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

minist

-0.76

 dissolved

-0.63

rm

-0.60

 irre

-0.60

phy

-0.59

 unwelcome

-0.59

 merging

-0.59

lvl

-0.58

gently

-0.58

âĹ¼

-0.58

POSITIVE LOGITS

å§«

0.82

0.81

ayn

0.75

ãĥ¼ãĥ³

0.74

 VIDE

0.70

puter

0.66

ï¿½ï¿½

0.64

ptions

0.63

 ILCS

0.63

ä½

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.