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

Head Attr Weights

0:0.08

1:0.09

2:0.08

3:0.09

4:0.08

5:0.08

6:0.07

7:0.08

8:0.09

9:0.07

10:0.07

11:0.08

Negative Logits

ーテ

-1.66

 Archdemon

-1.63

GH

-1.58

 PLUS

-1.53

 donkey

-1.53

 Haram

-1.49

ゴン

-1.49

 fructose

-1.46

isance

-1.42

 omission

-1.40

POSITIVE LOGITS

sed

2.08

mosp

1.95

onement

1.64

annon

1.58

natureconservancy

1.58

ishes

1.55

vati

1.54

arte

1.53

yrights

1.51

sts

1.49

Activations Density 0.000%

No Known Activations

This feature has no known activations.