INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

HG

-0.69

 Brav

-0.68

utterstock

-0.68

 Ares

-0.64

 Mermaid

-0.60

 Deborah

-0.60

 Byrne

-0.60

 discontinued

-0.59

 Mant

-0.59

DX

-0.58

POSITIVE LOGITS

kefeller

0.70

 Cosponsors

0.66

imil

0.65

otta

0.65

idel

0.63

=]

0.62

ħĭ

0.61

 taunt

0.60

ntil

0.60

ical

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.