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

2:0.10

3:0.08

4:0.07

5:0.06

6:0.09

7:0.10

8:0.08

9:0.07

10:0.09

11:0.08

Negative Logits

 caution

-1.95

 endif

-1.81

NetMessage

-1.76

 satur

-1.71

 mantra

-1.67

ted

-1.64

 magnification

-1.63

 compl

-1.62

 glam

-1.60

endif

-1.56

POSITIVE LOGITS

rafted

1.98

ヘ

1.76

early

1.76

owship

1.72

assad

1.67

arations

1.64

alus

1.63

otom

1.62

ansion

1.58

emaker

1.55

Activations Density 0.000%

No Known Activations

This feature has no known activations.