INDEX

Explanations

references to specific geographical locations or cultural landmarks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 purpoſe

-0.68

 Majefty

-0.64

 poffible

-0.54

Efq

-0.53

 Jefus

-0.52

 Perfon

-0.51

 Chriftian

-0.51

 Añade

-0.51

 ویکی‌پدیای

-0.51

 houſe

-0.51

POSITIVE LOGITS

quy

0.64

gia

0.61

CodeAttribute

0.61

Tho

0.59

LookAnd

0.58

 Particularly

0.58

tiendo

0.58

Thu

0.57

 Tien

0.56

Chham

0.56

Activations Density 0.011%

references to specific geographical locations or cultural landmarks

No Comments

No Known Activations