INDEX

Explanations

keywords related to navigating through a process or setting up configurations

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

words related to navigation and processes

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

Head Attr Weights

0:0.05

1:0.04

2:0.14

3:0.08

4:0.11

5:0.05

6:0.02

7:0.04

8:0.20

9:0.12

10:0.07

11:0.03

Negative Logits

cientious

-1.38

gemony

-1.29

arcity

-1.29

inguished

-1.28

cu

-1.26

nces

-1.24

ortunate

-1.22

ingers

-1.22

rists

-1.21

leck

-1.20

POSITIVE LOGITS

 Travels

1.60

 maze

1.53

ById

1.39

 labyrinth

1.32

 terrain

1.29

›

1.29

 bounds

1.29

 route

1.27

 directions

1.27

 boundaries

1.27

Activations Density 0.011%

keywords related to navigating through a process or setting up configurations

words related to navigation and processes

No Known Activations