iSarcasm TaskA — LoRA fine-tuning

Best checkpoint (seed = 999) from 5 seed runs on iSarcasm TaskA.

Results on test set (mean ± std across 5 seeds)

Metric Value
Accuracy 77.31 ± 4.25
Precision (macro) 62.51 ± 1.95
Recall (macro) 68.14 ± 0.50
F1 (macro) 63.58 ± 2.69
Precision (sarcastic) 33.42 ± 4.21
Recall (sarcastic) 55.30 ± 5.09
F1 (sarcastic) 41.30 ± 2.19

Seeds used: [2025, 42, 123, 7, 999] Best seed by val F1 macro: 999

Downloads last month
32
Safetensors
Model size
0.4B params
Tensor type
F16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support