INDEX

Explanations

phrases related to advertisements

oai_token-act-pair · gpt-3.5-turbo

instances of category labels for Pokémon types or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

forth

-0.80

 Nieto

-0.63

 beforehand

-0.63

enegger

-0.60

sic

-0.59

istically

-0.55

selves

-0.55

Leaks

-0.54

naires

-0.54

 regard

-0.54

POSITIVE LOGITS

BuyableInstoreAndOnline

0.85

thumbnails

0.78

 guiActive

0.72

 guiActiveUnfocused

0.72

 Featured

0.71

cloneembedreportprint

0.69

taboola

0.67

 thumbnail

0.66

 Manga

0.64

malink

0.64

Activations Density 0.534%

phrases related to advertisements

instances of category labels for Pokémon types or characteristics

No Comments

No Known Activations