INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

chy

-0.80

 exerc

-0.74

icent

-0.72

 helicop

-0.66

 reapp

-0.66

important

-0.66

aton

-0.65

 facult

-0.64

 sponge

-0.64

 besie

-0.64

POSITIVE LOGITS

edIn

0.75

Favorite

0.72

lished

0.72

KK

0.70

ractor

0.65

 Nadu

0.65

Tal

0.64

Ins

0.64

umers

0.62

 Handler

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.