INDEX

Explanations

the preposition "of."

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

Head Attr Weights

0:0.08

1:0.07

2:0.09

3:0.08

4:0.07

5:0.09

6:0.07

7:0.09

8:0.07

9:0.07

10:0.09

11:0.07

Negative Logits

improve

-3.00

SEA

-2.89

vag

-2.79

philis

-2.75

Tus

-2.65

cellence

-2.63

Score

-2.63

ivari

-2.58

Miami

-2.54

ever

-2.48

POSITIVE LOGITS

 Fein

2.78

 Citiz

2.63

 Aren

2.61

 Faction

2.58

 Header

2.58

 Borders

2.56

 Emblem

2.55

 Murdoch

2.55

 broadcaster

2.44

 Hobby

2.44

Activations Density 0.000%

the preposition "of."

No Known Activations