INDEX

Explanations

words indicating action or directive commands

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_L7_Hcat_z_lr1.20e-03_l11.10e+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.7.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

0:0.07

1:0.09

2:0.08

3:0.07

4:0.09

5:0.08

6:0.08

7:0.08

8:0.07

9:0.08

10:0.07

11:0.08

Negative Logits

ribute

-2.75

 concess

-2.66

 answ

-2.62

andem

-2.50

conservancy

-2.43

 advoc

-2.37

 prin

-2.36

plom

-2.28

ipal

-2.27

 rall

-2.25

POSITIVE LOGITS

 ageing

2.32

?,

2.22

 idiots

2.20

00000000

2.17

 assaulting

2.07

 exploding

2.06

 Bald

2.05

 screaming

2.05

 サーティワン

2.04

 Labyrinth

2.01

Activations Density 0.000%

words indicating action or directive commands

No Comments

No Known Activations