INDEX

Explanations

prepositions that signify locations or affiliations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.01

1:0.01

2:0.04

3:0.04

4:0.08

5:0.02

6:0.03

7:0.43

8:0.03

9:0.02

10:0.15

11:0.10

Negative Logits

alogue

-1.65

Offline

-1.55

メ

-1.54

runner

-1.42

earchers

-1.39

tracking

-1.39

 selfie

-1.37

wik

-1.37

 proxies

-1.36

GET

-1.34

POSITIVE LOGITS

 Engineering

1.63

 Oriental

1.54

Civ

1.53

 Harvard

1.51

 undergrad

1.47

 Sciences

1.46

NYU

1.46

 Advanced

1.45

MIT

1.45

 college

1.43

Activations Density 0.016%

prepositions that signify locations or affiliations

No Comments

No Known Activations