INDEX

Explanations

specific coded values or identifiers associated with a news article or individual

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_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

Not in Any Lists

No Comments

Head Attr Weights

0:0.02

1:0.14

2:0.08

3:0.06

4:0.05

5:0.12

6:0.08

7:0.08

8:0.08

9:0.07

10:0.09

11:0.09

Negative Logits

bats

-1.44

 aliases

-1.39

uyomi

-1.38

erness

-1.36

ikan

-1.35

 acron

-1.31

 initials

-1.30

 Railroad

-1.28

 Catalyst

-1.25

 metaphors

-1.25

POSITIVE LOGITS

 compr

1.41

 infl

1.31

 exha

1.28

 disp

1.24

 earthqu

1.20

ThumbnailImage

1.17

天

1.16

 exting

1.16

�

1.15

 metic

1.14

Activations Density 0.000%

specific coded values or identifiers associated with a news article or individual

No Comments

No Known Activations