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

0:0.07

1:0.05

2:0.08

3:0.08

4:0.10

5:0.07

6:0.08

7:0.09

8:0.08

9:0.09

10:0.08

11:0.08

Negative Logits

umbnails

-1.77

axies

-1.73

DVD

-1.68

 DVDs

-1.65

rift

-1.62

CD

-1.61

 dearly

-1.60

 Expend

-1.57

models

-1.56

pod

-1.53

POSITIVE LOGITS

 Shelley

1.67

 Agric

1.67

 Butcher

1.65

 Azerb

1.64

 Moroccan

1.61

iberal

1.53

 Colony

1.52

 Cornwall

1.50

 Stranger

1.49

 Humane

1.48

Activations Density 0.000%

No Known Activations

This feature has no known activations.