INDEX

Explanations

mentions of names starting with "Kat" followed by a combination of different characters

oai_token-act-pair · gpt-3.5-turbo

references to the name "Kat" and its variations

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

 relations

-0.70

 matter

-0.63

cill

-0.62

ENCE

-0.61

 intertw

-0.60

 Relations

-0.60

 contraceptives

-0.59

ENSE

-0.59

GBT

-0.59

cffff

-0.59

POSITIVE LOGITS

rina

1.27

apult

1.08

usha

1.03

anas

1.02

apesh

1.02

mand

0.97

rine

0.96

Kat

0.93

kat

0.92

inka

0.92

Activations Density 0.027%

mentions of names starting with "Kat" followed by a combination of different characters

references to the name "Kat" and its variations

No Comments

No Known Activations