INDEX

Explanations

references to individuals and their actions or experiences

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_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.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

0:0.16

1:0.10

2:0.09

3:0.05

4:0.03

5:0.03

6:0.14

7:0.10

8:0.02

9:0.04

10:0.13

11:0.05

Negative Logits

Frag

-3.31

 tank

-2.93

 Katy

-2.93

irit

-2.70

 mint

-2.65

 Frag

-2.64

 inert

-2.61

 biod

-2.61

tank

-2.54

bab

-2.48

POSITIVE LOGITS

 Cunningham

8.81

 Cary

3.99

 Compton

3.41

 Chung

3.40

Cox

3.33

 Swanson

3.30

 Dunn

3.24

Cov

3.13

 Durham

3.12

 Glover

3.04

Activations Density 0.002%

references to individuals and their actions or experiences

No Comments

No Known Activations