INDEX

Explanations

references to processes and structures in computer programming and economic critiques

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

1:0.01

2:0.35

3:0.09

4:0.11

5:0.02

6:0.03

7:0.06

8:0.06

9:0.05

10:0.10

11:0.06

Negative Logits

ivalry

-1.72

akia

-1.49

aze

-1.41

acity

-1.39

ahan

-1.38

natureconservancy

-1.37

itans

-1.34

 wonders

-1.34

udi

-1.34

ida

-1.32

POSITIVE LOGITS

Reloaded

1.54

EngineDebug

1.49

 Palest

1.49

 Mund

1.47

Lear

1.43

ometime

1.40

retty

1.39

Edited

1.39

ICE

1.32

 Defendants

1.31

Activations Density 0.949%

references to processes and structures in computer programming and economic critiques

No Comments

No Known Activations