INDEX

Explanations

phrases related to controversial or unethical practices

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.03

1:0.02

2:0.22

3:0.03

4:0.31

5:0.06

6:0.02

7:0.02

8:0.07

9:0.11

10:0.04

11:0.02

Negative Logits

wei

-1.35

away

-1.34

 outlet

-1.21

 fleet

-1.21

 stretch

-1.20

 Stanton

-1.19

plet

-1.18

 screen

-1.15

vertisement

-1.13

qus

-1.13

POSITIVE LOGITS

hement

1.64

 Practices

1.46

pmwiki

1.44

aimon

1.41

hops

1.41

Pract

1.36

ATURE

1.35

METHOD

1.32

OPLE

1.31

Management

1.30

Activations Density 0.005%

phrases related to controversial or unethical practices

No Comments

No Known Activations