INDEX

Explanations

prepositions indicating location or direction

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "in" referring to various groups or locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ijn

-0.68

utterstock

-0.66

olid

-0.64

erenn

-0.63

fml

-0.63

EngineDebug

-0.63

DragonMagazine

-0.62

icult

-0.62

Va

-0.61

imon

-0.61

POSITIVE LOGITS

 whom

0.74

adish

0.63

 Behind

0.62

Bok

0.62

behind

0.60

 vicinity

0.59

 whose

0.59

İĭ

0.58

whose

0.58

planet

0.57

Activations Density 0.239%

prepositions indicating location or direction

phrases that include the word "in" referring to various groups or locations

No Comments

No Known Activations