INDEX

Explanations

terms related to ideas or concepts being introduced or discussed

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_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

0:0.01

1:0.01

2:0.05

3:0.09

4:0.38

5:0.02

6:0.06

7:0.15

8:0.03

9:0.03

10:0.06

11:0.06

Negative Logits

 belie

-1.42

lid

-1.39

science

-1.38

 imagination

-1.34

 hemisphere

-1.32

 rollout

-1.31

 cred

-1.29

 taxp

-1.29

 contrace

-1.29

thumbnails

-1.28

POSITIVE LOGITS

 similarities

1.62

 resemb

1.54

ridges

1.49

Msg

1.46

Leaks

1.39

DonaldTrump

1.36

 Trace

1.35

 Flavoring

1.33

 Piercing

1.32

 Hidden

1.31

Activations Density 0.011%

terms related to ideas or concepts being introduced or discussed

No Comments

No Known Activations