INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.3.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

1:0.08

2:0.10

3:0.08

4:0.08

5:0.06

6:0.09

7:0.08

8:0.08

9:0.06

10:0.09

11:0.07

Negative Logits

 compass

-1.25

keeping

-1.15

 responsible

-1.13

 conscience

-1.13

fal

-1.12

 descended

-1.09

 returning

-1.07

 praying

-1.05

wcs

-1.05

 tolerant

-1.04

POSITIVE LOGITS

ilib

1.52

 VIDE

1.45

田

1.39

PLA

1.38

ダ

1.37

 SPORTS

1.37

actionDate

1.35

テ

1.34

UI

1.31

ウ

1.31

Activations Density 0.000%

No Known Activations

This feature has no known activations.