INDEX

Explanations

sequences of numbers that represent activation values

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

sizeCache

-0.50

Rohy

-0.45

Diweddarwch

-0.43

ToLoad

-0.42

astype

-0.42

bä

-0.41

曖昧さ回避

-0.41

トー

-0.41

 podio

-0.41

 running

-0.40

POSITIVE LOGITS

الدراسه

0.62

washer

0.59

NgModule

0.59

 ब्रेकडाउन

0.58

 ―――

0.56

 digress

0.55

Jeografia

0.55

LElement

0.54

amental

0.54

greja

0.54

Activations Density 0.078%

sequences of numbers that represent activation values

No Comments

No Known Activations