INDEX

Explanations

preferences and choices regarding various topics

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

 avoient

-0.50

 étoient

-0.47

WebElementEntity

-0.43

(".",

-0.40

 auroit

-0.37

 quæ

-0.37

 failures

-0.35

ucket

-0.35

 étoit

-0.35

 textStatus

-0.35

POSITIVE LOGITS

ंदीखरीदारी

0.55

 bParam

0.55

 optional

0.54

kuuta

0.53

 اختی

0.52

 Optional

0.51

好み

0.50

Optional

0.49

 preferencias

0.48

optional

0.48

Activations Density 0.508%

preferences and choices regarding various topics

No Comments

No Known Activations