INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

iaz

-0.75

dar

-0.72

mare

-0.71

henko

-0.68

eele

-0.68

inem

-0.67

udeau

-0.67

asin

-0.65

ombs

-0.65

lier

-0.64

POSITIVE LOGITS

signed

0.68

 poppy

0.62

 keynote

0.62

gnu

0.62

 contribut

0.62

',

0.62

 Signed

0.61

 allotted

0.61

 given

0.60

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.