INDEX

Explanations

situations involving accidents and injuries related to vehicles

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.12

1:0.03

2:0.10

3:0.04

4:0.06

5:0.06

6:0.25

7:0.03

8:0.07

9:0.15

10:0.02

11:0.02

Negative Logits

 gymn

-3.67

 baths

-3.33

metics

-3.32

 calibr

-3.31

 Ichigo

-3.27

 Bleach

-3.26

 Lauder

-3.23

scill

-3.20

circ

-3.19

 philos

-3.17

POSITIVE LOGITS

 Truck

8.43

 truck

8.34

 trucks

8.11

Tru

6.83

 pickup

6.11

 Jeep

5.98

 tractor

5.96

 trailer

5.74

 trailers

5.42

RV

5.25

Activations Density 0.026%

situations involving accidents and injuries related to vehicles

No Comments

No Known Activations