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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.07

1:0.07

2:0.09

3:0.08

4:0.08

5:0.08

6:0.07

7:0.08

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

soType

-3.21

issance

-2.75

actionDate

-2.67

AAP

-2.61

ana

-2.60

PB

-2.54

神

-2.51

Dal

-2.48

Aad

-2.46

amba

-2.44

POSITIVE LOGITS

♦

2.94

 Shannon

2.69

 airplanes

2.58

\/\/

2.56

 mastered

2.50

 fighter

2.48

 punches

2.47

utral

2.30

 punched

2.29

ighters

2.27

Activations Density 0.000%

No Known Activations

This feature has no known activations.