INDEX

Explanations

conditions or events that may prompt recommendations or actions

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

 tqdm

-0.54

 MainAxisSize

-0.53

RTDA

-0.46

BagConstraints

-0.46

centa

-0.45

 Gari

-0.45

**********/

-0.44

 preven

-0.43

 Stroll

-0.43

UnusedPrivate

-0.42

POSITIVE LOGITS

 zijne

0.59

 feroit

0.56

 then

0.56

 NSCoder

0.52

則

0.51

 allora

0.50

 mijne

0.49

 THEN

0.48

Then

0.48

 powin

0.48

Activations Density 0.421%

conditions or events that may prompt recommendations or actions

No Comments

No Known Activations