INDEX

Explanations

references to the concept of a "pivot" in a technical or programming context

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

center, pivot, origin, offset

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

 snippetHide

-0.63

IGraphics

-0.62

parsedMessage

-0.58

 surla

-0.56

帖最后由

-0.52

 Chriftian

-0.49

 Cæsar

-0.49

 utafitiHapana

-0.49

Hentet

-0.48

 demografica

-0.48

POSITIVE LOGITS

sidemargin

0.46

InsertCommand

0.39

 halfway

0.35

besos

0.35

 polaire

0.35

 comuniques

0.34

resizingMask

0.34

 besos

0.33

 point

0.32

 trasera

0.32

Activations Density 0.010%

references to the concept of a "pivot" in a technical or programming context

center, pivot, origin, offset

No Comments

No Known Activations