INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

0:0.09

1:0.08

2:0.09

3:0.08

4:0.10

5:0.07

6:0.08

7:0.08

8:0.07

9:0.07

10:0.07

11:0.07

Negative Logits

LIC

-1.72

 Canaver

-1.57

 Crowley

-1.53

 Paste

-1.47

 Ezek

-1.47

 VIDE

-1.46

gewater

-1.44

 veter

-1.40

oro

-1.39

onement

-1.36

POSITIVE LOGITS

footed

1.52

SpaceEngineers

1.50

 encountered

1.46

 stumble

1.41

ridden

1.40

got

1.39

 faced

1.38

 Nebula

1.33

inch

1.33

kept

1.33

Activations Density 0.000%

No Known Activations

This feature has no known activations.