INDEX

Explanations

phrases and language indicating conflict or competition

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.04

1:0.04

2:0.05

3:0.08

4:0.06

5:0.05

6:0.05

7:0.43

8:0.03

9:0.03

10:0.05

11:0.04

Negative Logits

 Colleges

-2.61

eneg

-2.56

 Obesity

-2.25

 Glacier

-2.21

gow

-2.19

 Jaguar

-2.15

culosis

-2.13

 Torn

-2.11

vernment

-2.11

gor

-2.10

POSITIVE LOGITS

00200000

3.23

 former

2.66

erville

2.54

之

2.52

dayName

2.40

 charm

2.39

 answ

2.36

Filename

2.34

uitous

2.29

プ

2.26

Activations Density 0.038%

phrases and language indicating conflict or competition

No Comments

No Known Activations