INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

sbm

-0.72

vg

-0.68

grades

-0.66

zl

-0.65

terday

-0.65

grading

-0.64

CSS

-0.63

SHARE

-0.61

saf

-0.61

leans

-0.61

POSITIVE LOGITS

\\\\\\\\\\\\\\\\

0.65

 Gould

0.63

eless

0.63

 Dragonbound

0.61

idential

0.61

uine

0.60

adel

0.59

 Georg

0.59

inary

0.59

 genuine

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.