Instructions to use rosieyzh/OLMo-150M-4xtg with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use rosieyzh/OLMo-150M-4xtg with Transformers:

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

pipe = pipeline("text-generation", model="rosieyzh/OLMo-150M-4xtg")

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("rosieyzh/OLMo-150M-4xtg")
model = AutoModelForCausalLM.from_pretrained("rosieyzh/OLMo-150M-4xtg", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use rosieyzh/OLMo-150M-4xtg with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rosieyzh/OLMo-150M-4xtg"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rosieyzh/OLMo-150M-4xtg",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/rosieyzh/OLMo-150M-4xtg

SGLang

How to use rosieyzh/OLMo-150M-4xtg 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 "rosieyzh/OLMo-150M-4xtg" \
    --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": "rosieyzh/OLMo-150M-4xtg",
		"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 "rosieyzh/OLMo-150M-4xtg" \
        --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": "rosieyzh/OLMo-150M-4xtg",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use rosieyzh/OLMo-150M-4xtg with Docker Model Runner:
```
docker model run hf.co/rosieyzh/OLMo-150M-4xtg
```

OLMo-150M-4xtg

150M OLMo model pretrained on 4 passes of the TinyGSM dataset.

Reading Model Names

Model names contain the contents of the pretraining dataset, delimited by underscores.

as: Algebraic Stack
fm3: FineMath3+
tg: TinyGSM
omi1: OpenMathInstruct1
omi2: OpenMathInstruct2

If there is a {n}x in front of the dataset abbreviation, that means it was repeated n times during pretraining. For instance, 2xtg refers to two passes over the TinyGSM dataset.

Downloads last month: 327

Safetensors

Model size

0.2B params

Tensor type

F32

Collection including rosieyzh/OLMo-150M-4xtg

OLMo-150M and OLMo-1B Pretrained Models

Collection

Pretrained models from scratch used in "Echo Chamber: RL Post-training Amplifies Behaviors Learned in Pretraining". • 12 items • Updated Jan 26 • 4