INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

akedown

-0.92

etary

-0.79

earch

-0.74

 bidder

-0.73

ppelin

-0.69

prises

-0.68

phabet

-0.67

vertis

-0.65

--+

-0.65

nery

-0.64

POSITIVE LOGITS

 Cube

0.74

0.70

ECT

0.69

onite

0.67

grain

0.66

 Rift

0.65

Cub

0.65

 Crate

0.64

 Juice

0.64

 Chic

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.