INDEX

Explanations

punctuation marks, specifically parentheses

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

network client errors

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

 kháng

-0.43

-------

-0.42

mazoo

-0.39

tonsoft

-0.38

nmgp

-0.37

bisoft

-0.36

CodedInputStream

-0.36

yyr

-0.35

ielsen

-0.35

Rha

-0.35

POSITIVE LOGITS

VersionUID

0.77

verifyException

0.69

 AssemblyProduct

0.68

WebServlet

0.65

memberof

0.65

 kasarigan

0.63

richTextPanel

0.63

SourceChecksum

0.62

 مرئيه

0.57

enumi

0.55

Activations Density 0.001%

punctuation marks, specifically parentheses

network client errors

No Comments

No Known Activations