INDEX

Explanations

mentions of specific organizations or companies

oai_token-act-pair · gpt-3.5-turbo

abbreviations or acronyms

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

ish

-0.78

aneously

-0.75

ously

-0.74

aneous

-0.71

istic

-0.70

adesh

-0.69

ries

-0.69

thinkable

-0.67

iser

-0.67

ning

-0.66

POSITIVE LOGITS

ettings

1.24

ocial

1.17

erver

1.11

ometimes

1.06

CRIP

1.03

afety

1.03

uggest

1.03

ync

1.02

ystem

1.02

DAQ

1.01

Activations Density 0.076%

mentions of specific organizations or companies

abbreviations or acronyms

No Comments

No Known Activations