INDEX

Explanations

occurrences of specific numerical sequences or identifiers

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

David Sinclair's blog posts

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

ReusableCell

-0.65

 للمعارف

-0.60

ésultats

-0.60

########.

-0.60

 JpaRepository

-0.59

فایل‌لار

-0.59

mpagne

-0.59

 تضيفلها

-0.59

جوايز

-0.57

TypedDataSet

-0.57

POSITIVE LOGITS

GMT

0.34

PointerException

0.33

utc

0.33

Â

0.33

GMT

0.33

Com

0.32

xhtml

0.31

zeit

0.31

Flag

0.30

potranspiration

0.30

Activations Density 0.013%

occurrences of specific numerical sequences or identifiers

David Sinclair's blog posts

No Comments

No Known Activations