INDEX

Explanations

references to status, reputation, and legitimacy in various contexts

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

1:0.02

2:0.08

3:0.13

4:0.02

5:0.04

6:0.06

7:0.07

8:0.09

9:0.21

10:0.05

11:0.17

Negative Logits

 Shutterstock

-1.29

 avid

-1.23

 alone

-1.16

uploads

-1.14

 videos

-1.13

 blindly

-1.12

 bombard

-1.07

 tabl

-1.07

udi

-1.05

 scams

-1.04

POSITIVE LOGITS

 victory

1.39

pletion

1.36

omen

1.35

 distinction

1.32

iction

1.29

inction

1.26

 clin

1.25

ointment

1.23

favorite

1.22

 winning

1.21

Activations Density 0.009%

references to status, reputation, and legitimacy in various contexts

No Comments

No Known Activations