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

 Invisible

-0.69

 publication

-0.65

 steroids

-0.65

 publishing

-0.65

 distribut

-0.64

Bott

-0.63

Virtual

-0.62

 contribut

-0.60

 Prof

-0.59

Und

-0.59

POSITIVE LOGITS

iencies

0.83

aru

0.73

orie

0.72

elaide

0.72

 Sabha

0.71

igue

0.71

enary

0.70

issan

0.70

cific

0.70

issions

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.