INDEX

Explanations

references to specific individuals or entities

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.05

2:0.12

3:0.04

4:0.04

5:0.04

6:0.20

7:0.04

8:0.07

9:0.16

10:0.04

11:0.05

Negative Logits

antry

-4.39

 Castle

-3.77

 Hogan

-3.58

 Barton

-3.53

aceous

-3.48

 Earl

-3.41

 Gilmore

-3.31

 castles

-3.28

Tro

-3.28

leck

-3.27

POSITIVE LOGITS

Neg

8.25

Neg

7.23

neg

6.99

neg

6.85

 negro

4.59

 negate

4.21

 Negro

4.12

 Toro

4.01

 Negative

4.01

 Mund

3.89

Activations Density 0.000%

references to specific individuals or entities

No Comments

No Known Activations