INDEX

Explanations

symbols and punctuation typically used in mathematical expressions or programming syntax

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

predicting mathematical definitions

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

Geplaatst

-0.73

AndEndTag

-0.69

expandindo

-0.68

 препратки

-0.66

הערות

-0.66

awtextra

-0.64

angliski

-0.63

 виправивши

-0.63

$_(

-0.61

 مشين

-0.60

POSITIVE LOGITS

IUrlHelper

0.33

NewGuid

0.28

Building

0.28

ôtel

0.27

 asociaciones

0.27

 Building

0.27

 hijas

0.25

 kaik

0.25

 którą

0.24

 aided

0.24

Activations Density 0.019%

symbols and punctuation typically used in mathematical expressions or programming syntax

predicting mathematical definitions

No Known Activations