INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Tou

-0.69

recy

-0.66

 Rabb

-0.65

 Crus

-0.64

 Bunny

-0.59

 Chains

-0.58

irmed

-0.58

liness

-0.57

ocalypse

-0.57

iful

-0.56

POSITIVE LOGITS

çīĪ

0.80

jen

0.77

chio

0.74

[&

0.72

ENG

0.72

BuyableInstoreAndOnline

0.70

onian

0.70

>>>>>>>>

0.69

eus

0.67

NPR

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.