INDEX

Explanations

phrases indicating inappropriate relationships or flirtation

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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.01

1:0.01

2:0.05

3:0.06

4:0.08

5:0.03

6:0.06

7:0.43

8:0.04

9:0.03

10:0.09

11:0.08

Negative Logits

stadt

-1.59

udeb

-1.48

Rated

-1.46

imil

-1.40

dam

-1.38

 stressed

-1.37

 Fukushima

-1.37

ighed

-1.37

example

-1.36

 Auschwitz

-1.35

POSITIVE LOGITS

 pardon

1.85

 brink

1.55

 bandwagon

1.54

 forgiveness

1.52

 renewal

1.51

 flirt

1.48

 sponsorship

1.47

 acceptance

1.46

 solicitation

1.41

appro

1.39

Activations Density 0.001%

phrases indicating inappropriate relationships or flirtation

No Comments

No Known Activations