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

0:0.07

1:0.08

2:0.09

3:0.08

4:0.08

5:0.09

6:0.08

7:0.07

8:0.07

9:0.08

10:0.09

11:0.07

Negative Logits

 Triangle

-2.67

Ri

-2.54

��

-2.47

HK

-2.45

Sao

-2.42

SQ

-2.37

match

-2.34

�

-2.31

�

-2.31

agonal

-2.29

POSITIVE LOGITS

 encyclopedia

2.91

 Lois

2.70

 biod

2.69

 acne

2.57

geries

2.50

Pse

2.47

archives

2.41

 pathogens

2.38

 Byron

2.36

 malware

2.36

Activations Density 0.000%

No Known Activations

This feature has no known activations.