INDEX

Explanations

references to shared entities or concepts across multiple instances

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

Tokens preceding names, titles, or symbols

np_acts-logits-general · gemini-2.0-flash

same name or title

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

 opar

-0.39

∑

-0.36

getMenuInflater

-0.36

Mor

-0.34

ssize

-0.34

Che

-0.33

 voraus

-0.32

UNRELATED

-0.32

En

-0.31

γω

-0.31

POSITIVE LOGITS

 disambiguazione

0.65

WithIOException

0.62

رشف

0.58

homonymie

0.57

 typelib

0.57

 otomatig

0.56

ंदीखरीदारी

0.55

EndGlobalSection

0.54

 виправивши

0.54

AutoField

0.53

Activations Density 0.862%

references to shared entities or concepts across multiple instances

Tokens preceding names, titles, or symbols

same name or title

No Comments

No Known Activations