INDEX

Explanations

mentions of the word "Smith"

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ADRA

-0.68

ktop

-0.67

ATING

-0.65

ctory

-0.64

nces

-0.64

Ã§a

-0.62

++++++++

-0.60

 dancer

-0.60

stract

-0.58

UGE

-0.58

POSITIVE LOGITS

sonian

1.75

son

1.02

field

0.94

smanship

0.93

sburg

0.90

 Barney

0.86

gren

0.85

ies

0.82

anity

0.79

ie

0.79

Activations Density 0.032%

mentions of the word "Smith"

No Comments

No Known Activations