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

2:0.11

3:0.07

4:0.09

5:0.08

6:0.08

7:0.07

8:0.09

9:0.06

10:0.08

11:0.09

Negative Logits

 punishments

-1.79

 layoffs

-1.74

 allowance

-1.71

 violations

-1.69

 detrimental

-1.69

 restraints

-1.67

 friction

-1.66

 discretionary

-1.64

 infring

-1.62

 transfers

-1.60

POSITIVE LOGITS

wat

2.01

Champ

1.88

代

1.76

ioch

1.74

yssey

1.73

tracks

1.71

 Tycoon

1.68

otos

1.66

cyclop

1.66

Engineers

1.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.