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

artifacts

-0.80

robe

-0.65

pha

-0.65

olla

-0.64

)...

-0.63

 beans

-0.62

 artifacts

-0.61

 infield

-0.60

isco

-0.58

LAT

-0.57

POSITIVE LOGITS

Reward

0.83

ãĤ§

0.77

ãĥĺ

0.76

ksh

0.73

Rs

0.72

Bir

0.69

SR

0.66

Rat

0.66

é¾įåĸļå£«

0.66

Measure

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.