INDEX

Explanations

names and specific mentions of individuals

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

1:0.05

2:0.07

3:0.07

4:0.05

5:0.04

6:0.20

7:0.05

8:0.07

9:0.25

10:0.02

11:0.04

Negative Logits

 Pear

-3.50

Tao

-3.48

 pardon

-3.35

 Flan

-3.13

panic

-3.09

 Kirin

-3.08

 Pony

-3.08

Pand

-3.07

pas

-3.03

 pard

-3.00

POSITIVE LOGITS

 Mechanical

3.30

arcity

3.12

ogging

3.01

Mech

2.98

ACS

2.97

 Clippers

2.83

 Fiber

2.77

 Mech

2.69

MW

2.69

ilities

2.66

Activations Density 0.000%

names and specific mentions of individuals

No Comments

No Known Activations