INDEX

Explanations

words related to direction or orientation

oai_token-act-pair · gpt-3.5-turbo

the word "ward" in various contexts related to directionality

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

ocular

-0.91

acea

-0.84

obook

-0.81

gdala

-0.78

anan

-0.71

tein

-0.70

colo

-0.68

azo

-0.68

coni

-0.66

itu

-0.63

POSITIVE LOGITS

robe

0.99

ward

0.93

ness

0.87

wards

0.82

ly

0.81

nesses

0.79

abouts

0.74

ments

0.72

 Bound

0.71

 Spiral

0.68

Activations Density 0.018%

words related to direction or orientation

the word "ward" in various contexts related to directionality

No Comments

No Known Activations