INDEX

Explanations

phrases related to particular names

oai_token-act-pair · gpt-3.5-turbo

names of locations or entities related to specific events and individuals

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

ifix

-0.72

risome

-0.68

tex

-0.64

 Archdemon

-0.62

fusc

-0.61

adesh

-0.61

 Rusty

-0.61

 headlights

-0.60

readable

-0.60

worm

-0.59

POSITIVE LOGITS

abet

0.70

ittal

0.66

estic

0.65

clerosis

0.65

uez

0.64

ials

0.64

 Mercy

0.62

Ü

0.61

amate

0.61

 Marketable

0.60

Activations Density 0.180%

phrases related to particular names

names of locations or entities related to specific events and individuals

No Comments

No Known Activations