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

2:0.08

3:0.08

4:0.07

5:0.07

6:0.09

7:0.09

8:0.08

9:0.09

10:0.08

11:0.08

Negative Logits

 Intro

-1.82

atography

-1.78

ahime

-1.74

ascript

-1.68

 サーティワン

-1.56

odan

-1.55

oir

-1.55

 Introduction

-1.55

anco

-1.49

oeuv

-1.48

POSITIVE LOGITS

BUG

2.10

 Panic

1.84

 havoc

1.74

CTV

1.73

 flooding

1.73

Days

1.73

 flood

1.66

 suing

1.63

soever

1.61

payers

1.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.