Instructions to use nilq/mistral-1L-tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nilq/mistral-1L-tiny with Transformers:

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

pipe = pipeline("text-generation", model="nilq/mistral-1L-tiny")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nilq/mistral-1L-tiny")
model = AutoModelForCausalLM.from_pretrained("nilq/mistral-1L-tiny")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use nilq/mistral-1L-tiny with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/nilq/mistral-1L-tiny

SGLang

How to use nilq/mistral-1L-tiny 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 "nilq/mistral-1L-tiny" \
    --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": "nilq/mistral-1L-tiny",
		"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 "nilq/mistral-1L-tiny" \
        --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": "nilq/mistral-1L-tiny",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use nilq/mistral-1L-tiny with Docker Model Runner:
```
docker model run hf.co/nilq/mistral-1L-tiny
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

mistral-1L-tiny

A tiny single-layer 35.1M parameter Mistral model, with a hidden size of 512, and an MLP intermediate size of 1024. This model is trained on the roneneldan/TinyStories dataset. It achieves the following results on the evaluation set:

Loss: 1.6868
Accuracy: 0.5792

Model description

This work is inspired by the 21M parameter one-layer GPT-Neo of the Tiny Stories paper. Results reproduced to acquire high-frequency checkpoints for further analysis.

Intended uses & limitations

Analysis of feature dynamics and emergence in real-world language models.

Training procedure

Trained for 90171 steps, corresponding to ~2 hours on a single H100.

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0006
train_batch_size: 64
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
num_epochs: 3.0

Training results

Quite consistent English text generation.

Framework versions

Transformers 4.38.1
Pytorch 2.2.0+cu121
Datasets 2.17.1
Tokenizers 0.15.2

Downloads last month: 91

Safetensors

Model size

35.1M params

Tensor type

F32

Model tree for nilq/mistral-1L-tiny

Merges

2 models

Quantizations

1 model

Dataset used to train nilq/mistral-1L-tiny

Space using nilq/mistral-1L-tiny 1

Collection including nilq/mistral-1L-tiny

Toy Models to Study

Collection

9 items • Updated Mar 17, 2024 • 2

Paper for nilq/mistral-1L-tiny

TinyStories: How Small Can Language Models Be and Still Speak Coherent English?

Paper • 2305.07759 • Published May 12, 2023 • 45

Evaluation results

Accuracy on roneneldan/TinyStories
self-reported

0.579