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

0:0.09

1:0.07

2:0.08

3:0.08

4:0.09

5:0.08

6:0.06

7:0.08

8:0.08

9:0.07

10:0.08

11:0.09

Negative Logits

 disagree

-1.97

ardless

-1.81

ento

-1.77

DERR

-1.63

 adultery

-1.57

ujah

-1.55

 snipp

-1.55

 disqualified

-1.54

cember

-1.53

athe

-1.53

POSITIVE LOGITS

 Maxim

1.85

 Provided

1.71

uyomi

1.68

 Gravity

1.61

 Zoom

1.59

 Known

1.58

 monop

1.57

Known

1.53

 Locked

1.52

WIND

1.50

Activations Density 0.000%

No Known Activations

This feature has no known activations.