INDEX

Explanations

references to technical specifications and platform independence

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

platform-related terminology and concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-att-kk

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

1:0.02

2:0.05

3:0.15

4:0.02

5:0.04

6:0.11

7:0.30

8:0.05

9:0.03

10:0.05

11:0.09

Negative Logits

netflix

-1.33

beck

-1.24

itual

-1.18

cox

-1.17

quist

-1.15

zyme

-1.13

ocamp

-1.11

 Werewolf

-1.11

 omission

-1.11

Pru

-1.09

POSITIVE LOGITS

taboola

1.42

 rails

1.30

 guiActiveUnfocused

1.26

 enterprises

1.13

dimension

1.09

rongh

1.09

 Hispan

1.08

 operates

1.08

 manufactures

1.08

Application

1.06

Activations Density 0.023%

references to technical specifications and platform independence

platform-related terminology and concepts

No Comments

No Known Activations