INDEX

Explanations

names of sports teams and cities

oai_token-act-pair · gpt-3.5-turbo

references to sports teams and related events

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 goddamn

-0.60

.)

-0.58

 fucking

-0.57

.–

-0.52

 answ

-0.51

"+

-0.51

 accordingly

-0.51

 twelve

-0.50

 Mandatory

-0.50

 fuckin

-0.50

POSITIVE LOGITS

etc

1.03

and

0.97

elsius

0.74

etc

0.73

ect

0.72

ADRA

0.65

or

0.62

merce

0.61

ibia

0.59

berra

0.57

Activations Density 0.473%

names of sports teams and cities

references to sports teams and related events

No Comments

No Known Activations