INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

cit

-0.69

 seller

-0.68

actionDate

-0.66

pts

-0.66

 Britann

-0.63

 Euros

-0.63

 Logged

-0.63

 purchaser

-0.62

 prejudice

-0.62

FG

-0.62

POSITIVE LOGITS

ADA

0.74

emetery

0.70

mass

0.69

vous

0.68

uchi

0.67

docs

0.67

orney

0.66

hest

0.66

urses

0.65

tackle

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.