INDEX

Explanations

phrases indicating denial or negation

oai_token-act-pair · gpt-3.5-turbo

strong negations and emphatic denials

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apters

-0.74

ariat

-0.72

orio

-0.68

ffen

-0.67

urers

-0.66

lahoma

-0.64

urer

-0.64

iary

-0.63

rooms

-0.63

rors

-0.63

POSITIVE LOGITS

LY

1.56

ALLY

1.54

ELY

1.50

THING

1.48

WHERE

1.44

ONE

1.44

HO

1.44

 THERE

1.41

ING

1.38

LESS

1.38

Activations Density 0.111%

phrases indicating denial or negation

strong negations and emphatic denials

No Comments

No Known Activations