INDEX

Explanations

statistics and measurements related to various topics, such as health care, economic conditions, energy production, and political support

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

statistical and numerical information related to social, economic, and political topics

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

quantitative data related to years, percentages, and financial figures

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.03

2:0.19

3:0.06

4:0.06

5:0.15

6:0.04

7:0.04

8:0.04

9:0.09

10:0.11

11:0.05

Negative Logits

ón

-1.23

iquette

-1.21

own

-1.17

 parted

-1.11

icent

-1.09

essor

-1.08

 Friedrich

-1.08

olor

-1.08

URI

-1.07

ptroller

-1.06

POSITIVE LOGITS

 crore

1.50

Ago

1.16

million

1.12

 Scotch

1.08

chenko

1.07

igslist

1.07

 disposable

1.06

 Sector

1.06

Gry

1.03

abytes

1.03

Activations Density 0.098%

statistics and measurements related to various topics, such as health care, economic conditions, energy production, and political support

statistical and numerical information related to social, economic, and political topics

quantitative data related to years, percentages, and financial figures

No Comments

No Known Activations