INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

IZE

-0.65

FTWARE

-0.64

 predec

-0.63

worms

-0.63

DERR

-0.63

worm

-0.63

iment

-0.62

iments

-0.61

KN

-0.60

XL

-0.60

POSITIVE LOGITS

 ..............

0.75

adj

0.74

course

0.73

thro

0.72

 Thro

0.72

robe

0.72

 ãĤµ

0.71

azard

0.70

 Evening

0.70

 Parade

0.69

Activations Density 0.000%

No Known Activations

This feature has no known activations.