INDEX

Explanations

events, places, and names of specific people in news and sports

oai_token-act-pair · gpt-3.5-turbo

references to major events and figures in sports and politics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

minist

-0.63

xual

-0.63

zsche

-0.59

ratom

-0.59

yss

-0.56

MpServer

-0.55

uscript

-0.54

 glim

-0.51

outine

-0.51

stories

-0.49

POSITIVE LOGITS

.''.

0.61

.).

0.58

.''

0.56

 fame

0.55

''.

0.54

 respectively

0.51

*.

0.50

]."

0.50

).

0.48

."

0.48

Activations Density 1.782%

events, places, and names of specific people in news and sports

references to major events and figures in sports and politics

No Comments

No Known Activations