INDEX

Explanations

comparative phrases that highlight differences between groups or scenarios

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

"more...than" comparisons

np_acts-logits-general · gemini-2.0-flash

are, is, do, did, were

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 Verſ

-0.44

 Anſ

-0.40

 Pantai

-0.38

 Reſ

-0.37

eradish

-0.37

 Stadtteil

-0.36

 Dabei

-0.36

SOUNDBITE

-0.36

Seriously

-0.35

seriously

-0.35

POSITIVE LOGITS

 مشين

0.77

IsMutable

0.72

OGND

0.60

SequentialGroup

0.57

WithIOException

0.54

Personendaten

0.54

PreferredItem

0.52

:^(

0.51

 localObject

0.51

complexContent

0.50

Activations Density 0.328%

comparative phrases that highlight differences between groups or scenarios

"more...than" comparisons

are, is, do, did, were

No Comments

No Known Activations