INDEX

Explanations

punctuation marks, specifically commas

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.07

1:0.09

2:0.08

3:0.08

4:0.08

5:0.08

6:0.07

7:0.08

8:0.08

9:0.08

10:0.09

11:0.07

Negative Logits

 periodic

-2.80

ldom

-2.74

 monop

-2.59

retty

-2.50

pson

-2.50

 yearly

-2.49

 quarterly

-2.47

atra

-2.47

armor

-2.42

stats

-2.36

POSITIVE LOGITS

YC

2.98

 Zucker

2.97

 Canaver

2.65

 bulldo

2.46

kens

2.45

 Palestinians

2.43

 Cuomo

2.41

 Maduro

2.39

 Hawking

2.33

 Kinder

2.32

Activations Density 0.000%

punctuation marks, specifically commas

No Comments

No Known Activations