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

1:0.06

2:0.09

3:0.08

4:0.08

5:0.09

6:0.09

7:0.08

8:0.08

9:0.08

10:0.07

11:0.06

Negative Logits

Forest

-1.82

Force

-1.73

 Genie

-1.67

 Polk

-1.58

DP

-1.52

Hope

-1.48

Policy

-1.48

Gov

-1.48

 Mountain

-1.47

Party

-1.47

POSITIVE LOGITS

xual

1.98

gaard

1.98

yip

1.91

sov

1.82

arnaev

1.82

akeru

1.79

rences

1.76

retty

1.73

uggle

1.72

oly

1.71

Activations Density 0.000%

No Known Activations

This feature has no known activations.