INDEX

Explanations

repeated phrases and words related to quantity and measurement

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

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

0:0.02

1:0.01

2:0.07

3:0.06

4:0.35

5:0.03

6:0.18

7:0.09

8:0.03

9:0.04

10:0.05

11:0.04

Negative Logits

 Sanskrit

-1.40

minecraft

-1.35

 Torah

-1.35

 Framework

-1.31

 teachings

-1.30

hari

-1.30

 monog

-1.29

 Warfare

-1.28

 Principles

-1.28

 pronounce

-1.27

POSITIVE LOGITS

urg

1.53

aneously

1.50

 illegally

1.45

oks

1.44

omaly

1.44

undown

1.42

ificantly

1.42

ideo

1.41

needed

1.36

urgical

1.36

Activations Density 0.002%

repeated phrases and words related to quantity and measurement

No Comments

No Known Activations