INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

llor

-0.84

bom

-0.74

ZE

-0.74

zza

-0.73

prus

-0.71

zn

-0.70

WT

-0.68

 Niet

-0.65

idium

-0.64

 wiser

-0.64

POSITIVE LOGITS

function

0.73

hetti

0.72

 offset

0.70

imating

0.67

wcsstore

0.66

imates

0.64

 performances

0.63

imated

0.61

 Integration

0.59

 seasons

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.