INDEX

Explanations

phrases relating to specific non-generalized concepts or terms in a technical or medical context

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

non‐, do not, result of, come I, artist has

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

ViewFeatures

-0.58

 AssemblyTitle

-0.52

adaptiveStyles

-0.52

GOTREF

-0.50

windowFixed

-0.50

 Numerade

-0.48

ientras

-0.47

LookAnd

-0.47

 saveiro

-0.47

 <<<<<<<<<<<<<<

-0.46

POSITIVE LOGITS

*/].

0.36

SWR

0.35

ætte

0.35

<?

0.33

 iNdEx

0.32

Rost

0.32

keyCode

0.32

典

0.32

__.__

0.31

Cell

0.31

Activations Density 0.960%

phrases relating to specific non-generalized concepts or terms in a technical or medical context

non‐, do not, result of, come I, artist has

No Comments

No Known Activations