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

 Practices

-0.75

Campaign

-0.67

æ©

-0.65

 Insp

-0.65

aughters

-0.63

 Indigenous

-0.62

nings

-0.61

nants

-0.60

Pis

-0.59

BILITIES

-0.59

POSITIVE LOGITS

gob

0.75

aily

0.70

croft

0.68

istor

0.68

icion

0.66

 guiActiveUn

0.64

paper

0.64

emp

0.62

ndra

0.61

razil

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.