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

uala

-0.85

emale

-0.72

alo

-0.70

ulty

-0.67

kinson

-0.67

ulk

-0.66

ither

-0.66

uming

-0.65

 pearl

-0.65

ETHOD

-0.64

POSITIVE LOGITS

 Wilde

0.75

 Kers

0.73

 deficits

0.63

 Hunts

0.60

 Benson

0.60

hare

0.60

Ø§ÙĦ

0.59

 Held

0.59

wer

0.59

Ub

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.