INDEX

Explanations

the word 'in'

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

2:0.07

3:0.08

4:0.11

5:0.08

6:0.09

7:0.07

8:0.08

9:0.07

10:0.06

11:0.08

Negative Logits

 Barton

-2.53

 Stamp

-2.50

 Buchanan

-2.45

 Tucker

-2.44

 Schultz

-2.25

 Stacy

-2.19

 starters

-2.18

 Ricardo

-2.15

 Jarrett

-2.11

 Prescott

-2.09

POSITIVE LOGITS

obyl

2.93

lyak

2.79

龍契士

2.68

oğ

2.67

hani

2.62

igmat

2.57

 exting

2.57

Iran

2.56

accompan

2.46

Rated

2.44

Activations Density 0.000%

the word 'in'

No Comments

No Known Activations