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

 Soup

-0.74

atown

-0.72

regor

-0.72

luster

-0.72

 Advent

-0.69

é¾įå¥ĳå£«

-0.67

 Chocobo

-0.66

 Clicker

-0.63

eper

-0.63

 Enough

-0.62

POSITIVE LOGITS

emb

0.69

 decom

0.69

oti

0.66

ing

0.64

arov

0.63

mut

0.61

imen

0.60

Attempts

0.60

âĢ¢âĢ¢âĢ¢âĢ¢

0.60

ERY

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.