INDEX

Explanations

acronyms related to government bodies or procedures

oai_token-act-pair · gpt-3.5-turbo

abbreviations and acronyms related to political or governmental organizations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

forth

-0.68

given

-0.66

lift

-0.66

king

-0.65

esville

-0.64

face

-0.63

wards

-0.63

bows

-0.63

 priv

-0.63

panic

-0.61

POSITIVE LOGITS

SA

1.13

ACA

1.11

IDS

1.06

AN

1.02

ENA

1.02

MA

1.01

HAEL

1.00

WA

1.00

KA

1.00

AR

1.00

Activations Density 0.123%

acronyms related to government bodies or procedures

abbreviations and acronyms related to political or governmental organizations

No Comments

No Known Activations