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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oma

-0.70

zers

-0.70

otherapy

-0.67

ocide

-0.67

 Older

-0.64

 flashback

-0.62

omy

-0.62

orie

-0.61

ograp

-0.61

FAR

-0.60

POSITIVE LOGITS

Lib

0.75

Po

0.75

illions

0.73

Spread

0.71

Bang

0.71

tons

0.67

ãĤ§

0.65

ulner

0.65

Scroll

0.63

stead

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.