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

 annex

-0.68

yip

-0.65

Lenin

-0.63

Aren

-0.63

 Cruise

-0.63

Iraq

-0.62

gie

-0.60

igen

-0.60

 cruise

-0.59

iHUD

-0.58

POSITIVE LOGITS

eport

0.83

 irony

0.82

 understatement

0.65

tha

0.65

olulu

0.63

urous

0.62

 Thro

0.62

lear

0.61

idence

0.60

ilation

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.