INDEX

Explanations

mentions of blood-related actions, such as putting blood on food or taking blood from a patient

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

references to blood and its various contexts, particularly in relation to health and crime

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-att-kk

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

Head Attr Weights

0:0.02

1:0.02

2:0.07

3:0.08

4:0.02

5:0.05

6:0.12

7:0.34

8:0.06

9:0.04

10:0.05

11:0.07

Negative Logits

ventory

-1.36

Eag

-1.35

merce

-1.34

方

-1.29

ultimate

-1.25

hig

-1.24

landish

-1.22

rosso

-1.22

Vers

-1.21

 successors

-1.21

POSITIVE LOGITS

Flu

1.39

 transplant

1.38

 semen

1.38

 kidneys

1.34

 siph

1.30

dye

1.29

 electroly

1.28

 circ

1.25

 blot

1.22

 chlorine

1.22

Activations Density 0.024%

mentions of blood-related actions, such as putting blood on food or taking blood from a patient

references to blood and its various contexts, particularly in relation to health and crime

No Known Activations