INDEX

Explanations

elements related to programming structures, data management, and system operations

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

Counts, limits, or numbers, often in code

np_acts-logits-general · gemini-2.0-flash

affected rows

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

zeczytaj

-0.49

kegaard

-0.47

rzez

-0.45

hiya

-0.45

phite

-0.44

:✨

-0.40

 通販

-0.40

ciela

-0.39

heimer

-0.39

zewod

-0.39

POSITIVE LOGITS

 count

1.80

 Count

1.59

 number

1.58

count

1.58

Count

1.56

 COUNT

1.47

 Anzahl

1.47

 numberOf

1.46

 total

1.45

COUNT

1.44

Activations Density 0.691%

elements related to programming structures, data management, and system operations

Counts, limits, or numbers, often in code

affected rows

No Comments

No Known Activations