INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

notes

-0.75

 offsets

-0.72

aways

-0.69

 claimant

-0.68

Downloadha

-0.66

 counting

-0.66

 haunt

-0.65

 Definitive

-0.64

 haun

-0.63

votes

-0.63

POSITIVE LOGITS

lance

0.76

ILLE

0.75

rang

0.73

angler

0.65

andel

0.64

urity

0.64

 Brass

0.63

rier

0.63

mal

0.60

rend

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.