INDEX

Explanations

instances of operational or functional capabilities and implications in various contexts

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

causes subsequent elaboration

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

 للمعارف

-0.48

IsContent

-0.41

脚注の使い方

-0.38

 لينك

-0.38

'{@

-0.37

invokeLater

-0.37

\{\\

-0.37

getOutputStream

-0.35

IUrlHelper

-0.35

orithm

-0.34

POSITIVE LOGITS

 lenker

0.69

 Biôgrafia

0.56

SharedDtor

0.51

 ejército

0.50

Sklici

0.50

IndentedString

0.49

formKey

0.49

 Numerade

0.47

хьтан

0.47

 rodz

0.46

Activations Density 0.086%

instances of operational or functional capabilities and implications in various contexts

causes subsequent elaboration

No Comments

No Known Activations