INDEX

Explanations

references to the word "cat"

oai_token-act-pair · gpt-3.5-turbo

references to cats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

mble

-0.76

eous

-0.73

htt

-0.72

undo

-0.71

indal

-0.71

 Sacrament

-0.70

 Seym

-0.69

oppable

-0.65

unda

-0.63

gur

-0.63

POSITIVE LOGITS

aclysm

1.44

heter

1.29

fish

1.05

chers

1.01

alogue

0.97

apult

0.94

cat

0.89

wal

0.89

hawk

0.88

cher

0.87

Activations Density 0.020%

references to the word "cat"

references to cats

No Comments

No Known Activations