INDEX

Explanations

quantitative data or numerical values

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

Numbers starting with 1

np_acts-logits-general · gemini-2.0-flash

numbers and years

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

gebob

-0.67

Ecotoxicity

-0.64

<bos>

-0.63

XtraBars

-0.62

 juſ

-0.60

memoized

-0.58

 ſche

-0.57

 itſelf

-0.56

 ***!

-0.56

omock

-0.56

POSITIVE LOGITS

st

0.77

mybatisplus

0.66

ST

0.59

day

0.59

 thing

0.56

st

0.54

 year

0.51

 חיצוניים

0.51

 minute

0.51

 piece

0.50

Activations Density 0.391%

quantitative data or numerical values

Numbers starting with 1

numbers and years

No Comments

No Known Activations