INDEX

Explanations

phrases related to past events or history

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to the past

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

starting

-0.65

 Skydragon

-0.62

 thereafter

-0.60

 GENERAL

-0.60

 clock

-0.58

 presently

-0.56

NEY

-0.56

strap

-0.55

 exponentially

-0.54

 tailor

-0.54

POSITIVE LOGITS

ebin

1.72

imes

1.31

ime

1.30

iche

1.29

ures

1.28

oral

1.22

orate

1.17

ries

1.16

illes

1.12

el

1.09

Activations Density 0.028%

phrases related to past events or history

references to the past

No Comments

No Known Activations