INDEX

Explanations

instances of significant numbers, particularly in data or research contexts

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

code-related queries

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

-0.51

LookAnd

-0.50

matchCondition

-0.48

 otomatig

-0.48

 étoile

-0.47

 spesa

-0.47

RenderAtEndOf

-0.46

 partenza

-0.45

TextWatcher

-0.45

enterOuterAlt

-0.44

POSITIVE LOGITS

 unsigned

0.37

\{\\

0.37

 News

0.37

PerformLayout

0.36

UnifiedTopology

0.36

revision

0.35

AnimationFrame

0.35

News

0.35

kunst

0.35

ciclop

0.35

Activations Density 0.030%

instances of significant numbers, particularly in data or research contexts

code-related queries

No Comments

No Known Activations