INDEX

Explanations

action movies and specific movie titles

oai_token-act-pair · gpt-3.5-turbo

references to notable events or milestones in popular culture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ĸļ

-0.75

annels

-0.66

enhagen

-0.53

otom

-0.51

sbm

-0.51

Opening

-0.49

 Chile

-0.49

PsyNetMessage

-0.48

roach

-0.47

à¨

-0.45

POSITIVE LOGITS

 Gleaming

0.66

 guiActiveUnfocused

0.60

xxxxxxxx

0.55

 bragging

0.54

 embell

0.53

 Furious

0.52

 Trog

0.52

winning

0.50

smanship

0.49

manship

0.49

Activations Density 1.672%

action movies and specific movie titles

references to notable events or milestones in popular culture

No Comments

No Known Activations