INDEX

Explanations

abbreviations and acronyms related to organizations or positions

oai_token-act-pair · gpt-3.5-turbo

acronyms or initialisms representing organizations or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 brill

-0.84

EStream

-0.84

 Horus

-0.75

 Incredible

-0.73

quickShipAvailable

-0.71

 Tant

-0.67

 joking

-0.67

 Martian

-0.67

 quantity

-0.66

 Kardashian

-0.66

POSITIVE LOGITS

CS

1.35

CA

1.30

SC

1.28

SF

1.28

PD

1.26

GC

1.24

SA

1.24

WC

1.23

NP

1.23

DC

1.21

Activations Density 0.253%

abbreviations and acronyms related to organizations or positions

acronyms or initialisms representing organizations or entities

No Comments

No Known Activations