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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Airways

-0.75

 Reloaded

-0.70

 Remastered

-0.69

 istg

-0.67

rogen

-0.67

YY

-0.63

BUS

-0.62

orian

-0.62

Adds

-0.60

ens

-0.60

POSITIVE LOGITS

hab

1.12

aeper

0.78

anamo

0.78

umbn

0.77

keley

0.77

ornia

0.75

encia

0.73

akespe

0.73

oided

0.73

kefeller

0.72

Activations Density 0.000%

No Known Activations

This feature has no known activations.