INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

abulary

-0.75

cano

-0.68

uyomi

-0.67

 Fired

-0.66

©¶æ

-0.62

ricular

-0.59

 Dungeons

-0.59

 precedent

-0.58

 ethics

-0.58

POSITIVE LOGITS

Trend

0.84

TRY

0.82

ificantly

0.75

SL

0.73

OTO

0.72

tackle

0.72

orno

0.70

XT

0.69

ROR

0.69

rums

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.