INDEX

Explanations

names and relationships between characters

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.3.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.02

1:0.08

2:0.08

3:0.06

4:0.02

5:0.08

6:0.06

7:0.08

8:0.19

9:0.10

10:0.08

11:0.09

Negative Logits

iasm

-1.43

ouble

-1.24

leness

-1.22

istries

-1.20

rompt

-1.20

 escalate

-1.19

 hither

-1.17

 adhere

-1.16

iquid

-1.16

 belie

-1.15

POSITIVE LOGITS

 Founding

1.32

woman

1.23

vill

1.23

 WWII

1.14

 ancestor

1.12

 Entity

1.11

├

1.10

 Statue

1.10

アル

1.07

 Fell

1.07

Activations Density 0.008%

names and relationships between characters

No Known Activations