INDEX

Explanations

sequences related to locations and events

oai_token-act-pair · gpt-3.5-turbo

numerical data or statistics related to locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Effective

-0.72

 Genie

-0.70

soDeliveryDate

-0.69

ERC

-0.68

 Genius

-0.68

 Cerberus

-0.65

SIGN

-0.64

 Knot

-0.64

LOAD

-0.63

 Serge

-0.62

POSITIVE LOGITS

berra

0.96

oenix

0.94

castle

0.91

elaide

0.89

borough

0.88

 islands

0.86

iverpool

0.86

bane

0.84

gow

0.81

bourne

0.80

Activations Density 0.037%

sequences related to locations and events

numerical data or statistics related to locations

No Comments

No Known Activations