INDEX

Explanations

instances of participation or involvement in various activities and events

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

0:0.02

1:0.01

2:0.08

3:0.06

4:0.14

5:0.03

6:0.03

7:0.38

8:0.02

9:0.03

10:0.09

11:0.06

Negative Logits

 pharmacies

-1.48

ibl

-1.45

 porous

-1.43

god

-1.40

 consistency

-1.39

haven

-1.38

 expiration

-1.37

iuses

-1.36

users

-1.34

 hindsight

-1.33

POSITIVE LOGITS

 festivities

1.82

 fray

1.70

 conversation

1.63

EngineDebug

1.60

 proceedings

1.59

 conference

1.49

 parade

1.49

 Caption

1.47

 discussions

1.47

ourney

1.46

Activations Density 0.010%

instances of participation or involvement in various activities and events

No Comments

No Known Activations