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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.07

2:0.09

3:0.07

4:0.07

5:0.06

6:0.09

7:0.09

8:0.08

9:0.08

10:0.09

11:0.07

Negative Logits

gor

-1.68

AMI

-1.65

mun

-1.63

Redditor

-1.59

sites

-1.56

itud

-1.55

advertisement

-1.55

�

-1.54

ophob

-1.53

chio

-1.51

POSITIVE LOGITS

 Cancel

1.57

 Mechdragon

1.56

 resign

1.54

 resigned

1.52

 Creed

1.52

 Execution

1.43

 Dish

1.41

 Reboot

1.41

CLOSE

1.40

 Reck

1.40

Activations Density 0.000%

No Known Activations

This feature has no known activations.