INDEX

Explanations

positive affirmations or statements

oai_token-act-pair · gpt-3.5-turbo

positive evaluations of situations or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

inav

-0.75

 largeDownload

-0.70

choes

-0.70

hend

-0.70

ohl

-0.69

cled

-0.67

hod

-0.66

 sockets

-0.65

ohyd

-0.64

uling

-0.64

POSITIVE LOGITS

 happened

0.97

 happening

0.92

 happens

0.89

 happen

0.88

 happ

0.83

 Happ

0.79

 imaginable

0.78

ifier

0.75

ially

0.75

ional

0.75

Activations Density 0.030%

positive affirmations or statements

positive evaluations of situations or actions

No Comments

No Known Activations