INDEX

Explanations

references to geopolitical conflicts and migration-related issues

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

0:0.03

1:0.03

2:0.06

3:0.04

4:0.05

5:0.03

6:0.41

7:0.04

8:0.03

9:0.03

10:0.12

11:0.08

Negative Logits

itely

-1.49

{"

-1.35

worldly

-1.28

oru

-1.26

jad

-1.24

 soph

-1.20

iquette

-1.20

Zup

-1.18

venth

-1.17

acci

-1.17

POSITIVE LOGITS

raltar

1.25

opa

1.22

orno

1.22

arium

1.20

odor

1.18

ifier

1.16

itives

1.15

urch

1.15

 refuge

1.14

 preservation

1.13

Activations Density 0.018%

references to geopolitical conflicts and migration-related issues

No Comments

No Known Activations