Individual results
View docsView in-depth performance of a single language model on a single test suite.
Region-by-region surprisal
Sample item for Negative Polarity Licensing (any; with subject relative clause)
The first item of the test suite is shown below for quick reference. Please visit the page for Negative Polarity Licensing (any; with subject relative clause) to see the full list of items.
| Item |
Condition
| Licensor | np | compl | rc_verb | rc_dp | rc_obj | matrix_v | npi | continuation |
|---|---|---|---|---|---|---|---|---|---|---|
| Item | Condition | Licensor | np | compl | rc_verb | rc_dp | rc_obj | matrix_v | npi | continuation |
| 1 | neg_pos | No | author | that | liked | the | senators | has had | any | success |
| 1 | neg_neg | No | author | that | liked | no | senators | has had | any | success |
| 1 | pos_pos | The | author | that | liked | the | senators | has had | any | success |
| 1 | pos_neg | The | author | that | liked | no | senators | has had | any | success |
Showing 1 to 4 of 4 entries
Prediction performance for GPT-2 on Negative Polarity Licensing (any; with subject relative clause)
| Accuracy |
Formula
| Description |
|---|---|---|
| Accuracy | Prediction | Description |
| 60.53% | neg_pos.npi < pos_neg.npi |
No description provided. |
| 92.11% | neg_neg.npi < pos_neg.npi |
No description provided. |
| 97.37% | neg_pos.npi < pos_pos.npi |
No description provided. |
Showing 1 to 3 of 3 entries