| |
Alpha This system is experimental. Scores and classifications are early-stage research and may be unreliable. Methodology → |
|
| Summary ~lite Article discusses Palisades Fire suspect's use of ChatGPT as evidence.
Lite evaluation by llama-4-scout-wai-psq · editorial channel only · no per-section breakdown available
| |
Longitudinal
· 5 evals | |
Audit Trail
12 entries | 2026-03-05 11:13 | eval_success | PSQ evaluated: g-PSQ=0.120 (3 dims) | - - | | 2026-03-05 11:13 |
eval
|
Evaluated by llama-4-scout-wai-psq: +0.12 (Mild positive) | | | 2026-03-05 11:06 | eval_success | PSQ evaluated: g-PSQ=-0.240 (3 dims) | - - | | 2026-03-05 11:06 |
eval
|
Evaluated by llama-3.3-70b-wai-psq: -0.24 (Mild negative) | | | 2026-02-28 15:03 | eval_success | Lite evaluated: Mild positive (0.10) | - - | | 2026-02-28 15:03 | model_divergence | Cross-model spread 0.40 exceeds threshold (2 models) | - - | | 2026-02-28 15:03 |
eval
|
Evaluated by llama-4-scout-wai: +0.10 (Mild positive) 0.00 | | | reasoning ED, neutral rights stance | | 2026-02-28 14:56 | eval_success | Lite evaluated: Mild positive (0.10) | - - | | 2026-02-28 14:56 |
eval
|
Evaluated by llama-4-scout-wai: +0.10 (Mild positive) | | | reasoning ED, neutral rights stance | | 2026-02-28 14:56 | model_divergence | Cross-model spread 0.40 exceeds threshold (2 models) | - - | | 2026-02-28 14:46 | eval_success | Lite evaluated: Moderate positive (0.50) | - - | | 2026-02-28 14:46 |
eval
|
Evaluated by llama-3.3-70b-wai: +0.50 (Moderate positive) | | | reasoning Investigative journalism on rights abuse | | |
| |