INDEX

Explanations

phrases indicating attribution or assigning responsibility

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate causation or blame

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

Dur

-0.82

esson

-0.67

 doesnt

-0.62

 Advertisement

-0.61

ById

-0.60

 Means

-0.60

 nets

-0.59

 didnt

-0.58

 belts

-0.57

SourceFile

-0.57

POSITIVE LOGITS

 blame

1.55

asted

1.43

asting

1.34

ying

1.27

wered

1.25

lled

1.24

iling

1.16

pless

1.13

gg

1.13

iled

1.10

Activations Density 0.081%

phrases indicating attribution or assigning responsibility

phrases that indicate causation or blame

No Comments

No Known Activations