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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ename

-0.74

ocalypse

-0.70

mination

-0.65

iations

-0.65

arettes

-0.61

 clearance

-0.61

ations

-0.60

ration

-0.60

rollers

-0.59

ingly

-0.59

POSITIVE LOGITS

ERC

0.86

WF

0.85

YC

0.75

ullivan

0.74

ooth

0.72

GN

0.71

Wikipedia

0.69

WP

0.69

@#&

0.68

LI

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.