INDEX

Explanations

references to locations and places

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

locations for activities or purposes

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

Filmography

-0.42

Welche

-0.37

BnF

-0.37

 fatigue

-0.36

 showers

-0.36

 cucchiaio

-0.36

UTIVE

-0.36

clases

-0.35

 Malgré

-0.35

abrigo

-0.35

POSITIVE LOGITS

 تانيه

0.54

DockStyle

0.52

errHandler

0.52

ResponseWriter

0.49

tlere

0.48

 للمعارف

0.47

 estekak

0.47

 poffible

0.47

 BoxDecoration

0.47

ReusableCell

0.46

Activations Density 0.012%

references to locations and places

locations for activities or purposes

No Comments

No Known Activations