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

2:0.08

3:0.09

4:0.09

5:0.07

6:0.08

7:0.08

8:0.07

9:0.07

10:0.09

11:0.09

Negative Logits

minecraft

-2.08

lime

-1.97

chrom

-1.91

MSN

-1.80

\-

-1.79

package

-1.78

Measure

-1.71

olor

-1.70

mine

-1.69

orb

-1.68

POSITIVE LOGITS

 Painter

1.81

 Bauer

1.80

���

1.74

 Sett

1.65

 Ancients

1.61

 Kaplan

1.59

 Frankfurt

1.52

 spoiled

1.52

 Klaus

1.51

��

1.51

Activations Density 0.000%

No Known Activations

This feature has no known activations.