INDEX

Explanations

references to motorcycles or related products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

énario

-0.78

neté

-0.74

findpost

-0.71

 سكانية

-0.71

Бележки

-0.69

 الحره

-0.64

InjectAttribute

-0.63

lankton

-0.62

erokee

-0.62

 &___

-0.62

POSITIVE LOGITS

 saddle

0.57

 Thorn

0.53

 bike

0.51

 Bike

0.51

DataPropertyName

0.50

ģ

0.50

 Bicycle

0.49

 ridden

0.49

 dera

0.49

 bicycle

0.49

Activations Density 0.087%

references to motorcycles or related products

No Comments

No Known Activations