Instructions to use oadams/turn-yield-detection with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use oadams/turn-yield-detection with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("unsloth/tinyllama-bnb-4bit") model = PeftModel.from_pretrained(base_model, "oadams/turn-yield-detection") - Notebooks
- Google Colab
- Kaggle
Commit History
-
Upgrading unsloth to 2025-01.
Oliver Adams6b207a5 -
Retraining with unsloth version @July-Llama-2024.
Oliver Adams7955a21
-
Any utterance ending in a filler is now a non-yield.
Oliver Adams77ebd53
-
Updated training data to reflect some issues Lisa found.
Oliver Adamsc55ba65
-
Bugfix in small question training data.
Oliver AdamsPreviously the yield status of these questions was erroneously False. It has now been retrained with them set to True.cb5bc41
-
Added small questions as positive yield examples.
Oliver AdamsSee df41838 in github.com:enfuse/koda_turn_yielding227802b
-
Updated for compatibility with more recent unsloth versions.
Oliver Adams36c704e
-
Upload 7 files
Verified signaturea1016d3 -
initial commit
Verified signature3c1a93c