Dialect-tuned LoRA adapter: en-AU seed 2024

This adapter was trained for sarcasm detection on en-AU using TinyLlama/TinyLlama-1.1B-Chat-v1.0.

Selection metric: same-dialect test macro-F1 = 0.7935. Validation macro-F1 at tuned threshold = 0.8402. Decision threshold for yes-minus-no logit scoring = 2.625000.

Downloads last month
6
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for vyshnav112233/task2-3-lora-sarcasm-en-au-seed-2024

Adapter
(1,614)
this model

Spaces using vyshnav112233/task2-3-lora-sarcasm-en-au-seed-2024 2