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

ilitary

-0.76

tic

-0.72

hift

-0.70

Token

-0.69

PKK

-0.65

 Anarch

-0.65

ado

-0.65

rones

-0.65

igham

-0.64

 Venezuel

-0.64

POSITIVE LOGITS

"></

0.71

 UNIVERS

0.68

 magnet

0.68

 Chevy

0.65

istries

0.63

 Cinem

0.63

Â¯Â¯Â¯Â¯Â¯Â¯Â¯Â¯

0.63

 Mehran

0.60

 Bethesda

0.60

 Cosponsors

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.