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

0:0.08

1:0.06

2:0.07

3:0.08

4:0.09

5:0.08

6:0.08

7:0.09

8:0.09

9:0.08

10:0.09

11:0.07

Negative Logits

 Seym

-1.64

ouse

-1.50

andre

-1.46

 Deliver

-1.45

geries

-1.42

 Wonderland

-1.41

demon

-1.40

asca

-1.40

heart

-1.39

 backdrop

-1.37

POSITIVE LOGITS

ggles

1.72

odox

1.70

ipedia

1.66

1.58

ciating

1.54

atel

1.49

irie

1.44

otor

1.38

peed

1.38

icion

1.33

Activations Density 0.000%

No Known Activations

This feature has no known activations.