INDEX

Explanations

phrases related to performing actions or giving instructions

oai_token-act-pair · gpt-3.5-turbo

phrases describing ongoing actions involving the second person perspective

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

ICES

-0.85

MpServer

-0.82

Zup

-0.79

 Cosponsors

-0.73

bably

-0.67

ILS

-0.65

 Lans

-0.65

onite

-0.64

IFIED

-0.64

ethy

-0.64

POSITIVE LOGITS

 distracted

0.78

 bask

0.72

 vacation

0.72

 touring

0.72

 researching

0.71

 immersed

0.70

 browsing

0.69

ggle

0.67

 filming

0.67

 idle

0.67

Activations Density 0.138%

phrases related to performing actions or giving instructions

phrases describing ongoing actions involving the second person perspective

No Comments

No Known Activations