INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

0:0.08

1:0.08

2:0.10

3:0.07

4:0.08

5:0.08

6:0.07

7:0.07

8:0.09

9:0.09

10:0.07

11:0.07

Negative Logits

waukee

-1.54

within

-1.53

onto

-1.47

osaurus

-1.44

foot

-1.42

Torrent

-1.41

feet

-1.41

MSN

-1.41

lear

-1.40

Squ

-1.37

POSITIVE LOGITS

 ende

1.68

 confirmation

1.50

ikh

1.45

 estimation

1.44

uthor

1.41

 amnesty

1.41

ayn

1.40

asha

1.38

 consultation

1.37

 excuse

1.34

Activations Density 0.000%

No Known Activations

This feature has no known activations.