant's training of a Llama 30B model.

Original model card

llama-30b-sft-7:
  dtype: fp16
  log_dir: "llama_log_30b"
  learning_rate: 1e-5
  model_name: /home/ubuntu/Open-Assistant/model/model_training/.saved/llama-30b-super-pretrain/checkpoint-3500
  #model_name: OpenAssistant/llama-30b-super-pretrain
  output_dir: llama_model_30b
  deepspeed_config: configs/zero3_config_sft.json
  weight_decay: 0.0
  residual_dropout: 0.0
  max_length: 2048
  use_flash_attention: true
  warmup_steps: 20
  gradient_checkpointing: true
  gradient_accumulation_steps: 12
  per_device_train_batch_size: 2
  per_device_eval_batch_size: 3
  eval_steps: 101
  save_steps: 485
  num_train_epochs: 4
  save_total_limit: 3
  use_custom_sampler: true
  sort_by_length: false
  #save_strategy: steps
  save_strategy: epoch
  datasets:
    - oasst_export:
        lang: "bg,ca,cs,da,de,en,es,fr,hr,hu,it,nl,pl,pt,ro,ru,sl,sr,sv,uk"
        input_file_path: 2023-04-12_oasst_release_ready_synth.jsonl.gz
        val_split: 0.05
    - vicuna:
        val_split: 0.05
        max_val_set: 800
        fraction: 1.0
    - dolly15k:
        val_split: 0.05
        max_val_set: 300
    - grade_school_math_instructions:
        val_split: 0.05
    - code_alpaca:
        val_split: 0.05
        max_val_set: 250

OASST dataset paper: https://arxiv.org/abs/2304.07327

Downloads last month: 15

Paper for gsaivinay/OpenAssistant-SFT-7-Llama-30B-HF

OpenAssistant Conversations -- Democratizing Large Language Model Alignment

Paper • 2304.07327 • Published Apr 14, 2023 • 10