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

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.08

2:0.08

3:0.07

4:0.07

5:0.07

6:0.07

7:0.08

8:0.07

9:0.08

10:0.09

11:0.08

Negative Logits

Bron

-3.18

 Decker

-3.10

Doe

-2.81

�

-2.60

ukemia

-2.60

Whit

-2.51

Rust

-2.51

Gil

-2.50

 Bron

-2.46

Hash

-2.44

POSITIVE LOGITS

 spons

2.65

gow

2.56

acca

2.56

 Lever

2.38

 Affect

2.32

Software

2.28

 Influence

2.27

 oblig

2.25

anca

2.23

 Swing

2.23

Activations Density 0.000%

No Known Activations

This feature has no known activations.