INDEX

Explanations

phrases related to scientific research and its implications

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

not only... but also

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

Попис

-0.59

 InputDecoration

-0.58

InputBorder

-0.57

AndEndTag

-0.56

 Exacts

-0.56

GTCX

-0.53

 lenker

-0.51

Manbalar

-0.49

IBOutlet

-0.48

 otomatig

-0.48

POSITIVE LOGITS

 présidenti

0.41

 merek

0.38

 extinción

0.35

 Jerusalén

0.35

 parezca

0.34

 présidentielle

0.34

 Nusantara

0.33

 certeza

0.33

 Unbe

0.32

Activité

0.32

Activations Density 0.022%

phrases related to scientific research and its implications

not only... but also

No Comments

No Known Activations