INDEX

Explanations

mentions of specific TV shows and movies

oai_token-act-pair · gpt-3.5-turbo

references to specific media or entertainment content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Blank

-0.48

 hindsight

-0.46

emort

-0.43

itialized

-0.40

ulous

-0.40

etta

-0.39

 complication

-0.38

Loading

-0.38

 depends

-0.38

 derby

-0.37

POSITIVE LOGITS

etheless

0.60

NetMessage

0.58

taboola

0.51

hens

0.49

ederation

0.48

¥ŀ

0.48

onductor

0.48

SourceFile

0.48

nown

0.47

displayText

0.47

Activations Density 5.409%

mentions of specific TV shows and movies

references to specific media or entertainment content

No Comments

No Known Activations