INDEX

Explanations

references to government agencies or departments

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

Not in Any Lists

No Comments

Head Attr Weights

0:0.05

1:0.02

2:0.16

3:0.13

4:0.04

5:0.06

6:0.09

7:0.04

8:0.06

9:0.25

10:0.03

11:0.03

Negative Logits

 Coul

-3.56

 Cameroon

-3.06

 convergence

-3.05

 Trials

-3.04

 Astral

-3.00

Fay

-2.92

aya

-2.91

rals

-2.83

pson

-2.79

 Brisbane

-2.79

POSITIVE LOGITS

Secretary

3.81

Department

3.74

DHS

3.68

HHS

3.54

soType

3.51

 Department

3.21

channelAvailability

3.19

 UCHIJ

3.07

DOJ

3.05

 secretary

3.04

Activations Density 0.001%

references to government agencies or departments

No Comments

No Known Activations