INDEX

Explanations

mentions of ducks

oai_token-act-pair · gpt-3.5-turbo

mentions of ducks

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 Palestin

-0.83

 occas

-0.77

ccording

-0.76

 agre

-0.72

 conflic

-0.71

Interstitial

-0.70

 condem

-0.70

 exercised

-0.69

 accur

-0.69

 conclud

-0.68

POSITIVE LOGITS

lings

1.19

weed

1.05

 Duck

1.00

 ducks

1.00

tails

0.99

 duck

0.98

bowl

0.95

fish

0.92

 Goose

0.88

lift

0.86

Activations Density 0.007%

mentions of ducks

mentions of ducks

No Known Activations