INDEX

Explanations

references to founding figures, organizations, and philosophical concepts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

Not in Any Lists

No Comments

Head Attr Weights

0:0.03

1:0.03

2:0.05

3:0.17

4:0.03

5:0.04

6:0.10

7:0.16

8:0.04

9:0.03

10:0.07

11:0.20

Negative Logits

isSpecialOrderable

-1.31

channelAvailability

-1.29

 unsu

-1.25

chuk

-1.14

xon

-1.10

monitor

-1.09

liga

-1.09

rez

-1.08

umatic

-1.08

 unprepared

-1.08

POSITIVE LOGITS

aughters

1.19

 principles

1.19

 Ideas

1.18

 Founders

1.17

 Peoples

1.17

 founding

1.15

 Principles

1.13

 Concepts

1.12

 Heritage

1.11

FORE

1.07

Activations Density 0.007%

references to founding figures, organizations, and philosophical concepts

No Comments

No Known Activations