INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

igl

-0.78

ylene

-0.73

¶ħ

-0.71

 offic

-0.61

edded

-0.58

ethy

-0.58

é¾

-0.57

ulp

-0.57

hex

-0.57

icist

-0.57

POSITIVE LOGITS

Interstitial

0.89

å§«

0.74

utters

0.73

isson

0.69

nikov

0.69

otos

0.66

ansky

0.65

Ø©

0.62

 Sisters

0.61

 Levant

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.