INDEX

Explanations

words related to locations or names

oai_token-act-pair · gpt-3.5-turbo

words related to elements of identity or nationality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

OWS

-0.70

 Ashes

-0.67

Subject

-0.65

drawn

-0.65

 Normandy

-0.64

 ILCS

-0.64

space

-0.64

AAP

-0.63

rences

-0.61

mark

-0.61

POSITIVE LOGITS

zzi

1.33

vernment

1.26

zzo

1.08

oco

1.03

zzle

0.96

onga

0.92

zo

0.90

cephal

0.89

opsis

0.89

berto

0.88

Activations Density 0.055%

words related to locations or names

words related to elements of identity or nationality

No Comments

No Known Activations