INDEX

Explanations

the word "want" at various levels of intensity

oai_token-act-pair · gpt-3.5-turbo

expressions of desire or intention

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

VERTISEMENT

-0.67

dor

-0.66

grim

-0.66

uum

-0.65

vision

-0.63

ammy

-0.63

 Echoes

-0.63

ricks

-0.62

errors

-0.62

voy

-0.61

POSITIVE LOGITS

urities

0.78

to

0.75

uld

0.66

 mate

0.65

nels

0.64

him

0.63

reprene

0.62

 quotas

0.61

htar

0.61

FontSize

0.61

Activations Density 0.077%

the word "want" at various levels of intensity

expressions of desire or intention

No Comments

No Known Activations