INDEX

Explanations

situations or phrases related to legal complaints and potential lawsuits

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

phrases and words related to social dynamics and personal connections

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

0:0.09

1:0.05

2:0.24

3:0.07

4:0.11

5:0.10

6:0.04

7:0.02

8:0.07

9:0.06

10:0.06

11:0.03

Negative Logits

andowski

-1.41

 Flavoring

-1.36

enser

-1.26

 Craigslist

-1.26

 circuitry

-1.26

quished

-1.24

berus

-1.22

oven

-1.21

DVD

-1.18

Detailed

-1.18

POSITIVE LOGITS

ás

1.26

onen

1.21

idine

1.14

 Monteneg

1.11

pas

1.10

na

1.07

tty

1.07

é

1.05

ana

1.05

 observ

1.04

Activations Density 0.002%

situations or phrases related to legal complaints and potential lawsuits

phrases and words related to social dynamics and personal connections

No Comments

No Known Activations