INDEX

Explanations

names of people and places, particularly focusing on specific names such as "William Gladstone" and "Gladys Kalemazoo"

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

references to the name "Gladstone."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-att-kk

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

0:0.07

1:0.03

2:0.23

3:0.09

4:0.13

5:0.06

6:0.04

7:0.03

8:0.07

9:0.11

10:0.06

11:0.03

Negative Logits

Obj

-1.29

 variables

-1.25

 classification

-1.24

 veter

-1.24

Topics

-1.21

TAG

-1.19

 scrut

-1.17

ccording

-1.16

 millenn

-1.15

awa

-1.12

POSITIVE LOGITS

eele

1.26

stice

1.22

beck

1.22

andise

1.22

stad

1.22

bey

1.18

vette

1.18

addin

1.17

antis

1.16

 Reloaded

1.16

Activations Density 0.001%

names of people and places, particularly focusing on specific names such as "William Gladstone" and "Gladys Kalemazoo"

references to the name "Gladstone."

No Comments

No Known Activations