INDEX

Explanations

phrases related to securing objects to prevent them from falling

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

4:0.07

5:0.08

6:0.08

7:0.07

8:0.08

9:0.09

10:0.08

11:0.09

Negative Logits

rals

-1.73

states

-1.71

rolet

-1.68

rete

-1.61

ipel

-1.61

orage

-1.61

cific

-1.57

ounty

-1.56

blers

-1.55

otton

-1.46

POSITIVE LOGITS

 hijab

1.68

 charism

1.60

 Kremlin

1.55

 sexism

1.53

 cynicism

1.51

 questioning

1.48

 homophobia

1.48

 pledging

1.47

▬▬

1.46

 salute

1.46

Activations Density 0.000%

phrases related to securing objects to prevent them from falling

No Known Activations