INDEX

Explanations

the letter 'R'

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

IFrame

Link

Head Attr Weights

0:0.13

1:0.24

2:0.06

3:0.06

4:0.05

5:0.03

6:0.05

7:0.06

8:0.06

9:0.05

10:0.10

11:0.06

Negative Logits

 Spect

-3.57

 Gran

-2.89

ís

-2.86

schild

-2.85

 Ghosts

-2.81

 Ichigo

-2.70

 shif

-2.69

olson

-2.68

inates

-2.68

 dyed

-2.63

POSITIVE LOGITS

Lup

5.60

Tup

3.04

urdue

2.91

Aub

2.85

 Pret

2.84

 Pleasant

2.80

male

2.79

po

2.78

 pneumonia

2.77

 caffe

2.65

Activations Density 0.000%

the letter 'R'

No Known Activations