Text Generation
Transformers
PyTorch
English
experimental
research
bit-level
transformer
reversible
safety
telemetry
language-modeling
Instructions to use WCNegentropy/BitTransformerLM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use WCNegentropy/BitTransformerLM with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="WCNegentropy/BitTransformerLM")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("WCNegentropy/BitTransformerLM", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use WCNegentropy/BitTransformerLM with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "WCNegentropy/BitTransformerLM" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "WCNegentropy/BitTransformerLM", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/WCNegentropy/BitTransformerLM
- SGLang
How to use WCNegentropy/BitTransformerLM 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 "WCNegentropy/BitTransformerLM" \ --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": "WCNegentropy/BitTransformerLM", "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 "WCNegentropy/BitTransformerLM" \ --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": "WCNegentropy/BitTransformerLM", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use WCNegentropy/BitTransformerLM with Docker Model Runner:
docker model run hf.co/WCNegentropy/BitTransformerLM
Remove nested directory: BitTransformerLM/tests/test_quantization.py
Browse files
BitTransformerLM/tests/test_quantization.py
DELETED
|
@@ -1,19 +0,0 @@
|
|
| 1 |
-
import os, sys; sys.path.insert(0, os.path.abspath(os.path.join(os.path.dirname(__file__), "..")))
|
| 2 |
-
import torch
|
| 3 |
-
from bit_transformer import BitTransformerLM, quantize_dynamic, prepare_qat_fx, convert_qat_fx
|
| 4 |
-
from bit_transformer.training import train_loop
|
| 5 |
-
|
| 6 |
-
def test_qat_matches_dynamic_quant():
|
| 7 |
-
data = torch.randint(0, 2, (16, 8), dtype=torch.long)
|
| 8 |
-
base = BitTransformerLM(d_model=16, nhead=4, num_layers=1, dim_feedforward=32, max_seq_len=8)
|
| 9 |
-
train_loop(base, data, epochs=1, log=False)
|
| 10 |
-
dyn = quantize_dynamic(base)
|
| 11 |
-
qat_model = BitTransformerLM(d_model=16, nhead=4, num_layers=1, dim_feedforward=32, max_seq_len=8)
|
| 12 |
-
qat_model.load_state_dict(base.state_dict())
|
| 13 |
-
prepare_qat_fx(qat_model)
|
| 14 |
-
convert_qat_fx(qat_model)
|
| 15 |
-
inp = torch.randint(0, 2, (10, 8), dtype=torch.long)
|
| 16 |
-
out_dyn, _ = dyn(inp)
|
| 17 |
-
out_qat, _ = qat_model(inp)
|
| 18 |
-
diff = (out_dyn - out_qat).abs().max().item()
|
| 19 |
-
assert diff < 0.6
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|