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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

icult

-0.84

fif

-0.80

lace

-0.75

oday

-0.73

IPS

-0.72

onom

-0.72

tsky

-0.72

put

-0.70

ke

-0.69

hend

-0.69

POSITIVE LOGITS

 scrut

0.86

 transformer

0.72

 reconc

0.71

 leasing

0.71

 lease

0.71

srf

0.69

 confir

0.65

 disabling

0.65

Sac

0.64

 freight

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.