INDEX

Explanations

references to network connections and their configurations

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

horseradish, user turns

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

NameInMap

-0.49

Бахар

-0.45

BufferException

-0.43

 متعلقه

-0.43

 المعيارى

-0.43

RenderAtEndOf

-0.41

 variances

-0.41

 generalizations

-0.40

YourGuide

-0.40

 THUR

-0.40

POSITIVE LOGITS

XmlAccessType

0.46

 beginnetje

0.43

 mapStateToProps

0.43

хьтан

0.42

 pernas

0.40

niosek

0.40

breng

0.40

Connection

0.39

 egna

0.38

Personendaten

0.38

Activations Density 0.001%

references to network connections and their configurations

horseradish, user turns

No Comments

No Known Activations