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

KEN

-0.76

atform

-0.75

oÄŁ

-0.71

 Mara

-0.69

ï¸ı

-0.69

ebus

-0.66

 Lauder

-0.66

Vie

-0.65

etz

-0.64

Gil

-0.64

POSITIVE LOGITS

nesday

0.75

Sup

0.70

busters

0.69

geist

0.69

izophren

0.69

punk

0.66

riot

0.64

riots

0.63

 pseud

0.62

 parity

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.