INDEX

Explanations

instances of the word "after" and related phrases indicating subsequent events or actions

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

after something happened

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

RTSC

-0.52

#+#

-0.50

IUrlHelper

-0.48

lename

-0.48

 NavController

-0.45

iecie

-0.45

seam

-0.43

casian

-0.42

BagConstraints

-0.41

iastes

-0.41

POSITIVE LOGITS

 AssemblyCulture

0.47

 što

0.45

 împ

0.43

InteropServices

0.42

 wład

0.41

 poiché

0.40

 ponieważ

0.40

 perquè

0.40

 incidents

0.40

 accidentally

0.40

Activations Density 0.014%

instances of the word "after" and related phrases indicating subsequent events or actions

after something happened

No Comments

No Known Activations