INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

++++

-0.68

éĹĺ

-0.63

 cleaners

-0.63

atari

-0.62

trap

-0.61

 pace

-0.60

phony

-0.60

odore

-0.59

insert

-0.59

 cleaner

-0.58

POSITIVE LOGITS

 Prosecut

0.82

uala

0.75

ilyn

0.71

istrates

0.67

ated

0.66

thood

0.66

ities

0.66

UTF

0.65

attribute

0.63

osen

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.