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

0:0.08

1:0.10

2:0.11

3:0.06

4:0.08

5:0.07

6:0.08

7:0.08

8:0.08

9:0.08

10:0.07

11:0.07

Negative Logits

{\

-1.72

onomy

-1.71

ctors

-1.67

thood

-1.64

DragonMagazine

-1.63

avery

-1.61

 Attributes

-1.60

 latitude

-1.58

lesi

-1.52

adoes

-1.51

POSITIVE LOGITS

OC

1.43

 corro

1.41

 outright

1.32

 outstanding

1.32

 booked

1.28

 secretly

1.28

 abroad

1.27

 overseas

1.27

 illegally

1.26

nab

1.26

Activations Density 0.000%

No Known Activations

This feature has no known activations.