INDEX

Explanations

text related to user interaction and referencing external resources

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

made decisions

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

<strong>

-0.69

TagMode

-0.62

SequentialGroup

-0.60

Personensuche

-0.57



-0.57

 surla

-0.56

AsUp

-0.56

 EconPapers

-0.55

<b>

-0.55

 defaultstate

-0.54

POSITIVE LOGITS

bibinfo

0.56

uxxxx

0.55

bibfield

0.52

BibitemShut

0.49

ienti

0.45

 melanin

0.44

ільки

0.43

 neuroscience

0.42

 hoax

0.41

omaly

0.41

Activations Density 0.013%

text related to user interaction and referencing external resources

made decisions

No Comments

No Known Activations