INDEX

Explanations

mentions of people and their perspectives or feelings

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

preceding descriptions of people

np_acts-logits-general · gemini-2.0-flash

people may ask

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

исленность

-0.46

ignoire

-0.45

 surla

-0.43

ifrance

-0.42

 nahilalakip

-0.38

 exigir

-0.38

pgterms

-0.38

 يتيمه

-0.37

łożyć

-0.36

 kurikulum

-0.35

POSITIVE LOGITS

 people

0.73

 peoples

0.61

people

0.59

 PEOPLE

0.57

 mensen

0.55

 людей

0.54

 människor

0.52

enumii

0.52

 الناس

0.52

People

0.51

Activations Density 0.326%

mentions of people and their perspectives or feelings

preceding descriptions of people

people may ask

No Comments

No Known Activations