INDEX

Explanations

programming-related terms and commands

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

compiling and debugging c++

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

-0.39

 útil

-0.32

 indispensable

-0.31

 ángeles

-0.30

חב

-0.29

 cofre

-0.29

↵

-0.29

 extra

-0.28

 valuable

-0.28

POSITIVE LOGITS

 snippetHide

0.76

parsedMessage

0.75

 مشين

0.75

AutoScaleMode

0.73

sizeCache

0.71

IUrlHelper

0.71

ReusableCell

0.69

endphp

0.69

RTGC

0.68

WebVitals

0.68

Activations Density 0.062%

programming-related terms and commands

compiling and debugging c++

No Comments

No Known Activations