INDEX

Explanations

various forms of word endings that suggest action or state

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

2:0.08

3:0.37

4:0.02

5:0.02

6:0.07

7:0.05

8:0.05

9:0.08

10:0.08

11:0.07

Negative Logits

�

-1.45

OLOGY

-1.22

ב

-1.19

מ

-1.18

ר

-1.13

ittees

-1.11

OLOG

-1.10

\\\\\\\\

-1.10

三

-1.09

 Holocaust

-1.09

POSITIVE LOGITS

idated

1.35

cit

1.34

odor

1.32

uating

1.29

erella

1.29

icio

1.28

nir

1.26

esse

1.22

athing

1.22

aby

1.21

Activations Density 0.017%

various forms of word endings that suggest action or state

No Comments

No Known Activations