Text Generation
Transformers
PyTorch
Norwegian
Norwegian Bokmål
Norwegian Nynorsk
text2text-generation
T5
NorT5
Norwegian
encoder-decoder
custom_code
Instructions to use ltg/nort5-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ltg/nort5-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="ltg/nort5-base", trust_remote_code=True)# Load model directly from transformers import AutoModelForSeq2SeqLM model = AutoModelForSeq2SeqLM.from_pretrained("ltg/nort5-base", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use ltg/nort5-base with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "ltg/nort5-base" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "ltg/nort5-base", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/ltg/nort5-base
- SGLang
How to use ltg/nort5-base 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 "ltg/nort5-base" \ --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": "ltg/nort5-base", "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 "ltg/nort5-base" \ --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": "ltg/nort5-base", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use ltg/nort5-base with Docker Model Runner:
docker model run hf.co/ltg/nort5-base
Update modeling_nort5.py
Browse files- modeling_nort5.py +2 -2
modeling_nort5.py
CHANGED
|
@@ -134,7 +134,7 @@ class DecoderLayer(nn.Module):
|
|
| 134 |
if past_key_value is not None:
|
| 135 |
self_attn_past_key_value = past_key_value[:2]
|
| 136 |
cross_attn_past_key_value = past_key_value[2:]
|
| 137 |
-
query_offset = self_attn_past_key_value[0].size(
|
| 138 |
else:
|
| 139 |
self_attn_past_key_value, cross_attn_past_key_value = None, None
|
| 140 |
|
|
@@ -570,7 +570,7 @@ class NorT5ForConditionalGeneration(NorT5Model):
|
|
| 570 |
output_hidden_states: Optional[bool] = None,
|
| 571 |
return_dict: Optional[bool] = None,
|
| 572 |
):
|
| 573 |
-
use_cache = use_cache if use_cache is not None else self.config
|
| 574 |
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
|
| 575 |
|
| 576 |
if encoder_outputs is None:
|
|
|
|
| 134 |
if past_key_value is not None:
|
| 135 |
self_attn_past_key_value = past_key_value[:2]
|
| 136 |
cross_attn_past_key_value = past_key_value[2:]
|
| 137 |
+
query_offset = self_attn_past_key_value[0].size(2)
|
| 138 |
else:
|
| 139 |
self_attn_past_key_value, cross_attn_past_key_value = None, None
|
| 140 |
|
|
|
|
| 570 |
output_hidden_states: Optional[bool] = None,
|
| 571 |
return_dict: Optional[bool] = None,
|
| 572 |
):
|
| 573 |
+
use_cache = use_cache if use_cache is not None else getattr(self.config, "use_cache", False)
|
| 574 |
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
|
| 575 |
|
| 576 |
if encoder_outputs is None:
|