INDEX

Explanations

numerical data related to dates or counts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

0:0.22

1:0.08

2:0.08

3:0.06

4:0.02

5:0.11

6:0.08

7:0.01

8:0.06

9:0.13

10:0.05

11:0.05

Negative Logits

minist

-1.84

nown

-1.80

leck

-1.69

�

-1.69

eele

-1.66

vous

-1.62

estern

-1.57

lyak

-1.56

oples

-1.53

ewitness

-1.51

POSITIVE LOGITS

1.83

1.35

�

1.35

1.31

)/

1.30

�

1.27

 Sessions

1.26

 offic

1.24

・

1.22

 guest

1.21

Activations Density 0.004%

numerical data related to dates or counts

No Comments

No Known Activations