INDEX

Explanations

action verbs and phrases related to user engagement or interaction

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

1:0.04

2:0.07

3:0.17

4:0.02

5:0.02

6:0.12

7:0.08

8:0.06

9:0.20

10:0.06

11:0.06

Negative Logits

ufact

-1.55

sed

-1.28

udeb

-1.22

��

-1.16

ccess

-1.16

ngth

-1.16

wered

-1.14

phony

-1.11

student

-1.11

arest

-1.10

POSITIVE LOGITS

MSN

1.29

Arcade

1.21

 inputs

1.13

 infrared

1.11

 heels

1.09

BILL

1.09

 toggle

1.04

 advis

1.04

 paran

1.02

 Unemployment

1.02

Activations Density 0.008%

action verbs and phrases related to user engagement or interaction

No Comments

No Known Activations