INDEX

Explanations

references to names and identities

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

НИК

-0.58

 Name

-0.52

FieldNumber

-0.51

<?

-0.49

pyx

-0.48

раздо

-0.48

WICK

-0.47

 rede

-0.47

 biens

-0.46

 defaultstate

-0.45

POSITIVE LOGITS

 names

1.83

names

1.50

 Names

1.45

 voices

1.45

 faces

1.41

 NAMES

1.32

 usernames

1.26

 Faces

1.16

 personalities

1.14

Names

1.12

Activations Density 0.090%

references to names and identities

No Comments

No Known Activations