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

0:0.07

1:0.07

2:0.09

3:0.08

4:0.09

5:0.08

6:0.08

7:0.08

8:0.08

9:0.08

10:0.08

11:0.07

Negative Logits

 Miles

-3.42

 Cable

-3.03

 Lane

-2.95

 Voltage

-2.93

 Maid

-2.88

MPG

-2.83

 Champ

-2.79

 Mile

-2.77

 Peggy

-2.77

Pry

-2.76

POSITIVE LOGITS

pse

3.12

inia

2.99

 overfl

2.84

aurus

2.84

 recol

2.72

ongyang

2.60

 confir

2.57

 opin

2.54

 primates

2.51

 includ

2.50

Activations Density 0.000%

No Known Activations

This feature has no known activations.