INDEX

Explanations

dates

oai_token-act-pair · gpt-3.5-turbo

dates and specific references to time

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

mble

-0.86

anwhile

-0.86

enhagen

-0.71

ä»£

-0.70

BOOK

-0.68

 tray

-0.66

ngth

-0.65

OPS

-0.64

ellation

-0.63

culosis

-0.63

POSITIVE LOGITS

riage

1.02

ried

1.00

athon

0.96

ette

0.96

itime

0.87

Mar

0.87

ital

0.86

cipled

0.83

 Abram

0.80

ced

0.79

Activations Density 0.011%

dates

dates and specific references to time

No Comments

No Known Activations