INDEX

Explanations

references to specific names

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "names."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

romy

-0.71

OPLE

-0.70

Bed

-0.68

yrinth

-0.67

irth

-0.65

UGE

-0.64

cffffcc

-0.63

utherford

-0.63

idth

-0.62

ARM

-0.60

POSITIVE LOGITS

paces

1.49

 names

1.24

plates

1.11

names

1.08

 aliases

1.04

pace

1.00

ames

0.94

paced

0.92

named

0.86

 Cosponsors

0.86

Activations Density 0.011%

references to specific names

occurrences of the word "names."

No Comments

No Known Activations