INDEX

Explanations

concepts related to online privacy and data security

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

0:0.09

1:0.04

2:0.15

3:0.15

4:0.05

5:0.15

6:0.03

7:0.09

8:0.03

9:0.04

10:0.08

11:0.03

Negative Logits

 Faul

-2.11

 Negro

-2.06

 Senegal

-1.96

 tents

-1.92

interstitial

-1.89

 Tropical

-1.80

 Construction

-1.78

 Burns

-1.75

 Staten

-1.75

 Muss

-1.74

POSITIVE LOGITS

encrypted

3.12

 password

2.98

 passwords

2.96

 encrypted

2.95

 securely

2.86

Password

2.86

 stored

2.71

 decrypt

2.67

 identifiers

2.60

 authentication

2.59

Activations Density 0.481%

concepts related to online privacy and data security

No Comments

No Known Activations