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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Sight

-0.76

oster

-0.69

Ïī

-0.67

 Prospect

-0.66

 collector

-0.63

 Countdown

-0.63

 liner

-0.62

 IMAGES

-0.58

matic

-0.58

icter

-0.58

POSITIVE LOGITS

ãĥį

0.76

apo

0.76

urb

0.69

unin

0.67

course

0.66

uph

0.66

URA

0.66

meet

0.64

ipal

0.63

romeda

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.