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

rences

-0.77

ļéĨĴ

-0.66

RIC

-0.65

erella

-0.61

lys

-0.61

eal

-0.60

":"/

-0.60

 Precision

-0.59

orter

-0.59

ras

-0.59

POSITIVE LOGITS

PsyNetMessage

0.73

pmwiki

0.72

 indo

0.70

<+

0.69

schild

0.68

sav

0.63

 Entered

0.62

wana

0.61

ãĤ©

0.61

 disappro

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.