INDEX

Explanations

semicolons and parentheses in code-related contexts

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

code statements and comments

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

ksikon

-0.47

Scénario

-0.47

 surla

-0.47

fato

-0.47

elola

-0.47

 Streit

-0.45

amarca

-0.45

nehin

-0.45

 modeled

-0.45

elar

-0.44

POSITIVE LOGITS

InputBorder

0.68

HtmlAttribute

0.50

reportWebVitals

0.49

 invokingState

0.46

ContentAsync

0.43

 HasFactory

0.42

 ModelExpression

0.40

SpringRunner

0.36

\{\\

0.35

 scolas

0.35

Activations Density 0.011%

semicolons and parentheses in code-related contexts

code statements and comments

No Comments

No Known Activations