INDEX

Explanations

occurrences and patterns of specific nouns and phrases

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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.21

1:0.03

2:0.05

3:0.26

4:0.01

5:0.09

6:0.02

7:0.04

8:0.03

9:0.01

10:0.18

11:0.02

Negative Logits

vernment

-2.32

Live

-2.32

 specialize

-2.25

urally

-2.00

 Surviv

-1.99

UTE

-1.88

 Athlet

-1.86

SPONSORED

-1.85

 VIDEOS

-1.85

 Practices

-1.84

POSITIVE LOGITS

 drawer

2.28

 sleeve

2.14

 wallet

2.12

shirt

2.04

 trunk

2.01

 bucket

2.01

 iceberg

1.97

 sleeves

1.89

 wallets

1.85

pants

1.83

Activations Density 0.030%

occurrences and patterns of specific nouns and phrases

No Comments

No Known Activations