INDEX

Explanations

text indicating significant changes or variations, often represented by highlighting or specific formatting

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

changes or significance

np_acts-logits-general · gemini-2.0-flash

explains symbols and changes

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

сылкі

-0.41

 Weiner

-0.37

Incident

-0.37

Obrázky

-0.36

ktop

-0.36

Secondly

-0.36

 Secondly

-0.35

 УК

-0.35

AccessorTable

-0.35

 igen

-0.35

POSITIVE LOGITS

 marked

0.57

0.52

istoitu

0.51

marked

0.48

 symbols

0.48

Autoritní

0.47

Ծանոթ

0.46

adaptiveStyles

0.46

互联网档案馆

0.46

 denote

0.45

Activations Density 0.706%

text indicating significant changes or variations, often represented by highlighting or specific formatting

changes or significance

explains symbols and changes

No Comments

No Known Activations