INDEX

Explanations

terms related to achievements and accomplishments in sports and artistic endeavors

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.05

1:0.02

2:0.05

3:0.45

4:0.02

5:0.05

6:0.02

7:0.04

8:0.04

9:0.02

10:0.16

11:0.03

Negative Logits

 poisons

-2.54

 terrorists

-2.36

 cleaners

-2.31

 environmentalists

-2.27

 gangs

-2.19

 programmers

-2.15

 thugs

-2.12

 idiots

-2.10

 protestors

-2.09

 mobs

-2.08

POSITIVE LOGITS

 outing

2.74

 anniversary

2.63

 foray

2.62

 audition

2.54

 debut

2.54

 assignment

2.51

 birthday

2.47

trial

2.42

 consecutive

2.42

ourney

2.41

Activations Density 0.118%

terms related to achievements and accomplishments in sports and artistic endeavors

No Comments

No Known Activations