INDEX

Explanations

proper nouns, specifically names

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

0:0.10

1:0.04

2:0.09

3:0.06

4:0.06

5:0.05

6:0.21

7:0.07

8:0.06

9:0.12

10:0.05

11:0.05

Negative Logits

 Pepsi

-3.95

Cola

-3.50

 dash

-3.34

 Nissan

-3.30

 Dash

-3.25

 Nept

-3.22

CAP

-3.14

 Blitz

-3.14

 Booker

-3.06

 Richardson

-3.02

POSITIVE LOGITS

lod

8.70

 lodge

7.81

Lod

7.75

 Lodge

7.16

 lodging

6.22

 lodged

4.18

 skeleton

3.83

 skeletons

3.74

ォ

3.63

 Masonic

3.57

Activations Density 0.000%

proper nouns, specifically names

No Comments

No Known Activations