INDEX

Explanations

phrases related to actions and their consequences

oai_token-act-pair · gpt-3.5-turbo

phrases related to processes and measurements of sound or distance

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

 Scenes

-0.67

bable

-0.60

 Tactics

-0.57

Reviewer

-0.55

estern

-0.54

Dialogue

-0.52

 ACTIONS

-0.51

estones

-0.51

aughters

-0.50

OIL

-0.49

POSITIVE LOGITS

out

0.54

arat

0.53

of

0.52

aken

0.52

 into

0.50

akening

0.50

uate

0.49

to

0.47

 onto

0.47

 return

0.46

Activations Density 0.587%

phrases related to actions and their consequences

phrases related to processes and measurements of sound or distance

No Comments

No Known Activations