INDEX

Explanations

references to significant historical figures and their contributions or achievements

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

0:0.03

1:0.07

2:0.10

3:0.02

4:0.02

5:0.03

6:0.07

7:0.06

8:0.26

9:0.11

10:0.06

11:0.13

Negative Logits

 confir

-1.62

 myself

-1.27

 attest

-1.22

obin

-1.22

itaire

-1.21

 sugg

-1.20

estyles

-1.11

uner

-1.11

 confess

-1.11

)?

-1.10

POSITIVE LOGITS

CVE

1.32

 Armenian

1.21

 Cosponsors

1.20

ould

1.18

NCT

1.15

leans

1.12

 deserved

1.09

Bad

1.08

apest

1.08

{*

1.07

Activations Density 0.016%

references to significant historical figures and their contributions or achievements

No Comments

No Known Activations