INDEX

Explanations

geographic names and locations

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

geographical and administrative place names

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

 soins

-0.43

 preocu

-0.38

舺

-0.38

 AssemblyCulture

-0.37

 Wortes

-0.37

ündigt

-0.37

haikusbot

-0.36

 differently

-0.36

 pulsera

-0.35

 beru

-0.34

POSITIVE LOGITS

DoubleQuotes

0.55

 BrowserModule

0.54

fillType

0.53

setAccessible

0.52

featureID

0.51

 Conſ

0.51

ClientSize

0.50

BrowserModule

0.50

__*/

0.50

]")]

0.49

Activations Density 0.088%

geographic names and locations

geographical and administrative place names

No Comments

No Known Activations