INDEX

Explanations

terms related to optimization or improvement

oai_token-act-pair · gpt-3.5-turbo

references to optimization techniques or processes

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

 Spears

-0.64

 specifically

-0.64

 ashore

-0.63

cloth

-0.61

 forcefully

-0.61

UGH

-0.60

 withd

-0.60

maid

-0.59

ously

-0.58

Elk

-0.57

POSITIVE LOGITS

imates

1.68

imal

1.54

imated

1.48

imate

1.40

imum

1.37

ima

1.34

ical

1.31

imation

1.30

imating

1.29

ics

1.21

Activations Density 0.026%

terms related to optimization or improvement

references to optimization techniques or processes

No Comments

No Known Activations