INDEX

Explanations

No Explanations Found

New Auto-Interp

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.08

2:0.09

3:0.08

4:0.08

5:0.07

6:0.07

7:0.06

8:0.08

9:0.08

10:0.08

11:0.08

Negative Logits

 Masquerade

-1.89

 Merchants

-1.87

 Summoner

-1.83

 Rockets

-1.74

 Beaut

-1.74

 McGr

-1.71

 Marines

-1.71

[|

-1.69

 Incarnation

-1.68

xp

-1.68

POSITIVE LOGITS

ril

1.75

ucer

1.74

vati

1.74

stein

1.71

 purportedly

1.65

allel

1.64

atible

1.64

bley

1.60

idan

1.58

 conducting

1.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.