INDEX

Explanations

No Explanations Found

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

ashington

-0.69

assetsadobe

-0.69

 Congress

-0.68

 pull

-0.65

western

-0.65

iss

-0.62

 Seeking

-0.62

 rust

-0.62

 Ahead

-0.60

cheat

-0.60

POSITIVE LOGITS

%%%%

0.80

EVA

0.77

lar

0.72

ament

0.70

aments

0.68

ricular

0.68

iatures

0.68

ties

0.68

hetic

0.66

rets

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.