INDEX

Explanations

key phrases that indicate the presence of events or actions

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

New Auto-Interp

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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.02

1:0.03

2:0.15

3:0.16

4:0.22

5:0.05

6:0.05

7:0.04

8:0.03

9:0.05

10:0.09

11:0.08

Negative Logits

enta

-1.37

�

-1.36

prus

-1.26

ogun

-1.25

��

-1.25

yt

-1.23

buquerque

-1.16

ixie

-1.16

psons

-1.15

scription

-1.14

POSITIVE LOGITS

 SHALL

1.53

shall

1.50

 shall

1.39

 marked

1.23

 risk

1.19

 locks

1.19

 disqualified

1.17

 contamination

1.17

 undesirable

1.16

 risks

1.16

Activations Density 0.013%

key phrases that indicate the presence of events or actions

No Comments

No Known Activations