INDEX

Explanations

references to fame and notoriety associated with individuals or places

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

1:0.02

2:0.06

3:0.05

4:0.09

5:0.03

6:0.03

7:0.38

8:0.03

9:0.04

10:0.10

11:0.10

Negative Logits

rial

-1.57

xit

-1.54

plet

-1.52

ascript

-1.50

ilage

-1.48

raints

-1.48

 plane

-1.48

umbs

-1.47

gradient

-1.47

 solution

-1.45

POSITIVE LOGITS

CLASS

1.63

��

1.56

 outlaw

1.50

 Crimes

1.50

 derogatory

1.45

Merit

1.45

 nick

1.43

 Theft

1.42

م

1.42

 obnoxious

1.39

Activations Density 0.008%

references to fame and notoriety associated with individuals or places

No Comments

No Known Activations