INDEX

Explanations

words associated with political corruption or racially charged sentiment

oai_token-act-pair · gemini-2.0-flash

Non-English

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ScopeManager

-0.52

Przypisy

-0.48

reht

-0.48

Cringe

-0.48

ferences

-0.47

expandindo

-0.47

Matters

-0.47

dys

-0.47

 Theories

-0.45

erci

-0.45

POSITIVE LOGITS

цездатний

0.66

 <>",

0.65

 Italijanski

0.60

ToBounds

0.60

 Normdatei

0.60

neté

0.60

UrlResolution

0.56

brigens

0.56

évaluateur

0.55

webElementXpaths

0.55

Activations Density 13.933%

words associated with political corruption or racially charged sentiment

Non-English

No Comments

No Known Activations