INDEX

Explanations

numerical values and quantities related to 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

0:0.02

1:0.02

2:0.12

3:0.08

4:0.35

5:0.02

6:0.11

7:0.05

8:0.04

9:0.03

10:0.05

11:0.05

Negative Logits

ORGE

-1.45

venge

-1.35

orget

-1.32

ONT

-1.31

ezvous

-1.26

efer

-1.24

together

-1.21

uador

-1.21

trak

-1.20

mble

-1.20

POSITIVE LOGITS

 standpoint

1.36

 guiIcon

1.35

}.

1.33

 respective

1.27

 occasions

1.27

 foregoing

1.24

 basis

1.22

 until

1.21

 Supplementary

1.21

max

1.20

Activations Density 0.009%

numerical values and quantities related to various contexts

No Comments

No Known Activations