INDEX

Explanations

words related to actors or characters in the context of movies or media

oai_token-act-pair · gpt-3.5-turbo

recurring mentions of the term "star"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

odcast

-0.83

esville

-0.69

essor

-0.69

apons

-0.68

mble

-0.67

idences

-0.66

Downloadha

-0.64

berra

-0.63

 subp

-0.63

heastern

-0.62

POSITIVE LOGITS

star

1.27

stars

1.06

liner

0.85

stri

0.84

burst

0.82

 stars

0.82

Stars

0.82

vation

0.81

fare

0.77

bucks

0.77

Activations Density 0.006%

words related to actors or characters in the context of movies or media

recurring mentions of the term "star"

No Comments

No Known Activations