INDEX

Explanations

geographic locations and their descriptors

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

locations and proximity

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

 escaleras

-0.43

 lentejuelas

-0.43

 acepta

-0.42

 abuelo

-0.41

íně

-0.40

 alles

-0.39

 bailarina

-0.39

 participación

-0.38

 Küchen

-0.38

 seuil

-0.37

POSITIVE LOGITS

 nakalista

0.52

TagMode

0.51

 uParam

0.51

tangentMode

0.48

RTGC

0.47

RectangleBorder

0.45

DockStyle

0.44

ніципалі

0.43

Maj

0.43

 outlying

0.43

Activations Density 0.024%

geographic locations and their descriptors

locations and proximity

No Comments

No Known Activations