INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Tasman

-0.61

 Bowen

-0.60

 vitamin

-0.60

 signs

-0.59

arling

-0.59

CONT

-0.58

 foul

-0.58

 acid

-0.58

erenn

-0.57

 vitamins

-0.57

POSITIVE LOGITS

mberg

0.77

rera

0.65

ãĥ¼ãĥĨ

0.63

Origin

0.63

learn

0.63

 certific

0.62

apt

0.61

ãĥ¯ãĥ³

0.61

EEE

0.60

":"/

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.