INDEX

Explanations

No Explanations Found

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

0:0.08

1:0.09

2:0.09

3:0.09

4:0.07

5:0.08

6:0.07

7:0.07

8:0.07

9:0.06

10:0.09

11:0.09

Negative Logits

unknown

-3.15

wagen

-2.74

auder

-2.73

eln

-2.71

igr

-2.69

ussen

-2.62

ctuary

-2.61

oubted

-2.54

eller

-2.53

merce

-2.48

POSITIVE LOGITS

 Loaded

2.67

 Twist

2.50

rig

2.46

…………

2.45

……………………

2.39

 Flex

2.36

 feminism

2.36

dad

2.35

…."

2.34

 dads

2.32

Activations Density 0.000%

No Known Activations

This feature has no known activations.