INDEX

Explanations

references to specific data structures or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 برانيه

-0.54

 kasarigan

-0.52

 مشين

-0.49

########.

-0.46

 Chwiliwch

-0.46

UIControlState

-0.44

囗

-0.42

ัพท์

-0.41

awtextra

-0.40

 isInitialized

-0.40

POSITIVE LOGITS

 întâ

0.48

actéristi

0.47

 AssemblyCompany

0.46

 okuyayım

0.45

EXHIB

0.44

BibitemShut

0.43

Eugen

0.42

Zá

0.42

AuthContext

0.42

loroethene

0.41

Activations Density 0.315%

references to specific data structures or identifiers

No Comments

No Known Activations