INDEX

Explanations

references to code organization and reuse in programming

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

used or references

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

:✨

-0.43

PyExc

-0.41

 tár

-0.40

 surla

-0.39

DockStyle

-0.37

uttore

-0.35

 huellas

-0.35

 proef

-0.34

MENAFN

-0.34

 مؤرشف

-0.34

POSITIVE LOGITS

GIVEREF

0.50

IGraphics

0.48

 invokingState

0.47

 NUKAT

0.44

Транспорт

0.43

+#+#

0.43

感冒

0.42

 utafitiHapana

0.41

AndEndTag

0.41

EDEFAULT

0.41

Activations Density 0.069%

references to code organization and reuse in programming

used or references

No Comments

No Known Activations