INDEX

Explanations

phrases related to events occurring repeatedly or multiple times

oai_token-act-pair · gpt-3.5-turbo

mentions of repeated instances or occurrences of events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

bern

-0.70

XT

-0.70

yright

-0.65

DEP

-0.63

OST

-0.62

zbek

-0.61

agra

-0.61

UST

-0.61

talk

-0.61

street

-0.60

POSITIVE LOGITS

 occasions

1.20

cale

0.87

theless

0.85

 occasion

0.78

 periods

0.77

uations

0.76

 experien

0.75

\/\/

0.75

urnal

0.73

 intervals

0.73

Activations Density 0.009%

phrases related to events occurring repeatedly or multiple times

mentions of repeated instances or occurrences of events

No Comments

No Known Activations