INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_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.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.09

1:0.08

2:0.07

3:0.08

4:0.08

5:0.07

6:0.08

7:0.08

8:0.07

9:0.09

10:0.07

11:0.08

Negative Logits

 Career

-3.00

 Quentin

-2.85

Chu

-2.83

 Orient

-2.81

 Exploration

-2.74

 Treasure

-2.73

 Solo

-2.72

 Minerva

-2.68

Pag

-2.66

 Bonus

-2.66

POSITIVE LOGITS

 broadcasting

2.88

 broadcasters

2.87

 radios

2.71

 hurd

2.66

 pestic

2.66

 cultured

2.63

 teasp

2.57

 mathemat

2.53

 convol

2.50

 latch

2.48

Activations Density 0.000%

No Known Activations

This feature has no known activations.