INDEX

Explanations

phrases related to travel and exploration

oai_token-act-pair · gpt-3.5-turbo

references to travel destinations and exploration

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

berman

-0.79

abase

-0.72

ufact

-0.68

imer

-0.67

wake

-0.67

zhen

-0.67

aimon

-0.67

ortium

-0.65

opoly

-0.65

âĢİ

-0.65

POSITIVE LOGITS

å¤§

0.58

 Diabetes

0.55

 Schwarzenegger

0.55

 Isis

0.55

?????-?????-

0.54

 collapsed

0.53

é¾įåĸļå£«

0.52

 Hurricane

0.52

 dynamically

0.51

 scor

0.51

Activations Density 0.138%

phrases related to travel and exploration

references to travel destinations and exploration

No Comments

No Known Activations