INDEX

Explanations

terms that indicate strong emotional reactions or criticisms related to moral issues

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

0:0.05

1:0.07

2:0.13

3:0.18

4:0.02

5:0.02

6:0.15

7:0.07

8:0.09

9:0.07

10:0.04

11:0.05

Negative Logits

plugins

-1.28

 condem

-1.18

edIn

-1.17

 slic

-1.11

 restraints

-1.11

fts

-1.11

 maneuvers

-1.08

hement

-1.07

 Tags

-1.05

 printers

-1.04

POSITIVE LOGITS

cffffcc

1.34

 guiActive

1.33

️

1.29

ツ

1.27

 Awakens

1.18

版

1.17

\)

1.17

ヤ

1.16

�

1.15

Lenin

1.13

Activations Density 0.007%

terms that indicate strong emotional reactions or criticisms related to moral issues

No Comments

No Known Activations