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

0:0.09

1:0.07

2:0.09

3:0.08

4:0.08

5:0.08

6:0.08

7:0.07

8:0.07

9:0.09

10:0.08

11:0.07

Negative Logits

puted

-1.58

ripp

-1.52

iscal

-1.50

interrupted

-1.49

gross

-1.47

spring

-1.46

 collectively

-1.46

terness

-1.45

 dissolved

-1.44

 peacefully

-1.44

POSITIVE LOGITS

 Thieves

1.71

 Chau

1.62

obiles

1.60

epad

1.59

 Finder

1.57

atown

1.54

sle

1.50

andals

1.47

owler

1.47

 Courier

1.44

Activations Density 0.000%

No Known Activations

This feature has no known activations.