INDEX

Explanations

terms related to veterans and healthcare access

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.21

1:0.08

2:0.07

3:0.09

4:0.07

5:0.03

6:0.10

7:0.09

8:0.03

9:0.05

10:0.07

11:0.07

Negative Logits

 Boko

-3.32

 Wake

-3.29

UFC

-3.25

 Raleigh

-3.07

Dur

-3.04

 Surf

-3.03

Bey

-2.95

 Tunis

-2.92

 Braz

-2.92

Cyp

-2.83

POSITIVE LOGITS

 Bohem

3.81

 Malk

3.27

 Tibetan

3.25

 Holo

3.19

 Ambro

3.14

 Slovakia

2.96

 Stanley

2.95

 Slov

2.94

resso

2.94

Pav

2.92

Activations Density 0.001%

terms related to veterans and healthcare access

No Known Activations