INDEX

Explanations

double quotation marks

oai_token-act-pair · gpt-3.5-turbo

quotations and punctuation marks indicating dialogue or speech

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Flavoring

-0.80

 accomp

-0.79

inois

-0.79

 hoard

-0.75

 angered

-0.74

 chair

-0.73

 surpr

-0.72

 behav

-0.72

Klu

-0.71

 agre

-0.71

POSITIVE LOGITS

BuyableInstoreAndOnline

1.08

Hello

1.04

false

1.04

Hey

0.99

normal

0.98

SELECT

0.94

hey

0.94

wcsstore

0.92

true

0.92

evil

0.89

Activations Density 0.168%

double quotation marks

quotations and punctuation marks indicating dialogue or speech

No Comments

No Known Activations