INDEX

Explanations

statements or descriptions with a strong or impactful tone

oai_token-act-pair · gpt-3.5-turbo

the term "bold" used to emphasize statements, policies, or actions

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

 Cheong

-0.94

OTOS

-0.84

yip

-0.73

nesota

-0.67

enfranch

-0.67

 DISTRICT

-0.67

apolis

-0.65

rogens

-0.64

uters

-0.63

aito

-0.62

POSITIVE LOGITS

faced

1.06

ness

1.03

er

0.98

face

0.97

 bold

0.95

bold

0.93

word

0.84

est

0.82

 Ital

0.79

nesses

0.79

Activations Density 0.026%

statements or descriptions with a strong or impactful tone

the term "bold" used to emphasize statements, policies, or actions

No Comments

No Known Activations