INDEX

Explanations

commonalities or similarities between different entities

oai_token-act-pair · gpt-3.5-turbo

instances of the word "similarities" and related concepts such as resemblance and comparison

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

til

-0.70

bern

-0.69

gress

-0.65

ved

-0.65

adish

-0.63

Sky

-0.63

 boarding

-0.63

 helicop

-0.62

leanor

-0.60

FT

-0.60

POSITIVE LOGITS

 similarities

1.10

lihood

1.06

 resemblance

0.86

 twins

0.84

 between

0.83

 resemb

0.80

alities

0.80

 similarity

0.79

liness

0.78

alogy

0.78

Activations Density 0.031%

commonalities or similarities between different entities

instances of the word "similarities" and related concepts such as resemblance and comparison

No Comments

No Known Activations