INDEX

Explanations

elements related to editorial roles and author attributions

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

 يتيمه

-0.88

 OMITBAD

-0.76

 فريبيس

-0.66

verwijspagina

-0.65

 queſta

-0.62

 للمعارف

-0.60

ロウィン

-0.60

 referenties

-0.57

 propOrder

-0.57

awaiter

-0.56

POSITIVE LOGITS

ViewImports

0.35

 certified

0.30

 licensed

0.30

 established

0.29

seed

0.26

 Taktlose

0.26

dim

0.26

ownik

0.25

WriteTo

0.25

Licensed

0.25

Activations Density 0.145%

elements related to editorial roles and author attributions

No Comments

No Known Activations