INDEX

Explanations

references to the "Star Wars" franchise

oai_token-act-pair · gpt-3.5-turbo

mentions of the "Star" universe franchises, particularly "Star Wars" and "Star Trek"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

utics

-0.79

terday

-0.74

LIA

-0.74

GMT

-0.70

essee

-0.70

ipop

-0.70

Downloadha

-0.69

xual

-0.69

UME

-0.68

inventoryQuantity

-0.68

POSITIVE LOGITS

vation

1.14

ring

1.09

light

1.07

buck

1.07

fish

1.06

 Wars

1.04

burst

1.02

 Trek

1.00

bucks

0.98

fighter

0.92

Activations Density 0.031%

references to the "Star Wars" franchise

mentions of the "Star" universe franchises, particularly "Star Wars" and "Star Trek"

No Comments

No Known Activations