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

0:0.05

1:0.06

2:0.08

3:0.09

4:0.08

5:0.09

6:0.09

7:0.10

8:0.06

9:0.07

10:0.08

11:0.09

Negative Logits

�

-1.68

redo

-1.61

>>

-1.59

mans

-1.57

:=

-1.55

�

-1.53

uckles

-1.53

lvl

-1.53

�

-1.51

======

-1.51

POSITIVE LOGITS

 anonymously

1.68

 Euras

1.58

 Falk

1.56

 anonym

1.50

 Nikola

1.46

 Antarctica

1.46

dating

1.44

 faked

1.43

 UFOs

1.42

 Crimea

1.40

Activations Density 0.000%

No Known Activations

This feature has no known activations.