INDEX

Explanations

text related to explaining, describing, discussing, or covering something in detail

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ief

-0.68

WT

-0.68

Hon

-0.65

Bridge

-0.64

mouth

-0.64

cil

-0.64

Ã±

-0.62

corn

-0.62

achable

-0.62

ometers

-0.60

POSITIVE LOGITS

why

1.20

how

1.18

WHY

0.99

why

0.98

 some

0.93

 aspects

0.92

 similarities

0.89

 examples

0.88

 exactly

0.86

 specifics

0.85

Activations Density 0.197%

text related to explaining, describing, discussing, or covering something in detail

No Comments

No Known Activations