INDEX

Explanations

references to brands, products, or specific types of merchandise

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

-0.37

estias

-0.34

Und

-0.33

往

-0.33

dim

-0.32

Dim

-0.32

 хо

-0.32

 Superhosts

-0.32

Ums

-0.31

DIM

-0.31

POSITIVE LOGITS

تقاوى

0.76

 Numerade

0.71

fromnode

0.67

 للمعارف

0.65

Демографія

0.64

tagHelperRunner

0.64

 ſind

0.63

 nonUne

0.62

InjectAttribute

0.60

BibitemShut

0.60

Activations Density 0.159%

references to brands, products, or specific types of merchandise

No Comments

No Known Activations