INDEX

Explanations

programming language syntax elements related to boolean operations and function definitions

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

rendering UI elements

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

ioutil

-0.50

OGND

-0.49

 الحره

-0.49

createTextNode

-0.48

WriteBarrier

-0.45

canestro

-0.44

----</

-0.44

 FontFamily

-0.42

TagMode

-0.41

Tikang

-0.41

POSITIVE LOGITS

homonymie

0.50

+#+#

0.40

IVersion

0.39

 Sünde

0.39

transQ

0.38

BackStack

0.38

 invokingState

0.38

目的

0.37

gtest

0.37

brigens

0.37

Activations Density 0.015%

programming language syntax elements related to boolean operations and function definitions

rendering UI elements

No Comments

No Known Activations