INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

references to individuals and specific names in discussions

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

 contiguous

-0.67

Downloadha

-0.66

è¦ļéĨĴ

-0.66

 continents

-0.65

ppel

-0.64

Ô

-0.62

ragon

-0.62

 Izan

-0.61

 sugg

-0.61

 exacerbate

-0.60

POSITIVE LOGITS

ournal

0.72

Quotes

0.68

rics

0.66

×ľ

0.64

oppers

0.63

andise

0.63

aine

0.61

 Cooke

0.60

Ken

0.59

Kay

0.59

Activations Density 0.195%

names of individuals

references to individuals and specific names in discussions

No Comments

No Known Activations