INDEX

Explanations

sentences with the word "see" followed by specific descriptions or actions

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

Head Attr Weights

0:0.06

1:0.03

2:0.04

3:0.05

4:0.04

5:0.05

6:0.05

7:0.10

8:0.23

9:0.05

10:0.17

11:0.09

Negative Logits

ufact

-1.18

theless

-1.16

 Recomm

-1.13

 NETWORK

-1.02

atern

-1.00

 sugg

-0.96

asonic

-0.95

agonist

-0.94

 Prevention

-0.94

angan

-0.93

POSITIVE LOGITS

 similarities

1.21

 firsthand

1.18

 unfold

1.15

 cartoons

1.03

��

1.01

CVE

0.98

Tokens

0.98

 VIDEOS

0.96

Snap

0.95

 mockery

0.95

Activations Density 0.204%

sentences with the word "see" followed by specific descriptions or actions

No Known Activations