INDEX

Explanations

words related to health problems or conditions

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

0:0.03

1:0.03

2:0.07

3:0.05

4:0.04

5:0.05

6:0.12

7:0.27

8:0.04

9:0.05

10:0.14

11:0.06

Negative Logits

 induct

-1.64

 promoters

-1.56

 promoter

-1.55

ttp

-1.43

 endorsement

-1.38

 solicit

-1.37

 transistor

-1.37

 mailing

-1.36

 alarms

-1.36

 dispos

-1.36

POSITIVE LOGITS

hattan

1.67

ueller

1.62

renheit

1.52

nces

1.52

yrinth

1.48

nan

1.45

north

1.43

sq

1.39

apsed

1.39

grim

1.37

Activations Density 0.000%

words related to health problems or conditions

No Comments

No Known Activations