INDEX

Explanations

mathematical notation and expressions related to shapes and arrays

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

rows, increasing, diagrams

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

EndContext

-0.46

BeginContext

-0.43

 kaarangay

-0.41

prnewswire

-0.39

gupta

-0.38

ElementException

-0.37

fortawesome

-0.37

 Valera

-0.37

assertRaises

-0.37

mo

-0.35

POSITIVE LOGITS

richTextPanel

0.52

ChildScrollView

0.50

∬

0.50

 незавершена

0.48

AndEndTag

0.47

 horizont

0.45

">—

0.44

 laterally

0.44

 itſelf

0.44

setHorizontal

0.44

Activations Density 0.347%

mathematical notation and expressions related to shapes and arrays

rows, increasing, diagrams

No Comments

No Known Activations