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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.06

2:0.10

3:0.07

4:0.09

5:0.07

6:0.06

7:0.06

8:0.08

9:0.09

10:0.09

11:0.09

Negative Logits

stals

-1.56

iven

-1.42

Reviewer

-1.33

reviewed

-1.32

ther

-1.24

 accessed

-1.23

 resides

-1.20

Cert

-1.19

reads

-1.18

POSITIVE LOGITS

alion

1.64

hai

1.51

 Shogun

1.50

️

1.45

senal

1.45

──

1.43

ogun

1.39

 grunt

1.39

Pwr

1.37

 Reck

1.37

Activations Density 0.000%

No Known Activations

This feature has no known activations.