INDEX

Explanations

mentions of festivals and related events

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

2:0.06

3:0.11

4:0.03

5:0.05

6:0.10

7:0.29

8:0.05

9:0.05

10:0.08

11:0.08

Negative Logits

xon

-1.28

amination

-1.16

oxide

-1.15

cro

-1.15

lder

-1.15

gaard

-1.14

 tampering

-1.13

odan

-1.12

axy

-1.10

aron

-1.09

POSITIVE LOGITS

estival

1.37

 Ramadan

1.35

Eid

1.32

 festivities

1.32

 celebration

1.28

Fest

1.27

 honors

1.25

 rites

1.24

 themed

1.20

 commem

1.20

Activations Density 0.033%

mentions of festivals and related events

No Comments

No Known Activations