INDEX

Explanations

mock-ups of designs and products

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

New Auto-Interp

Top Features by Cosine Similarity

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

0:0.04

1:0.03

2:0.17

3:0.09

4:0.19

5:0.05

6:0.03

7:0.02

8:0.13

9:0.13

10:0.04

11:0.03

Negative Logits

 GOODMAN

-1.51

 actionGroup

-1.51

 redistributed

-1.29

 earners

-1.28

 taxable

-1.23

 dividends

-1.23

 sanitation

-1.22

 taxed

-1.20

 Forestry

-1.19

 indexed

-1.19

POSITIVE LOGITS

 caric

1.52

gery

1.50

reck

1.47

onite

1.33

 mock

1.31

tics

1.29

 Mock

1.28

 Lann

1.27

wolves

1.26

ragon

1.23

Activations Density 0.009%

mock-ups of designs and products

No Comments

No Known Activations