INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³

-0.84

cro

-0.73

Forge

-0.66

Ward

-0.65

 cluster

-0.64

ctica

-0.62

 Harvest

-0.62

ARS

-0.62

MENTS

-0.62

culosis

-0.62

POSITIVE LOGITS

agic

0.91

agos

0.87

addle

0.83

iott

0.79

olved

0.74

ited

0.73

aukee

0.67

 emoji

0.66

olesc

0.65

arded

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.