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

athing

-0.82

laughs

-0.64

 Dreams

-0.62

selves

-0.62

 Gaming

-0.61

death

-0.61

verages

-0.61

emo

-0.60

ravis

-0.60

bsite

-0.60

POSITIVE LOGITS

ustomed

0.72

QR

0.68

ãģ®å®

0.65

ãĥĥãĥĪ

0.62

rand

0.62

izons

0.61

ãģ®é

0.61

 tabl

0.61

pter

0.61

canon

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.