INDEX

Explanations

functions and classes related to specific software packages or libraries

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

code file paths

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

 ویکی‌پدیای

-0.50

 pungkas

-0.46

 nakalista

-0.45

 tää

-0.45

 satunya

-0.45

AnchorStyles

-0.44

 kokona

-0.43

 halben

-0.43

 tärke

-0.43

eterangan

-0.42

POSITIVE LOGITS

 intptr

0.72

 FontWeight

0.63

 nonatomic

0.59

 iota

0.57

ISI

0.55

 Corporation

0.55

 Aris

0.55

휀

0.54

 initComponents

0.52

 Hedge

0.52

Activations Density 0.013%

functions and classes related to specific software packages or libraries

code file paths

No Comments

No Known Activations