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

1:0.07

2:0.07

3:0.10

4:0.08

5:0.07

6:0.08

7:0.07

8:0.09

9:0.07

10:0.08

11:0.07

Negative Logits

 infancy

-1.60

idia

-1.60

fur

-1.59

�

-1.59

 wears

-1.57

�

-1.54

*/

-1.52

 surpassed

-1.52

�

-1.50

��

-1.50

POSITIVE LOGITS

ablishment

1.87

 Procedures

1.87

Commerce

1.82

eals

1.63

lished

1.62

ourning

1.61

 suspending

1.60

uates

1.59

Kings

1.58

wards

1.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.