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

imar

-0.84

NetMessage

-0.84

pedia

-0.80

DAQ

-0.77

xit

-0.75

Pierre

-0.71

avorite

-0.71

 Farage

-0.69

ItemTracker

-0.68

anyahu

-0.67

POSITIVE LOGITS

ouch

0.69

amn

0.69

ocent

0.69

ains

0.67

 Unch

0.65

bey

0.64

hew

0.62

onne

0.61

hab

0.60

pu

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.