INDEX

Explanations

words related to alignment and coordination

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_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

Not in Any Lists

No Comments

Head Attr Weights

0:0.03

1:0.03

2:0.08

3:0.23

4:0.03

5:0.03

6:0.08

7:0.15

8:0.05

9:0.09

10:0.06

11:0.09

Negative Logits

��

-1.24

irin

-1.16

hops

-1.15

iatric

-1.12

kids

-1.12

Pod

-1.10

hift

-1.10

laugh

-1.10

Dad

-1.09

brew

-1.09

POSITIVE LOGITS

ativity

1.24

lishes

1.20

aign

1.20

 Illum

1.19

 srfAttach

1.13

ーテ

1.12

ement

1.11

pole

1.09

 Fedora

1.06

 disparate

1.06

Activations Density 0.002%

words related to alignment and coordination

No Comments

No Known Activations