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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.09

2:0.08

3:0.07

4:0.08

5:0.07

6:0.09

7:0.07

8:0.07

9:0.09

10:0.08

11:0.07

Negative Logits

 tennis

-3.06

 Tennis

-3.03

 colourful

-2.94

�

-2.93

Sport

-2.67

pu

-2.66

 Teddy

-2.60

 LOVE

-2.53

Scot

-2.50

 warm

-2.50

POSITIVE LOGITS

lihood

2.78

 substitutes

2.76

CMS

2.73

 Cosponsors

2.67

Ary

2.59

ramid

2.56

SPONSORED

2.55

 Mald

2.44

hovah

2.42

 Instruments

2.41

Activations Density 0.000%

No Known Activations

This feature has no known activations.