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

0:0.07

1:0.05

2:0.08

3:0.07

4:0.09

5:0.06

6:0.08

7:0.07

8:0.10

9:0.09

10:0.09

11:0.08

Negative Logits

 Australians

-1.57

 Miko

-1.46

 Gabriel

-1.44

WWF

-1.41

 Denis

-1.39

CES

-1.39

 Dalai

-1.39

 brothers

-1.36

 Advocate

-1.34

vous

-1.34

POSITIVE LOGITS

esthesia

1.51

illac

1.50

 lookup

1.44

ifference

1.43

 disappear

1.41

VERTISEMENT

1.40

 whit

1.40

lished

1.39

 populated

1.39

tenance

1.38

Activations Density 0.000%

No Known Activations

This feature has no known activations.