INDEX

Explanations

items that are described with comparisons or similarities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 térmica

-0.40

 équilibr

-0.37

 stabilisation

-0.34

 kości

-0.34

 başar

-0.33

 gewä

-0.32

 thermique

-0.32

 duvar

-0.32

ษัท

-0.32

arrêt

-0.32

POSITIVE LOGITS

 autorytatywna

0.60

 ProtoMessage

0.59

 AssemblyCulture

0.59

like

0.56

ंदीखरीदारी

0.55

 Normdatei

0.55

liken

0.52

ggish

0.52

richTextPanel

0.52

iform

0.51

Activations Density 0.955%

items that are described with comparisons or similarities

No Comments

No Known Activations