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

0:0.08

1:0.08

2:0.08

3:0.09

4:0.07

5:0.09

6:0.07

7:0.08

8:0.08

9:0.07

10:0.07

11:0.09

Negative Logits

 leaks

-3.20

 redund

-3.05

 drip

-3.02

 offsets

-2.99

 restraints

-2.99

 compromise

-2.98

 roomm

-2.90

 renters

-2.89

 dues

-2.87

 leak

-2.86

POSITIVE LOGITS

fman

3.14

�

3.12

Kh

2.84

Kem

2.83

Ik

2.81

エル

2.71

ニ

2.69

─

2.67

KK

2.65

heid

2.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.