INDEX

Explanations

the word "stands" often followed by a prepositional phrase or additional context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "stands" used in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ricular

-0.86

ient

-0.75

vertis

-0.70

illian

-0.70

Lear

-0.68

RESULTS

-0.65

è¦ļéĨĴ

-0.63

ffic

-0.63

versely

-0.62

orgetown

-0.62

POSITIVE LOGITS

still

0.94

 stands

0.86

 stand

0.80

 tall

0.79

endor

0.71

 stood

0.70

ands

0.70

hof

0.69

 challeng

0.69

terday

0.69

Activations Density 0.010%

the word "stands" often followed by a prepositional phrase or additional context

instances of the word "stands" used in various contexts

No Comments

No Known Activations