INDEX

Explanations

phrases indicating consideration or acknowledgement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

.*")]

-0.59

 Sitz

-0.58

AndEndTag

-0.55

__':

-0.54

__":

-0.53

 racc

-0.50

 paces

-0.50

 Fass

-0.48

 sinal

-0.48

примеча

-0.48

POSITIVE LOGITS

 regard

1.40

regard

1.17

 regards

1.11

 valued

1.02

 Regard

1.01

 REGARD

1.00

 regarded

0.99

 valuable

0.98

 cherished

0.90

regards

0.90

Activations Density 0.097%

phrases indicating consideration or acknowledgement

No Comments

No Known Activations