INDEX

Explanations

names of people or entities

oai_token-act-pair · gpt-3.5-turbo

the letter 's' 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

ilater

-0.67

ritical

-0.66

SIGN

-0.64

EStream

-0.62

 GOODMAN

-0.62

protect

-0.62

 placebo

-0.59

Compare

-0.58

 multiplier

-0.57

flame

-0.57

POSITIVE LOGITS

ullivan

0.97

aying

0.84

wered

0.83

uez

0.81

ayers

0.79

ays

0.75

kaya

0.75

ued

0.75

 grandson

0.74

terday

0.74

Activations Density 0.206%

names of people or entities

the letter 's' in various contexts

No Comments

No Known Activations