INDEX

Explanations

countries and proper nouns

oai_token-act-pair · gpt-3.5-turbo

references to countries or geographical locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

xual

-0.63

 tremend

-0.60

 citiz

-0.58

 Lauder

-0.57

 prayer

-0.52

ĸļ

-0.50

osexual

-0.47

worldly

-0.47

urga

-0.47

dden

-0.46

POSITIVE LOGITS

thur

0.65

iasco

0.63

SpaceEngineers

0.58

ombat

0.55

emp

0.55

à¦

0.55

Äį

0.53

ourgeois

0.52

ucket

0.52

ĵĺ

0.50

Activations Density 1.073%

countries and proper nouns

references to countries or geographical locations

No Comments

No Known Activations