INDEX

Explanations

sequences of programming syntax or structure

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

code statements

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

XmlAccessType

-0.49

 DevComponents

-0.49

 plug

-0.44

 Kendall

-0.43

sno

-0.43

 Jude

-0.42

ListModel

-0.42

 push

-0.42

SelectedItem

-0.42

 Forder

-0.41

POSITIVE LOGITS

GOTREF

0.57

InitVars

0.48

 Normdatei

0.47

BeginContext

0.47

RTEX

0.44

 resourceCulture

0.43

RTGC

0.41

Personendaten

0.40

 <<<<<<<<<<<<<<

0.40

 Comprometido

0.40

Activations Density 0.017%

sequences of programming syntax or structure

code statements

No Comments

No Known Activations