INDEX

Explanations

words related to people's names

oai_token-act-pair · gpt-3.5-turbo

the presence of specific names and terms related to brands or businesses

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

ĪĴ

-0.75

 variance

-0.72

dayName

-0.68

Zen

-0.66

 looph

-0.65

BS

-0.65

ĨĴ

-0.65

llah

-0.63

 showc

-0.62

 horm

-0.61

POSITIVE LOGITS

rences

0.84

ocket

0.80

enges

0.79

doms

0.74

velt

0.72

ening

0.72

aiden

0.72

nder

0.71

emort

0.71

hood

0.69

Activations Density 0.074%

words related to people's names

the presence of specific names and terms related to brands or businesses

No Comments

No Known Activations