INDEX

Explanations

the presence of the name "Abdul" and variations of it in various contexts

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

1:0.03

2:0.05

3:0.07

4:0.06

5:0.05

6:0.15

7:0.06

8:0.04

9:0.30

10:0.03

11:0.05

Negative Logits

 Bryce

-3.23

 Coulter

-3.14

 Duffy

-3.13

Cla

-3.12

Mau

-3.07

 Ariel

-3.01

Mn

-2.98

 Christine

-2.92

 Randy

-2.89

 Cliff

-2.85

POSITIVE LOGITS

 suscept

2.84

 guiActiveUnfocused

2.79

 meats

2.78

Rail

2.73

ctors

2.69

 Fitz

2.67

ains

2.67

steam

2.65

arth

2.60

vable

2.60

Activations Density 0.003%

the presence of the name "Abdul" and variations of it in various contexts

No Comments

No Known Activations