INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.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

1:0.08

2:0.07

3:0.08

4:0.06

5:0.08

6:0.08

7:0.09

8:0.09

9:0.08

10:0.08

11:0.08

Negative Logits

ortal

-2.76

ahon

-2.75

ardless

-2.66

ascript

-2.65

enfranch

-2.62

_(

-2.61

 largeDownload

-2.60

itially

-2.60

ordial

-2.58

Christ

-2.57

POSITIVE LOGITS

Mug

2.89

 Pyramid

2.86

 Tsukuyomi

2.75

Sau

2.73

 Splash

2.72

 Rarity

2.71

Tus

2.66

Sph

2.65

 Quantity

2.60

 Travels

2.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.