INDEX

Explanations

the name "Fred"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Fred."

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

BOOK

-0.78

 inhibitors

-0.68

ĵĺ

-0.67

Demand

-0.67

 distingu

-0.65

 compr

-0.64

ModLoader

-0.63

 forth

-0.63

 compounded

-0.62

chnology

-0.62

POSITIVE LOGITS

rik

1.18

erick

1.13

rick

1.12

rique

1.09

dy

1.06

ric

1.03

ricks

1.00

dies

0.99

riks

0.97

die

0.96

Activations Density 0.009%

the name "Fred"

mentions of the name "Fred."

No Comments

No Known Activations