INDEX

Explanations

references to a specific subject or topic

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

"Subject" followed by "matter"

np_acts-logits-general · gemini-2.0-flash

subject matter

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 bianche

-0.58

 Timberlake

-0.52

раздо

-0.51

hysema

-0.50

Noi

-0.49

()");

-0.49

WithIOException

-0.49

✨:

-0.48

íticos

-0.47

😭😭

-0.47

POSITIVE LOGITS

 matter

1.86

matter

1.58

 Matter

1.48

 MATTER

1.43

Matter

1.42

 subject

1.18

 subjects

1.15

 Subject

1.08

setSubject

1.06

subject

1.06

Activations Density 0.102%

references to a specific subject or topic

"Subject" followed by "matter"

subject matter

No Comments

No Known Activations