INDEX

Explanations

questions and expressions of curiosity or anticipation about future actions

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

release summary

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

Ӕ

-0.64

WebElementEntity

-0.57

Derbyniad

-0.53

adaptiveStyles

-0.53

neros

-0.51

 betweenstory

-0.51

iſchen

-0.50

béco

-0.49

uxxxx

-0.48

 autorytatywna

-0.47

POSITIVE LOGITS

AddTagHelper

0.37

borders

0.28

ValueStyle

0.28

ViewImports

0.28

astify

0.27

 preghi

0.27

 wachten

0.27

rennen

0.26

czegó

0.26

Erstellt

0.25

Activations Density 0.076%

questions and expressions of curiosity or anticipation about future actions

release summary

No Comments

No Known Activations