INDEX

Explanations

places or locations that are known for specific activities or qualities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "hots" and variations related to heated spots 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 @ 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

ŃĶ

-0.76

yss

-0.74

PRESS

-0.73

xit

-0.65

cision

-0.65

 Yard

-0.63

 Sweeney

-0.62

 restraint

-0.61

UCT

-0.61

ogy

-0.60

POSITIVE LOGITS

pots

1.46

pot

1.28

 hots

1.24

 Hots

1.07

combe

0.97

atile

0.94

hots

0.88

pur

0.87

visor

0.76

lord

0.76

Activations Density 0.008%

places or locations that are known for specific activities or qualities

occurrences of the term "hots" and variations related to heated spots in various contexts

No Comments

No Known Activations