INDEX

Explanations

symbols and syntax related to programming or code structure

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

syntax, punctuation, or code structure

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

 فريبيس

-0.78

 كومونز

-0.75

RenderAtEndOf

-0.71

 betweenstory

-0.70

########.

-0.68

addGap

-0.67

MENAFN

-0.62

$_(

-0.60

addPreferredGap

-0.60

 ſind

-0.59

POSITIVE LOGITS

 same

0.43

same

0.41

 mismo

0.37

 mêmes

0.36

 же

0.35

 mesma

0.35

berfläche

0.34

BTW

0.34

btw

0.32

Same

0.32

Activations Density 0.062%

symbols and syntax related to programming or code structure

syntax, punctuation, or code structure

No Comments

No Known Activations