Instructions to use inverse-scaling/opt-13b_eval with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use inverse-scaling/opt-13b_eval with Transformers:

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

pipe = pipeline("text-generation", model="inverse-scaling/opt-13b_eval")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("inverse-scaling/opt-13b_eval")
model = AutoModelForCausalLM.from_pretrained("inverse-scaling/opt-13b_eval")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use inverse-scaling/opt-13b_eval with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "inverse-scaling/opt-13b_eval"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "inverse-scaling/opt-13b_eval",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/inverse-scaling/opt-13b_eval

SGLang

How to use inverse-scaling/opt-13b_eval 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 "inverse-scaling/opt-13b_eval" \
    --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": "inverse-scaling/opt-13b_eval",
		"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 "inverse-scaling/opt-13b_eval" \
        --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": "inverse-scaling/opt-13b_eval",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use inverse-scaling/opt-13b_eval with Docker Model Runner:
```
docker model run hf.co/inverse-scaling/opt-13b_eval
```

Add evaluation results on the mathemakitten--winobias_antistereotype_test_cot_v4 config and test split of mathemakitten/winobias_antistereotype_test_cot_v4

#13

by autoevaluator HF Staff - opened Nov 7, 2022

base: refs/heads/main

←

from: refs/pr/13

Discussion Files changed

+17

-0

Files changed (1) hide show

README.md +17 -0

README.md CHANGED Viewed

@@ -111,6 +111,23 @@ model-index:
       type: loss
       value: 1.539870785999474
       verified: true
 ---
 # OPT : Open Pre-trained Transformer Language Models

       type: loss
       value: 1.539870785999474
       verified: true
+  - task:
+      type: zero-shot-classification
+      name: Zero-Shot Text Classification
+    dataset:
+      name: mathemakitten/winobias_antistereotype_test_cot_v4
+      type: mathemakitten/winobias_antistereotype_test_cot_v4
+      config: mathemakitten--winobias_antistereotype_test_cot_v4
+      split: test
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.3131067961165049
+      verified: true
+    - name: Loss
+      type: loss
+      value: 1.4315469591985621
+      verified: true
 ---
 # OPT : Open Pre-trained Transformer Language Models