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

abouts

-0.73

VEN

-0.69

rius

-0.67

eur

-0.66

necess

-0.64

 readings

-0.63

dx

-0.62

cellence

-0.62

Ã«

-0.62

dL

-0.61

POSITIVE LOGITS

merce

0.68

iced

0.63

 Mehran

0.63

 skelet

0.61

Tow

0.60

ahi

0.60

 prosecut

0.60

 contrace

0.60

 Somali

0.60

CLASSIFIED

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.