INDEX

Explanations

words related to a specific geographic location: Florida

oai_token-act-pair · gpt-3.5-turbo

references to Florida and related terms

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

 Redemption

-0.83

 buckle

-0.74

CHR

-0.71

 bulls

-0.70

ELL

-0.67

PC

-0.67

 thumbs

-0.64

 stroke

-0.63

 ACTION

-0.63

selves

-0.62

POSITIVE LOGITS

idian

0.93

iculture

0.92

hess

0.88

iane

0.86

asma

0.84

mingham

0.82

rier

0.81

ately

0.81

inate

0.80

idan

0.80

Activations Density 0.010%

words related to a specific geographic location: Florida

references to Florida and related terms

No Comments

No Known Activations