INDEX

Explanations

Java logging framework references

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

programmatic keywords and mathematical notation

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

expandindo

-0.53

 <<<<<<<<<<<<<<

-0.46

WaitGroup

-0.46

 defStyleAttr

-0.43

transQ

-0.40

UserScript

-0.39

TagMode

-0.39

inSlope

-0.39

ành

-0.39

 PrintWriter

-0.38

POSITIVE LOGITS

unhofer

0.49

 المعيارى

0.46

 defaultstate

0.42

ophageal

0.42

휀

0.40

tisan

0.40

∬

0.39

Personensuche

0.39

 للمعارف

0.39

ogaster

0.39

Activations Density 0.000%

Java logging framework references

programmatic keywords and mathematical notation

No Comments

No Known Activations