INDEX

Explanations

prepositions and phrases indicating relationships and connections

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

filtering through membranes

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

 Stande

-0.51

AsUp

-0.48

flashdata

-0.46

PullParser

-0.41

FirstResponder

-0.41

 EVERYTHING

-0.40

 continuidad

-0.39

 Italijani

-0.39

 matchs

-0.39

 confirmer

-0.39

POSITIVE LOGITS

__(/*!

0.63

最快更新

0.58

RenderAtEndOf

0.54

rungsseite

0.53

$_(

0.52

PyExc

0.51

fjspx

0.49

tonode

0.47

 myſelf

0.47

PerformLayout

0.47

Activations Density 0.053%

prepositions and phrases indicating relationships and connections

filtering through membranes

No Comments

No Known Activations