INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.07

1:0.09

2:0.08

3:0.08

4:0.07

5:0.08

6:0.07

7:0.09

8:0.09

9:0.08

10:0.07

11:0.08

Negative Logits

edded

-2.61

lation

-2.44

Coun

-2.29

EXP

-2.27

estate

-2.26

 renovations

-2.26

ocument

-2.25

 renovation

-2.24

ESSION

-2.24

CRIPTION

-2.23

POSITIVE LOGITS

wake

2.61

eele

2.39

ilaterally

2.34

 Mane

2.31

 harb

2.31

icol

2.30

 Vand

2.29

Vaj

2.25

Fib

2.17

 dstg

2.10

Activations Density 0.000%

No Known Activations

This feature has no known activations.