INDEX

Explanations

programming constructs and syntax elements

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

code structure and foreign languages

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

bootstrapcdn

-0.72

oa̍t

-0.66

majánló

-0.65

rungsseite

-0.65

 autorytatywna

-0.63

 kaarangay

-0.63

 pinulongan

-0.62

 StringTokenizer

-0.62

IUrlHelper

-0.61

ðsíða

-0.60

POSITIVE LOGITS

 useEffect

0.49

useEffect

0.46

 updates

0.33

TRAILING

0.32

 Erişim

0.32

 plaisir

0.32

 czeka

0.31

 actualidad

0.30

 conduite

0.30

先

0.29

Activations Density 0.002%

programming constructs and syntax elements

code structure and foreign languages

No Comments

No Known Activations