INDEX

Explanations

HTML-related tags and attributes

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

Tokens followed by a number

np_acts-logits-general · gemini-2.0-flash

numbers, units, and class terms

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

SharedDtor

-0.72

 ſondern

-0.69

 otomatig

-0.68

 kasarigan

-0.66

iſchen

-0.66

 ſei

-0.63

 queſta

-0.63

 imagui

-0.62

 snippetHide

-0.60

出版年

-0.60

POSITIVE LOGITS

SuppressLint

0.35

 rain

0.33

 efectivamente

0.29

asi

0.29

es

0.29

ój

0.28

lement

0.28

Välislingid

0.28

 Krieges

0.28

中

0.28

Activations Density 1.059%

HTML-related tags and attributes

Tokens followed by a number

numbers, units, and class terms

No Comments

No Known Activations