INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

2:0.08

3:0.08

4:0.06

5:0.07

6:0.08

7:0.07

8:0.08

9:0.07

10:0.08

11:0.09

Negative Logits

York

-1.38

tro

-1.35

Loading

-1.28

Veter

-1.26

Tact

-1.26

Op

-1.25

Combat

-1.25

alore

-1.24

Dead

-1.24

Tro

-1.23

POSITIVE LOGITS

ertodd

1.69

 Disclosure

1.51

��

1.47

paio

1.46

etheless

1.45

��

1.32

 Lieberman

1.32

 Consumers

1.32

 cannibal

1.30

 indemn

1.28

Activations Density 0.000%

No Known Activations

This feature has no known activations.