INDEX

Explanations

phrases related to formal documentation and statements

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

0:0.17

1:0.08

2:0.05

3:0.04

4:0.06

5:0.02

6:0.16

7:0.13

8:0.01

9:0.05

10:0.15

11:0.02

Negative Logits

 Zimmerman

-3.09

 Marin

-3.05

 Robinson

-2.99

 Winston

-2.92

 colony

-2.77

 Nero

-2.75

Nan

-2.74

 Rays

-2.66

pac

-2.64

 Santiago

-2.63

POSITIVE LOGITS

 Grassley

8.05

 Hatch

3.92

Gr

3.88

 Grass

3.75

 Judiciary

3.69

 Chuck

3.33

Iowa

3.33

grass

3.21

 Iowa

3.15

 Leah

3.12

Activations Density 0.001%

phrases related to formal documentation and statements

No Known Activations