INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ials

-0.77

mong

-0.72

ga

-0.70

ites

-0.70

iaries

-0.70

atorium

-0.69

loo

-0.67

heter

-0.66

soever

-0.66

edin

-0.66

POSITIVE LOGITS

ãĥĥãĥī

0.72

 demol

0.70

WAYS

0.67

Ô

0.67

 MISS

0.66

xit

0.65

ãĥŁ

0.63

Ĭ±

0.63

ãĥį

0.62

¥µ

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.