INDEX

Explanations

the word "only"

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

0:0.08

1:0.07

2:0.08

3:0.08

4:0.08

5:0.06

6:0.08

7:0.08

8:0.08

9:0.09

10:0.09

11:0.09

Negative Logits

 Grape

-2.91

 Nikola

-2.77

 Watkins

-2.69

 Pepe

-2.61

 Meteor

-2.52

 Nets

-2.51

lete

-2.51

 Elon

-2.48

Wiz

-2.47

 Vacc

-2.47

POSITIVE LOGITS

CI

2.85

ALSE

2.67

thood

2.63

nam

2.59

ansen

2.58

isations

2.54

abad

2.47

alin

2.46

 relates

2.42

psons

2.36

Activations Density 0.000%

the word "only"

No Comments

No Known Activations