Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# BLOOM, a version for Petals
|
2 |
+
|
3 |
+
This model is a version of [bigscience/bloom-7b1](https://huggingface.co/bigscience/bloom-7b1)
|
4 |
+
post-processed to be run at home using the [Petals](https://github.com/bigscience-workshop/petals#readme) swarm.
|
5 |
+
|
6 |
+
**Note:** Petals is developed to run 100B+ models
|
7 |
+
like the [full-scale BLOOM](https://huggingface.co/bigscience/bloom-petals) or [BLOOMZ](https://huggingface.co/bigscience/bloomz-petals).
|
8 |
+
This model is provided for testing purposes only. It may be more efficient to run the original version of it locally.
|
9 |
+
|
10 |
+
Please check out:
|
11 |
+
|
12 |
+
- The [original model card](https://huggingface.co/bigscience/bloom-7b1)
|
13 |
+
to learn about the model's capabilities, specifications, and terms of use.
|
14 |
+
- The [Petals repository](https://github.com/bigscience-workshop/petals#readme)
|
15 |
+
to learn how to install Petals and run this model over the Petals swarm.
|
16 |
+
|
17 |
+
We provide minimal code examples below.
|
18 |
+
|
19 |
+
## Using the model
|
20 |
+
|
21 |
+
```python
|
22 |
+
from petals import DistributedBloomForCausalLM
|
23 |
+
|
24 |
+
model = DistributedBloomForCausalLM.from_pretrained("bigscience/bloom-7b1-petals")
|
25 |
+
# Embeddings & prompts are on your device, BLOOM blocks are distributed across the Internet
|
26 |
+
|
27 |
+
inputs = tokenizer("A cat sat", return_tensors="pt")["input_ids"]
|
28 |
+
outputs = model.generate(inputs, max_new_tokens=5)
|
29 |
+
print(tokenizer.decode(outputs[0])) # A cat sat on a mat...
|
30 |
+
```
|
31 |
+
|
32 |
+
## Serving the model blocks
|
33 |
+
|
34 |
+
```bash
|
35 |
+
python -m petals.cli.run_server bigscience/bloom-7b1-petals
|
36 |
+
```
|