INDEX

Explanations

references to measurement units and study details

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

 snippetHide

-0.55

WithIOException

-0.55

RTSC

-0.46

makeText

-0.45

 utafitiHapana

-0.43

 conftest

-0.43

neck

-0.42

 doInBackground

-0.41

posedge

-0.41

 afstand

-0.40

POSITIVE LOGITS

AnchorStyles

0.50

цезда

0.45

 interiors

0.44

gameserver

0.43

égories

0.42

 either

0.41

 Chwiliwch

0.40

Aholisi

0.40

 kaarangay

0.39

上の

0.38

Activations Density 1.150%

references to measurement units and study details

No Comments

No Known Activations