INDEX

Explanations

mentions of the term "anime."

oai_token-act-pair · gpt-3.5-turbo

terms related to anime and its adaptations

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

stone

-0.76

asser

-0.74

 Rockefeller

-0.72

stones

-0.71

staking

-0.71

lah

-0.71

cair

-0.71

lig

-0.71

riers

-0.71

pse

-0.70

POSITIVE LOGITS

 adaptation

0.91

 anime

0.90

 manga

0.87

emis

0.83

 fandom

0.81

Japan

0.80

 streaming

0.75

 Anime

0.74

 Manga

0.74

 studios

0.74

Activations Density 0.027%

mentions of the term "anime."

terms related to anime and its adaptations

No Comments

No Known Activations