INDEX

Explanations

phrases related to adjusting or increasing something, specifically related to prediction accuracy or physical products

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

phrases associated with adjusting or modifying settings

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

0:0.05

1:0.03

2:0.24

3:0.06

4:0.16

5:0.05

6:0.02

7:0.02

8:0.09

9:0.14

10:0.05

11:0.02

Negative Logits

 Salem

-1.44

 Alam

-1.27

noon

-1.24

ework

-1.24

 Lent

-1.22

ndra

-1.20

idav

-1.16

ammy

-1.15

ukong

-1.13

 conclud

-1.13

POSITIVE LOGITS

 guiActiveUn

1.57

Enhanced

1.43

Dynamic

1.32

Os

1.25

Hyper

1.24

Interface

1.21

ById

1.16

YP

1.15

onics

1.15

ysis

1.14

Activations Density 0.005%

phrases related to adjusting or increasing something, specifically related to prediction accuracy or physical products

phrases associated with adjusting or modifying settings

No Comments

No Known Activations