INDEX

Explanations

references to organizations and institutions, particularly those related to medical research or dementia

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

Kjelder

-0.57

ituary

-0.57

анд

-0.55

sego

-0.55

findpost

-0.55

tryk

-0.54

Datuak

-0.53

%</

-0.53

月号

-0.52

erg

-0.51

POSITIVE LOGITS

 autorytatywna

0.65

ABR

0.63

 ISNI

0.63

 utafitiHapana

0.61

wurst

0.61

Hentet

0.61

 Brü

0.57

 Broughton

0.57

 knuckles

0.56

iLocal

0.56

Activations Density 0.211%

references to organizations and institutions, particularly those related to medical research or dementia

No Comments

No Known Activations