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

hesis

-0.81

 Manit

-0.72

neau

-0.70

 bothering

-0.63

heses

-0.63

 occurring

-0.63

hips

-0.61

hig

-0.61

 Koen

-0.60

nant

-0.60

POSITIVE LOGITS

iolet

0.77

SPA

0.68

0.65

RFC

0.65

 Harbour

0.63

VM

0.62

forestation

0.62

vill

0.62

âĶĢâĶĢâĶĢâĶĢ

0.62

fman

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.