INDEX

Explanations

dates in various formats

oai_token-act-pair · gpt-3.5-turbo

segments of text related to updates or revisions, particularly those formatted with timestamps

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 mustard

-0.65

 grop

-0.64

 Butcher

-0.63

ļéĨĴ

-0.62

 creep

-0.61

 voic

-0.60

 chants

-0.57

 Gavin

-0.57

 Mellon

-0.56

 Chop

-0.56

POSITIVE LOGITS

Aug

1.36

Feb

1.35

1.30

1.25

1.24

1.23

1.22

Apr

1.21

Activations Density 0.018%

dates in various formats

segments of text related to updates or revisions, particularly those formatted with timestamps

No Comments

No Known Activations