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

obin

-0.77

fulness

-0.70

PsyNetMessage

-0.68

nown

-0.67

lez

-0.67

zees

-0.65

pins

-0.63

 mascul

-0.63

forth

-0.63

ãĥ£

-0.60

POSITIVE LOGITS

1.21

iasis

0.65

advertisement

0.65

ota

0.65

tun

0.65

TB

0.65

AG

0.64

peg

0.64

agg

0.64

tif

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.