INDEX

Explanations

judicial terminology and courtroom proceedings

oai_token-act-pair · gpt-4o-mini Triggered by @bot

judge speaking or ruling

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

reuung

-0.47

hoeddwyd

-0.47

rungsseite

-0.39

 bầu

-0.37

 gentes

-0.36

 onCreateView

-0.36

Personendaten

-0.35

 horizons

-0.35

 angels

-0.35

mmung

-0.34

POSITIVE LOGITS

IntoConstraints

0.61

 CURIAM

0.52

WriteBarrier

0.49

 SwitchCompat

0.47

posedge

0.47

Judge

0.47

 defaultstate

0.46

complexType

0.44

 ſind

0.43

 judge

0.43

Activations Density 0.057%

judicial terminology and courtroom proceedings

judge speaking or ruling

No Comments

No Known Activations