INDEX

Explanations

recording or documentation references

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

actions and classifications

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

bcryptjs

-0.57

 onlyOwner

-0.43

msgSender

-0.41

UnsafeEnabled

-0.41

webElementXpaths

-0.41

ItemBackground

-0.40

قایناقلار

-0.40

permitAll

-0.39

 Latter

-0.38

Rptr

-0.37

POSITIVE LOGITS

 your

0.55

 yourself

0.52

chieht

0.52

tentu

0.50

 yourselves

0.48

 Escolar

0.48

 keker

0.48

 táctil

0.48

éez

0.47

 honom

0.47

Activations Density 0.115%

recording or documentation references

actions and classifications

No Comments

No Known Activations