INDEX

Explanations

terms related to matrix operations and properties

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

Code, matrix, or numeric formatting

np_acts-logits-general · gemini-2.0-flash

work tapes

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

 '\\;'

-0.63

featureID

-0.63

Vanjske

-0.61

 оригіналу

-0.59

 ProtoMessage

-0.57

remera

-0.56

 estekak

-0.55

 ویکی‌پدی

-0.54

 defStyle

-0.53

makeText

-0.52

POSITIVE LOGITS

 Work

0.55

work

0.54

WORK

0.53

 work

0.53

 WORK

0.52

Work

0.51

 Tapes

0.49

########.

0.49

logical

0.48

 tapes

0.46

Activations Density 0.127%

terms related to matrix operations and properties

Code, matrix, or numeric formatting

work tapes

No Comments

No Known Activations