INDEX

Explanations

the start of content sections or new paragraphs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 […]

-0.71

 [...]

-0.47

-0.46

[…]

-0.41

่านั้น

-0.39

…

-0.38

็ง

-0.38

-0.37

[...]

-0.37

POSITIVE LOGITS

Personensuche

1.51

 autorytatywna

1.22

 Савезне

1.18

kloped

1.18

tagHelperRunner

1.13

SourceChecksum

1.11

:✨

1.11

InjectAttribute

1.08

 unknownFields

1.06

adaptiveStyles

1.06

Activations Density 0.000%

the start of content sections or new paragraphs

No Comments

No Known Activations