INDEX

Explanations

phrases related to marking events or actions

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

marking or labeling

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

+#+#

-0.60

nyata

-0.57

Beskrivning

-0.53

CWE

-0.52

CollectionUtils

-0.51

ChildScrollView

-0.50

Copia

-0.50

 copertura

-0.49

 AllAfrica

-0.49

Referências

-0.48

POSITIVE LOGITS

 marks

1.30

 mark

1.23

 marking

1.13

 marked

1.13

 MARKS

1.09

 Marks

1.09

 Marking

1.08

marks

1.03

marking

1.02

MARKS

1.01

Activations Density 0.109%

phrases related to marking events or actions

marking or labeling

No Comments

No Known Activations