INDEX

Explanations

punctuation marks and formatting symbols indicative of structured or technical writing

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

parentheses, commas, periods

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

 autorytatywna

-0.56

CopyWith

-0.53

SharedCtor

-0.50



-0.49

 nahilalakip

-0.49

 <>",

-0.48

oa̍t

-0.45

FunctionFlags

-0.45

Portail

-0.43

verwijspagina

-0.43

POSITIVE LOGITS

 writer

0.41

 tiroirs

0.40

無事

0.40

 pilot

0.39

 surla

0.39

 Employee

0.39

 worker

0.38

 Worker

0.38

 frequ

0.38

 committed

0.38

Activations Density 0.285%

punctuation marks and formatting symbols indicative of structured or technical writing

parentheses, commas, periods

No Comments

No Known Activations