INDEX

Explanations

links or prompts for further reading or action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

)_/¯

-0.61

帖最后由

-0.47

AntiForgeryToken

-0.45

CreateMap

-0.45

onAttach

-0.44

)("

-0.44

//

-0.42

 createSprite

-0.41

thrift

-0.40

invokeLater

-0.40

POSITIVE LOGITS

 estekak

0.59

 שוליים

0.53

:✨

0.48

rungsseite

0.47

 Numerade

0.46

 Autorisations

0.40

 оригіналу

0.40

 Мексичка

0.39

Clik

0.38

BeginInit

0.37

Activations Density 0.067%

links or prompts for further reading or action

No Comments

No Known Activations