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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ritch

-0.74

ĵĺ

-0.72

aughed

-0.69

 darkest

-0.69

usky

-0.68

thren

-0.68

olved

-0.66

orney

-0.65

ocry

-0.64

¬¼

-0.64

POSITIVE LOGITS

stores

0.64

 Synd

0.63

flush

0.62

adr

0.62

Les

0.61

breaking

0.61

destroy

0.61

features

0.60

custom

0.60

print

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.