INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ogly

-0.82

hiba

-0.82

licts

-0.79

obook

-0.76

urches

-0.76

irtual

-0.74

arily

-0.73

ounds

-0.72

reetings

-0.72

requently

-0.72

POSITIVE LOGITS

 Rudd

0.81

 Calder

0.72

 siph

0.69

OTE

0.68

Rai

0.66

rost

0.66

 Shinzo

0.66

 Skydragon

0.65

YL

0.63

 melan

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.