INDEX

Explanations

phrases related to controversial or sensitive topics, such as LGBTQ+ rights, political issues, and social justice

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to individuals and their personal connections or identities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-att-kk

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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.12

1:0.03

2:0.12

3:0.14

4:0.06

5:0.09

6:0.04

7:0.04

8:0.06

9:0.11

10:0.08

11:0.05

Negative Logits

��

-1.24

etheus

-1.17

 skelet

-1.15

Published

-1.05

 Spiegel

-1.04

isphere

-1.04

PDATE

-1.03

GBT

-1.02

FontSize

-1.00

PsyNetMessage

-1.00

POSITIVE LOGITS

flies

1.21

illin

1.18

 Chicken

1.18

oglu

1.18

 Farms

1.11

 Drops

1.09

ynski

1.08

uan

1.07

ille

1.06

 Slime

1.06

Activations Density 0.036%

phrases related to controversial or sensitive topics, such as LGBTQ+ rights, political issues, and social justice

references to individuals and their personal connections or identities

No Comments

No Known Activations