INDEX

Explanations

indefinite articles ('an')

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "an."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Frazier

-0.63

çĭ

-0.62

 branches

-0.60

 Bott

-0.56

 certain

-0.55

 given

-0.54

âĸº

-0.54

 McKay

-0.53

 sorts

-0.53

Carter

-0.53

POSITIVE LOGITS

cients

1.08

onym

1.05

notation

1.03

agram

1.00

abolic

0.99

onymous

0.99

ointed

0.97

notations

0.96

iversary

0.93

omaly

0.93

Activations Density 0.057%

indefinite articles ('an')

occurrences of the word "an."

No Comments

No Known Activations