INDEX

Explanations

actions related to viewing, reading, and interacting with media content

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_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.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

0:0.01

1:0.08

2:0.14

3:0.05

4:0.02

5:0.09

6:0.08

7:0.08

8:0.11

9:0.08

10:0.11

11:0.09

Negative Logits

rists

-1.08

netflix

-1.05

Downloadha

-1.02

bably

-0.98

ItemTracker

-0.96

pecially

-0.96

ocally

-0.95

milo

-0.95

γ

-0.95

 alike

-0.95

POSITIVE LOGITS

click

1.34

 webpage

1.25

 click

1.25

 permissions

1.23

 captcha

1.19

 homepage

1.15

 publisher

1.14

 slideshow

1.14

 file

1.11

 login

1.11

Activations Density 0.046%

actions related to viewing, reading, and interacting with media content

No Comments

No Known Activations