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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 onward

-0.98

 onwards

-0.93

 spoiled

-0.71

abilia

-0.68

BILITIES

-0.68

za

-0.66

ostics

-0.63

 cease

-0.63

alys

-0.61

dden

-0.60

POSITIVE LOGITS

 territ

0.79

otom

0.69

otomy

0.68

 vulner

0.68

 decap

0.67

peror

0.65

 seiz

0.62

Chuck

0.62

soDeliveryDate

0.61

 Purg

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.