INDEX

Explanations

academic language indicating the continuation or progression of a research paper

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

introductions or sections

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

FunctionFlags

-0.45

WebElementEntity

-0.43

dération

-0.42

󠁣

-0.42

aarrggbb

-0.40

 fluor

-0.40

 mandatario

-0.39

 AssemblyProduct

-0.39

 Eval

-0.38

multirow

-0.38

POSITIVE LOGITS

setVerticalGroup

0.59

pository

0.36

实现了

0.36

 sufic

0.35

 обсу

0.35

Espèce

0.34

UIManager

0.34

extension

0.34

sufficient

0.34

 söyledi

0.33

Activations Density 0.341%

academic language indicating the continuation or progression of a research paper

introductions or sections

No Comments

No Known Activations