INDEX

Explanations

items that represent structured data types or structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 المعيارى

-0.61

ſhip

-0.52

besch

-0.49

hlás

-0.48

 cellpadding

-0.48

atura

-0.48

Descrip

-0.47

PAG

-0.47

 מח

-0.46

 secu

-0.46

POSITIVE LOGITS

item

1.91

 item

1.44

 Item

1.18

Item

1.12

 Items

1.02

 ITEM

1.01

items

0.98

 ITEMS

0.97

ITEM

0.97

 items

0.95

Activations Density 0.011%

items that represent structured data types or structures

No Comments

No Known Activations