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

jury

-0.69

slot

-0.66

©¶æ

-0.66

usage

-0.66

temp

-0.64

 foreigner

-0.63

CAR

-0.62

HK

-0.60

pkg

-0.59

VIS

-0.59

POSITIVE LOGITS

ymes

0.77

ypes

0.74

apixel

0.71

itars

0.69

jri

0.66

vous

0.66

Gry

0.65

vable

0.65

yrics

0.65

iddles

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.