INDEX

Explanations

blocks of code or programming structures indicating significant events or elements in a script

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

programmatic code snippets

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

elemField

-0.55

 autorytatywna

-0.52

rungsseite

-0.52

delwed

-0.50

Datuak

-0.49

ConstraintMaker

-0.49

şört

-0.47

Šaltiniai

-0.47

RuleContext

-0.47

 Theſe

-0.46

POSITIVE LOGITS

 monoxide

0.40

awarkan

0.38

("")]

0.37

urantes

0.37

 abuses

0.36

 Front

0.36

expandindo

0.36

++;

0.36

 /></

0.36

azi

0.36

Activations Density 0.007%

blocks of code or programming structures indicating significant events or elements in a script

programmatic code snippets

No Comments

No Known Activations

blocks of code or programming structures indicating significant events or elements in a script

programmatic code snippets

No Comments

No Known Activations