INDEX

Explanations

phrases expressing strong emotional reactions or indifference towards certain topics

oai_token-act-pair · gpt-3.5-turbo

expressions of frustration or indifference

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

orthy

-0.83

obook

-0.76

 kindred

-0.74

NetMessage

-0.67

arcity

-0.63

south

-0.63

inen

-0.63

erville

-0.62

jri

-0.62

MpServer

-0.62

POSITIVE LOGITS

 fuck

0.87

 kidding

0.81

 sorry

0.75

hhh

0.73

 Shit

0.73

amn

0.71

fuck

0.71

 Fuck

0.71

yeah

0.70

hh

0.70

Activations Density 0.341%

phrases expressing strong emotional reactions or indifference towards certain topics

expressions of frustration or indifference

No Comments

No Known Activations