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

Head Attr Weights

0:0.09

1:0.07

2:0.08

3:0.07

4:0.08

5:0.08

6:0.07

7:0.08

8:0.09

9:0.08

10:0.07

11:0.07

Negative Logits

 Thumbnails

-1.83

 compliment

-1.82

 slideshow

-1.80

 timestamp

-1.74

 compliments

-1.71

 retweet

-1.58

 stamp

-1.55

 greets

-1.52

 Comments

-1.49

 salute

-1.47

POSITIVE LOGITS

uca

1.71

igm

1.70

 immersion

1.65

 adversary

1.65

aimon

1.59

inges

1.58

NetMessage

1.57

 Cantor

1.52

venants

1.52

 temptation

1.51

Activations Density 0.000%

No Known Activations

This feature has no known activations.