INDEX

Explanations

comparative phrases that highlight differences between entities or concepts

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_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.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

1:0.10

2:0.06

3:0.02

4:0.01

5:0.03

6:0.07

7:0.07

8:0.15

9:0.27

10:0.07

11:0.07

Negative Logits

 shoots

-1.16

lies

-1.15

 basis

-1.14

}}

-1.13

 Bruins

-1.13

 trainer

-1.10

 uncertainties

-1.09

 foundation

-1.06

 spoke

-1.04

 Italian

-1.04

POSITIVE LOGITS

perty

1.76

SPONSORED

1.57

outer

1.41

emis

1.39

ubis

1.32

jriwal

1.30

icket

1.27

ottest

1.26

indal

1.26

oute

1.24

Activations Density 0.066%

comparative phrases that highlight differences between entities or concepts

No Comments

No Known Activations