INDEX

Explanations

action words related to offering assistance or making positive changes

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

0:0.07

1:0.02

2:0.15

3:0.19

4:0.03

5:0.09

6:0.03

7:0.01

8:0.23

9:0.05

10:0.05

11:0.02

Negative Logits

 Huntington

-1.36

 branches

-1.33

 estates

-1.33

 fields

-1.31

earchers

-1.30

 births

-1.29

plings

-1.29

ources

-1.28

ages

-1.28

oved

-1.28

POSITIVE LOGITS

oser

1.64

izzy

1.48

kish

1.42

ENSE

1.41

Poké

1.40

assian

1.40

 rethink

1.39

icum

1.39

 berth

1.37

erick

1.34

Activations Density 0.041%

action words related to offering assistance or making positive changes

No Comments

No Known Activations