INDEX

Explanations

terms related to protective measures and defense mechanisms

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_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

0:0.02

1:0.01

2:0.07

3:0.05

4:0.05

5:0.03

6:0.05

7:0.11

8:0.03

9:0.04

10:0.33

11:0.14

Negative Logits

 guiActiveUn

-1.65

rawdownloadcloneembedreportprint

-1.51

OVA

-1.44

Filename

-1.38

umbnails

-1.38

eday

-1.37

archive

-1.37

Synopsis

-1.37

furt

-1.33

vier

-1.33

POSITIVE LOGITS

 protects

1.68

 shield

1.66

 protecting

1.60

 shielding

1.54

 protect

1.53

 armor

1.46

 effectiveness

1.44

 mantle

1.43

 Protect

1.42

 protection

1.40

Activations Density 0.279%

terms related to protective measures and defense mechanisms

No Comments

No Known Activations