INDEX

Explanations

dates and numbers with an emphasis on scores or quantities

oai_token-act-pair · gpt-3.5-turbo

numerical data related to statistics or votes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

oops

-0.69

idad

-0.67

ibe

-0.66

aido

-0.66

ingen

-0.65

waters

-0.65

aneers

-0.64

ĳå£«

-0.64

istries

-0.64

iaz

-0.62

POSITIVE LOGITS

enda

0.65

0.62

 Colon

0.61

>"

0.61

0.60

 interceptions

0.59

 Grav

0.59

 Honor

0.59

0.58

pts

0.58

Activations Density 0.167%

dates and numbers with an emphasis on scores or quantities

numerical data related to statistics or votes

No Comments

No Known Activations