Instructions to use ryandt/dfe-l3.1-writing-strats with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ryandt/dfe-l3.1-writing-strats with Transformers:

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

pipe = pipeline("text-generation", model="ryandt/dfe-l3.1-writing-strats")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ryandt/dfe-l3.1-writing-strats")
model = AutoModelForCausalLM.from_pretrained("ryandt/dfe-l3.1-writing-strats", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ryandt/dfe-l3.1-writing-strats with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ryandt/dfe-l3.1-writing-strats"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ryandt/dfe-l3.1-writing-strats",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ryandt/dfe-l3.1-writing-strats

SGLang

How to use ryandt/dfe-l3.1-writing-strats 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 "ryandt/dfe-l3.1-writing-strats" \
    --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": "ryandt/dfe-l3.1-writing-strats",
		"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 "ryandt/dfe-l3.1-writing-strats" \
        --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": "ryandt/dfe-l3.1-writing-strats",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ryandt/dfe-l3.1-writing-strats with Docker Model Runner:
```
docker model run hf.co/ryandt/dfe-l3.1-writing-strats
```

Model Card for Model ID

This is a quick test at building a model that engages in a more "pedagogically grounded" rhetoric when helping students brainstorm.

It was developed in about a day as a proof of concept.

Model Description

This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.

Developed by: Ryan Tannenbaum and For.Education
Model type: Llama 3.1 8B
Language(s) (NLP): English
License: MIT
Finetuned from model: Llama 3.1 8B

Model Sources [optional]

Demo [optional]: Coming soon

Uses

The model uses the following formatting:

### USER: <What the user says>

### ASSISTANT: <The bot response>

...

### TERMINATE

The model is programmed to "TERMINATE" the session when it reaches the end of its conversation

Bias, Risks, and Limitations

This model uses an incredibly small dataset to tackle a very specific use case. It is a proof of concept.

Training Data

Essay Dataset

Training Procedure

Trained with AutoTrain locally on 4090 card:

Training Hyperparameters

Epochs: 5 Learning Rate: 2e-5 Train Batch Size: 2 Mixed Precision: fp16 Quantization: int8

Model Card Authors [optional]

Ryan Tannenbaum (ryandt)

Model Card Contact

Ryan Tannenbaum

Downloads last month: 5

Safetensors

Model size

8B params

Tensor type

F16

ryandt
/

dfe-l3.1-writing-strats