INDEX

Explanations

phrases related to temporal aspects and relationships in discussions

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_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

0:0.04

1:0.03

2:0.07

3:0.05

4:0.06

5:0.03

6:0.42

7:0.06

8:0.04

9:0.03

10:0.07

11:0.05

Negative Logits

️

-1.28

inav

-1.23

 sshd

-1.22

itsch

-1.22

 bleach

-1.18

 upstream

-1.17

lessly

-1.15

 Diesel

-1.14

jab

-1.14

bud

-1.13

POSITIVE LOGITS

urai

1.49

�

1.35

enment

1.35

ndra

1.32

alian

1.32

ylum

1.29

�

1.27

 Poverty

1.24

�

1.24

alia

1.22

Activations Density 0.003%

phrases related to temporal aspects and relationships in discussions

No Comments

No Known Activations