INDEX

Explanations

terms related to environmental or ecological concepts, particularly those involving carbon emissions and sustainability

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.02

1:0.05

2:0.09

3:0.24

4:0.02

5:0.02

6:0.08

7:0.10

8:0.06

9:0.11

10:0.04

11:0.12

Negative Logits

olson

-1.28

 reckoning

-1.28

drm

-1.22

velt

-1.18

 Katz

-1.14

 skate

-1.13

 extremes

-1.12

 pens

-1.07

 reflection

-1.07

enko

-1.06

POSITIVE LOGITS

Enlarge

1.29

 Corpus

1.26

¶

1.18

lishes

1.14

cius

1.12

olated

1.11

ん

1.08

urally

1.08

 Supported

1.04

ctions

1.03

Activations Density 0.002%

terms related to environmental or ecological concepts, particularly those involving carbon emissions and sustainability

No Known Activations