INDEX

Explanations

the word "Wild" with varying degrees of activations

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Wild" and its variations in various contexts

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

 mathemat

-0.88

uyomi

-0.82

akeru

-0.81

uters

-0.80

ayers

-0.77

acists

-0.77

berus

-0.77

adian

-0.76

uckland

-0.74

xus

-0.74

POSITIVE LOGITS

erness

1.07

lings

0.99

flower

0.91

fires

0.89

fire

0.89

Tang

0.85

rose

0.80

 Horses

0.79

tro

0.78

 Wings

0.77

Activations Density 0.007%

the word "Wild" with varying degrees of activations

mentions of the word "Wild" and its variations in various contexts

No Comments

No Known Activations