INDEX

Explanations

phrases or references related to political strategy and legislative actions

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_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.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.11

1:0.02

2:0.07

3:0.35

4:0.04

5:0.08

6:0.02

7:0.06

8:0.07

9:0.02

10:0.09

11:0.04

Negative Logits

ateurs

-2.09

 Subtle

-2.03

nesday

-1.97

 Advertisement

-1.94

 Reincarn

-1.94

ruary

-1.88

ustom

-1.86

 Ideal

-1.84

cellence

-1.84

 Rational

-1.83

POSITIVE LOGITS

FCC

2.01

 firewall

1.97

 congressional

1.90

FBI

1.89

 Grassley

1.85

FBI

1.76

FAA

1.75

DOD

1.75

 wire

1.74

 inhibitor

1.74

Activations Density 0.035%

phrases or references related to political strategy and legislative actions

No Known Activations