INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

1:0.05

2:0.08

3:0.08

4:0.09

5:0.08

6:0.08

7:0.07

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

 signatures

-1.98

Rated

-1.68

 warrants

-1.58

 votes

-1.54

 precedent

-1.53

 provisions

-1.52

 trophies

-1.52

 petitions

-1.50

 Spike

-1.47

phies

-1.47

POSITIVE LOGITS

gdala

2.11

sych

1.83

amia

1.76

etus

1.73

imir

1.73

cius

1.72

�

1.70

ı

1.67

ocent

1.67

�

1.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.