Instructions to use EleutherAI/pile-t5-large with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use EleutherAI/pile-t5-large with Transformers:

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

pipe = pipeline("text-generation", model="EleutherAI/pile-t5-large")

# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("EleutherAI/pile-t5-large")
model = AutoModelForSeq2SeqLM.from_pretrained("EleutherAI/pile-t5-large", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use EleutherAI/pile-t5-large with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "EleutherAI/pile-t5-large"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EleutherAI/pile-t5-large",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/EleutherAI/pile-t5-large

SGLang

How to use EleutherAI/pile-t5-large 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 "EleutherAI/pile-t5-large" \
    --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": "EleutherAI/pile-t5-large",
		"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 "EleutherAI/pile-t5-large" \
        --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": "EleutherAI/pile-t5-large",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use EleutherAI/pile-t5-large with Docker Model Runner:
```
docker model run hf.co/EleutherAI/pile-t5-large
```

lintang commited on Apr 15, 2024

Commit

d42d873

verified ·

1 Parent(s): fcf253e

Update README.md

Browse files

Files changed (1) hide show

README.md +9 -6

README.md CHANGED Viewed

@@ -6,10 +6,11 @@ language:
 pipeline_tag: text2text-generation
 tags:
 - t5x
-- encode-decoder
 ---
 Pile-T5 Large is an Encoder-Decoder model trained on [the Pile](https://pile.eleuther.ai/) using the [T5x](https://github.com/google-research/t5x) library. The model was trained for 2 million steps or roughly 2 trillion tokens using MLM-objective similar to the original T5 model.
 ### Model Details
@@ -30,7 +31,7 @@ ai](mailto:contact@eleuther.ai).
 | Hyperparameter             | Value       |
 | -------------------------- | ----------- |
-| n<sub>parameters</sub>     |             |
 | n<sub>encoder layers</sub> | 24          |
 | n<sub>decoder layers</sub> | 24          |
 | d<sub>model</sub>          | 2816        |
@@ -133,16 +134,18 @@ checkpoints that can be used for finetuning with the T5x library, refer to [here
 ### Evaluations
-TBD
 ### BibTeX
 ```
-@article{2024t5v2,
   author  = {Lintang Sutawika and Aran Komatsuzaki and Colin Raffel},
-  title   = {Pile T5, an update of T5},
   year    = {2024},
-  url     = {}
 }
 ```

 pipeline_tag: text2text-generation
 tags:
 - t5x
+- encoder-decoder
 ---
 Pile-T5 Large is an Encoder-Decoder model trained on [the Pile](https://pile.eleuther.ai/) using the [T5x](https://github.com/google-research/t5x) library. The model was trained for 2 million steps or roughly 2 trillion tokens using MLM-objective similar to the original T5 model.
+The HF version of Pile-T5 Large borrows UMT5's model implementation as it uses scalable model implementation from T5x and uses `LlamaTokenizer`.
 ### Model Details
 | Hyperparameter             | Value       |
 | -------------------------- | ----------- |
+| n<sub>parameters</sub>     | 783173632   |
 | n<sub>encoder layers</sub> | 24          |
 | n<sub>decoder layers</sub> | 24          |
 | d<sub>model</sub>          | 2816        |
 ### Evaluations
+Pile-T5 Large was evaluated on SuperGLUE, CodeXGLUE. A Flan-finetuned version was evaluated on Flan Held In tasks.
+Results can be seen in the [blogpost](https://blog.eleuther.ai/pile-t5/)
 ### BibTeX
 ```
+@misc{2024PileT5,
   author  = {Lintang Sutawika and Aran Komatsuzaki and Colin Raffel},
+  title   = {Pile-T5},
   year    = {2024},
+  url     = {https://blog.eleuther.ai/pile-t5/},
+  note    = {Blog post},
 }
 ```