INDEX

Explanations

mentions of a specific topic or concept spread everywhere, especially on social media platforms like Twitter

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

repeated references to places or situations occurring globally or universally

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-att-kk

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

0:0.01

1:0.03

2:0.13

3:0.04

4:0.01

5:0.05

6:0.10

7:0.11

8:0.08

9:0.06

10:0.08

11:0.25

Negative Logits

®

-1.11

kamp

-1.10

Cly

-1.06

arty

-1.04

 Subtle

-1.01

bottom

-1.00

imity

-0.98

hua

-0.97

meier

-0.97

 Archive

-0.96

POSITIVE LOGITS

 imaginable

1.52

 except

1.44

 includ

1.42

ween

1.25

 alike

1.23

inois

1.22

imeters

1.18

 mathemat

1.17

phalt

1.15

 pregn

1.15

Activations Density 0.009%

mentions of a specific topic or concept spread everywhere, especially on social media platforms like Twitter

repeated references to places or situations occurring globally or universally

No Known Activations

mentions of a specific topic or concept spread everywhere, especially on social media platforms like Twitter

repeated references to places or situations occurring globally or universally

No Known Activations