INDEX

Explanations

adjectives describing positive attributes

oai_token-act-pair · gpt-3.5-turbo

words associated with positive attributes and emotional states

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Tanz

-0.68

 Eston

-0.67

 Lans

-0.65

 Edison

-0.63

 Aren

-0.62

 Continental

-0.62

 Reloaded

-0.60

Bed

-0.60

 Finnish

-0.59

 federally

-0.58

POSITIVE LOGITS

fully

1.82

ful

1.78

fulness

1.56

full

1.55

lessly

1.53

less

1.42

ably

1.35

lessness

1.32

abl

1.31

FUL

1.30

Activations Density 0.316%

adjectives describing positive attributes

words associated with positive attributes and emotional states

No Comments

No Known Activations