INDEX

Explanations

phrases indicating levels of intensity or impact in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

0:0.04

1:0.10

2:0.03

3:0.03

4:0.02

5:0.40

6:0.16

7:0.01

8:0.05

9:0.06

10:0.03

11:0.02

Negative Logits

antry

-1.89

 mathemat

-1.55

isers

-1.52

olving

-1.49

 monop

-1.49

 blankets

-1.48

���

-1.46

ipers

-1.46

 wrapping

-1.46

 Airl

-1.39

POSITIVE LOGITS

bh

1.87

Ann

1.86

1.76

pause

1.75

scar

1.73

Ry

1.69

sbm

1.68

Am

1.65

Ur

1.64

Torrent

1.64

Activations Density 0.069%

phrases indicating levels of intensity or impact in various contexts

No Comments

No Known Activations