INDEX

Explanations

references to timelines and events related to significant dates or historical contexts

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 ModelExpression

-0.46

providedIn

-0.42

tvguidetime

-0.41

.*")]

-0.41

躇

-0.40

setVerticalGroup

-0.40

 Drapeau

-0.40

IsMutable

-0.39

Lc

-0.39

WriteLiteral

-0.38

POSITIVE LOGITS

 disambiguazione

0.55

AndEndTag

0.51

 Савезне

0.49

 ProtoMessage

0.47

="@+

0.45

 Wikimedijinoj

0.45

 kaarangay

0.42

 незавершена

0.41

 allegedly

0.40

 voer

0.40

Activations Density 0.306%

references to timelines and events related to significant dates or historical contexts

No Comments

No Known Activations