Text Classification
Transformers
Safetensors
modernbert
agent-safety
tool-calling
long-context
distillation
Eval Results (legacy)
text-embeddings-inference
Instructions to use ProCreations/auto-200m-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ProCreations/auto-200m-2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="ProCreations/auto-200m-2")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("ProCreations/auto-200m-2") model = AutoModelForSequenceClassification.from_pretrained("ProCreations/auto-200m-2", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download training/chain.py from ProCreations/auto-200m-2: direct link, hf CLI and curl.
- Browser
- Download file 880 Bytes
-
https://huggingface.co/ProCreations/auto-200m-2/resolve/main/training/chain.py
- Command line
-
hf download hf://ProCreations/auto-200m-2/training/chain.py
-
curl -L -o chain.py https://huggingface.co/ProCreations/auto-200m-2/resolve/main/training/chain.py
880 Bytes
| """Unattended order: download -> prepare -> RoPE sweep/init -> smoke -> reference reproduction -> LR pilots -> p1 -> p2 | |
| -> validation-only selection -> single final evaluation. Every step is idempotent, so a rerun resumes.""" | |
| import subprocess, sys | |
| from common import * | |
| STEPS = [['download.py'], ['prepare.py'], ['rope.py'], ['smoke.py'], ['evaluate.py', 'reference'], ['train.py', 'pilots'], | |
| ['train.py', 'p1_short'], ['train.py', 'p2_long'], ['choose.py'], ['evaluate.py']] | |
| def main(): | |
| stop_after = sys.argv[1] if len(sys.argv) > 1 else None | |
| for step in STEPS: | |
| event('step_started', step=' '.join(step)) | |
| subprocess.run([sys.executable, str(ROOT/step[0]), *step[1:]], check=True, cwd=str(ROOT)) | |
| event('step_complete', step=' '.join(step)) | |
| if stop_after and ' '.join(step) == stop_after: break | |
| if __name__ == '__main__': main() | |