INDEX

Explanations

references to blood alcohol content and legal limits

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

0:0.02

1:0.01

2:0.11

3:0.06

4:0.03

5:0.03

6:0.04

7:0.21

8:0.03

9:0.04

10:0.28

11:0.10

Negative Logits

projects

-1.78

issance

-1.77

 tutorials

-1.62

uci

-1.59

 automate

-1.58

knit

-1.58

updated

-1.57

 deleg

-1.54

umn

-1.54

hai

-1.54

POSITIVE LOGITS

alcohol

2.18

 saliva

2.06

 clen

1.88

 serum

1.80

 steroid

1.76

 adrenaline

1.70

 gasoline

1.69

TNT

1.66

 spiked

1.64

 intox

1.63

Activations Density 0.000%

references to blood alcohol content and legal limits

No Comments

No Known Activations