INDEX

Explanations

references to societal issues and perceptions

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

0:0.02

1:0.03

2:0.09

3:0.04

4:0.02

5:0.06

6:0.20

7:0.29

8:0.05

9:0.02

10:0.07

11:0.05

Negative Logits

️

-1.48

itone

-1.39

pload

-1.30

heet

-1.28

chan

-1.27

tein

-1.26

vous

-1.21

chini

-1.17

illet

-1.14

RET

-1.13

POSITIVE LOGITS

trust

1.22

 trusts

1.11

��

1.11

 itself

1.07

 built

1.06

 envy

1.06

 ideals

1.06

 republican

1.05

 acad

1.05

 democracies

1.05

Activations Density 0.039%

references to societal issues and perceptions

No Comments

No Known Activations