INDEX

Explanations

negative contractions and phrases indicating limitation or disapproval

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 typelib

-0.76

DoubleQuotes

-0.65

:✨

-0.60

sizeCache

-0.59

konomi

-0.58

kháu

-0.58

 SIMBAD

-0.57

那个

-0.57

 astore

-0.55

 الحره

-0.55

POSITIVE LOGITS

ſelf

0.65

 fernández

0.60

 GIPHY

0.60

ſelves

0.59

 ―――――

0.58

 ་་

0.56

 managing

0.56

Portale

0.56

DEG

0.55

 kiin

0.55

Activations Density 0.347%

negative contractions and phrases indicating limitation or disapproval

No Comments

No Known Activations