INDEX

Explanations

mentions of anime

oai_token-act-pair · gpt-3.5-turbo

mentions of anime

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Ö¼

-0.81

LOAD

-0.77

pse

-0.73

lig

-0.67

 Hurricanes

-0.66

asser

-0.66

lain

-0.66

Gro

-0.65

driving

-0.65

tons

-0.64

POSITIVE LOGITS

 anime

1.03

 manga

0.95

 Anime

0.86

avorite

0.85

oka

0.81

 Manga

0.81

emis

0.79

 traged

0.78

umerable

0.78

uno

0.78

Activations Density 0.008%

mentions of anime

mentions of anime

No Comments

No Known Activations