INDEX

Explanations

references to sexual abuse and assault

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

0:0.07

1:0.03

2:0.07

3:0.24

4:0.05

5:0.05

6:0.06

7:0.10

8:0.03

9:0.05

10:0.11

11:0.07

Negative Logits

SON

-1.31

LESS

-1.24

States

-1.19

arius

-1.18

TRY

-1.18

soType

-1.16

hai

-1.14

ELD

-1.13

angular

-1.13

ciation

-1.09

POSITIVE LOGITS

 abusive

1.24

 harming

1.14

tarian

1.14

 harmful

1.13

 endangered

1.12

 horrend

1.12

 prost

1.12

ogyn

1.11

zbollah

1.10

 torture

1.10

Activations Density 0.010%

references to sexual abuse and assault

No Comments

No Known Activations