INDEX

Explanations

symbols, punctuation, and formatting elements in the text

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

also plays

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

Negative Logits

❸

-0.44

 convertView

-0.42

 arşivlendi

-0.41

 TestBed

-0.41

WaitGroup

-0.40

Biografia

-0.40

TestingModule

-0.39

 StatelessWidget

-0.39

Weblinks

-0.39

cinogenicity

-0.38

POSITIVE LOGITS

 OMITBAD

0.60

TokenNameLBRACE

0.40

 SwitchCompat

0.40

astify

0.40

 tempor

0.38

AddAttribute

0.38

 المعيارى

0.37

 Савезне

0.36

postsleuth

0.36

//{

0.36

Activations Density 0.038%

symbols, punctuation, and formatting elements in the text

also plays

No Known Activations