INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.07

1:0.07

2:0.08

3:0.07

4:0.08

5:0.08

6:0.07

7:0.08

8:0.07

9:0.08

10:0.10

11:0.08

Negative Logits

 corro

-1.68

 manure

-1.60

umenthal

-1.59

dod

-1.56

urized

-1.40

 pedest

-1.39

 pupp

-1.39

fman

-1.37

 methane

-1.37

 destroyer

-1.36

POSITIVE LOGITS

Overview

1.63

grade

1.62

 Fiscal

1.58

Appearance

1.55

oji

1.54

 Overview

1.53

onomy

1.52

 Actions

1.49

 Chapters

1.49

aders

1.48

Activations Density 0.000%

No Known Activations

This feature has no known activations.