INDEX

Explanations

topics related to disciplinary actions and compliance issues within a fire department

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.18

1:0.06

2:0.05

3:0.07

4:0.04

5:0.02

6:0.15

7:0.16

8:0.02

9:0.04

10:0.07

11:0.08

Negative Logits

 drib

-3.14

 embargo

-3.13

 centrif

-3.12

milo

-3.10

 imprison

-3.07

pir

-3.04

 rapists

-2.85

ollah

-2.77

isse

-2.74

 Guant

-2.70

POSITIVE LOGITS

 firefighter

5.79

 firefighters

5.11

EMS

4.67

 paramed

4.11

 paramedics

3.75

 Fire

3.66

Fire

3.50

 Volunteer

3.36

 fire

3.31

 ambulance

3.28

Activations Density 0.015%

topics related to disciplinary actions and compliance issues within a fire department

No Known Activations