INDEX

Explanations

phrases related to personal experiences or opinions

oai_token-act-pair · gpt-3.5-turbo

references to personal experiences or events that are significant to the speaker

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

Not in Any Lists

No Comments

Negative Logits

 pockets

-0.76

 Dayton

-0.73

hips

-0.68

 quarters

-0.67

lev

-0.66

 Polk

-0.64

 ears

-0.63

Priv

-0.62

nda

-0.61

itatively

-0.59

POSITIVE LOGITS

self

1.02

alian

1.01

chy

0.99

zbollah

0.96

ueller

0.88

iner

0.88

chwitz

0.87

unes

0.87

ain

0.87

asca

0.86

Activations Density 0.416%

phrases related to personal experiences or opinions

references to personal experiences or events that are significant to the speaker

No Comments

No Known Activations