INDEX

Explanations

names of specific places or organizations

oai_token-act-pair · gpt-3.5-turbo

names of entities or proper nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

PDATE

-0.87

PT

-0.74

COM

-0.74

LIN

-0.73

 Polk

-0.71

Ply

-0.68

DET

-0.66

 tune

-0.66

FORMATION

-0.64

TERN

-0.64

POSITIVE LOGITS

1.68

aic

1.15

aq

1.11

aum

1.08

aa

1.00

0.99

as

0.98

amate

0.97

aea

0.97

ģ

0.97

Activations Density 0.103%

names of specific places or organizations

names of entities or proper nouns

No Comments

No Known Activations