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

 Herz

-0.72

helle

-0.71

Reloaded

-0.68

 Mull

-0.67

avascript

-0.65

Marginal

-0.65

 Fenrir

-0.65

 WARN

-0.63

 Witch

-0.62

ISTER

-0.61

POSITIVE LOGITS

rans

0.76

orno

0.73

 Cipher

0.71

enger

0.68

verts

0.67

 bidder

0.66

interstitial

0.65

 reper

0.64

oat

0.63

src

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.