INDEX

Explanations

programming-related syntax and structure

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

keyword followed by comparison

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

Autoritní

-0.69

WithIOException

-0.66

 snippetHide

-0.63

 OMITBAD

-0.60

脚注の使い方

-0.60

abestanden

-0.60

 مرئيه

-0.60

Jeografia

-0.59

 للمعارف

-0.59

endphp

-0.57

POSITIVE LOGITS

is

0.41

TestBed

0.41

之上

0.38

 Consulte

0.35

 mentioned

0.34

 depicted

0.33

 Decke

0.32

 otherwise

0.31

 stated

0.31

之

0.30

Activations Density 0.025%

programming-related syntax and structure

keyword followed by comparison

No Comments

No Known Activations