INDEX

Explanations

references to receiving newsletters or updates in an inbox

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

New Auto-Interp

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

1:0.03

2:0.20

3:0.09

4:0.06

5:0.05

6:0.12

7:0.06

8:0.07

9:0.04

10:0.07

11:0.14

Negative Logits

inventoryQuantity

-1.92

 Flavoring

-1.66

ファ

-1.65

ibrary

-1.58

イ

-1.55

rack

-1.54

ragon

-1.50

 Cosponsors

-1.49

DIV

-1.46

actionDate

-1.40

POSITIVE LOGITS

 outlining

1.48

 impacting

1.48

 resembling

1.45

 RELEASE

1.42

 detailing

1.42

tid

1.41

 pertaining

1.37

 affecting

1.30

 relating

1.27

 pleasing

1.24

Activations Density 0.001%

references to receiving newsletters or updates in an inbox

No Comments

No Known Activations