INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals, particularly those with the last name "Wen"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 hypoc

-0.68

 offense

-0.66

 Apex

-0.65

 pumps

-0.65

 punishment

-0.65

Cr

-0.64

 Bravo

-0.64

â

-0.63

 spikes

-0.62

 pump

-0.62

POSITIVE LOGITS

wen

4.62

wyn

1.46

wan

1.32

wu

1.31

wy

1.20

Wan

1.17

wer

1.15

wl

1.12

abeth

1.08

wart

1.06

Activations Density 0.011%

proper nouns

mentions of specific individuals, particularly those with the last name "Wen"

No Comments

No Known Activations