Instructions to use Salesforce/ctrl with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Salesforce/ctrl with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="Salesforce/ctrl")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Salesforce/ctrl", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use Salesforce/ctrl with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "Salesforce/ctrl" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "Salesforce/ctrl", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/Salesforce/ctrl
- SGLang
How to use Salesforce/ctrl 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 "Salesforce/ctrl" \ --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": "Salesforce/ctrl", "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 "Salesforce/ctrl" \ --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": "Salesforce/ctrl", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use Salesforce/ctrl with Docker Model Runner:
docker model run hf.co/Salesforce/ctrl
Update config.json
Browse files- config.json +27 -0
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"attn_pdrop": 0.1,
|
| 3 |
+
"dff": 8192,
|
| 4 |
+
"embd_pdrop": 0.1,
|
| 5 |
+
"finetuning_task": null,
|
| 6 |
+
"initializer_range": 0.02,
|
| 7 |
+
"layer_norm_epsilon": 1e-06,
|
| 8 |
+
"n_ctx": 512,
|
| 9 |
+
"n_embd": 1280,
|
| 10 |
+
"n_head": 16,
|
| 11 |
+
"n_layer": 48,
|
| 12 |
+
"n_positions": 50000,
|
| 13 |
+
"num_labels": 1,
|
| 14 |
+
"output_attentions": false,
|
| 15 |
+
"output_hidden_states": false,
|
| 16 |
+
"pruned_heads": {},
|
| 17 |
+
"resid_pdrop": 0.1,
|
| 18 |
+
"summary_activation": null,
|
| 19 |
+
"summary_first_dropout": 0.1,
|
| 20 |
+
"summary_proj_to_labels": true,
|
| 21 |
+
"summary_type": "cls_index",
|
| 22 |
+
"summary_use_proj": true,
|
| 23 |
+
"torchscript": false,
|
| 24 |
+
"use_bfloat16": false,
|
| 25 |
+
"vocab_size": 246534,
|
| 26 |
+
"from_tf": false
|
| 27 |
+
}
|