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

encies

-0.84

ency

-0.70

ilege

-0.70

 minority

-0.68

 Fake

-0.67

 Corpus

-0.65

conserv

-0.64

ĵĺ

-0.64

pmwiki

-0.64

 locality

-0.63

POSITIVE LOGITS

ificial

0.79

cery

0.76

ertodd

0.70

alth

0.70

odes

0.69

icro

0.69

pport

0.64

 Capture

0.64

istries

0.64

ary

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.