INDEX

Explanations

words and phrases related to signaling, indication, and relationships

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Efq

-0.44

Personensuche

-0.44

Бахар

-0.41

obiety

-0.40

ighborhood

-0.40

飽

-0.40

FlatAppearance

-0.39

eluarkan

-0.39

 CONTEM

-0.39

cektir

-0.38

POSITIVE LOGITS

 AssemblyTitle

0.50

 propOrder

0.49

 autorytatywna

0.46

>{@

0.44

 فريبيس

0.43

outState

0.43

IndentedString

0.42

meaning

0.42

WaitGroup

0.42

seamnă

0.40

Activations Density 0.066%

words and phrases related to signaling, indication, and relationships

No Comments

No Known Activations