INDEX

Explanations

proper nouns or entities such as names of countries, companies, and people, especially focusing on partnerships

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

key indicators of institutional presence and activity in various sectors

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

0:0.03

1:0.16

2:0.11

3:0.02

4:0.02

5:0.03

6:0.07

7:0.05

8:0.28

9:0.07

10:0.05

11:0.06

Negative Logits

xual

-1.23

emic

-1.21

eworthy

-1.21

auri

-1.21

eneg

-1.19

.?

-1.14

plom

-1.14

‎

-1.12

razen

-1.11

urrency

-1.10

POSITIVE LOGITS

 progresses

1.41

Pie

1.21

 unfolds

1.18

jit

1.17

Lie

1.16

["

1.15

 Hunt

1.15

 enters

1.13

 tries

1.13

龍�

1.11

Activations Density 0.093%

proper nouns or entities such as names of countries, companies, and people, especially focusing on partnerships

key indicators of institutional presence and activity in various sectors

No Known Activations