Instructions to use FlyinGodzilla/Falcon3-3B-Instruct-Conceptizer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use FlyinGodzilla/Falcon3-3B-Instruct-Conceptizer with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("tiiuae/Falcon3-3B-Instruct") model = PeftModel.from_pretrained(base_model, "FlyinGodzilla/Falcon3-3B-Instruct-Conceptizer") - Notebooks
- Google Colab
- Kaggle
Commit History
Yesterday
-
Model card: link the paper on arXiv
Verified signatureCo-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>7dd4c6d
10 days ago
-
Model card: the paper's result, library versions, where to train your own model
Verified signatureCo-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>5219c0d -
Shorter model card
Verified signatureCo-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>2c73856 -
Add the Conceptizer LoRA adapters (training seeds 0, 37 and 42)
Verified signatureFinal checkpoints (50,000 steps) evaluated in the paper; each adapter byte-identical to the evaluated file. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>9721004 -
initial commit
Verified signature3f659ce