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

ulp

-0.74

gur

-0.72

wrap

-0.70

shall

-0.66

agate

-0.65

rights

-0.64

asca

-0.63

rue

-0.63

aryn

-0.62

eri

-0.62

POSITIVE LOGITS

OTAL

0.85

çļ

0.85

 thence

0.84

 theirs

0.79

%%%%

0.78

RESULTS

0.77

 THEIR

0.75

iors

0.74

 They

0.72

 Their

0.72

Activations Density 0.000%

No Known Activations

This feature has no known activations.